# Brew — Complete Marketing Site Content > Brew is the AI-native email service provider (ESP) for modern teams and their AI agents. Brand lives in markdown files Brew reads at generation time, so every email ships on-brand from the app, the API, or any agent over MCP. This document inlines every public marketing page. For anything beyond marketing content, use these instead: - [Index of everything](https://brew.new/llms.txt) — the navigable llms.txt index - [Product documentation](https://docs.brew.new) — guides, recipes, integrations, changelog - [Public API v1 agent guide](https://brew.new/api/v1/llms.txt) — endpoints, error codes, required ordering - [OpenAPI 3.1 spec](https://brew.new/openapi.json) — machine-readable API contract - [OpenAPI YAML](https://brew.new/api/openapi.yaml) — same spec at the /api/openapi.yaml well-known path --- # Brew.new: the AI-native ESP for modern teams and their AI agents Brew is the AI-native email service provider (ESP) for modern teams and their AI agents. Brand lives in markdown files Brew reads at generation time, so every email ships on-brand from the app, the API, or any agent over MCP. Describe an email, automation, or audience in plain English. Brew.new generates the on-brand design, sends it from your verified domain, and reports on delivery. The same capabilities are in the web app, the Public API, the TypeScript SDK, and a hosted MCP server. ## What Brew.new does Brew.new is a full email service provider, not a layer on another ESP. It stores contacts, authenticates your sending domain with SPF, DKIM, and DMARC, and owns delivery. Brand voice, palette, typography, and imagery are learned from your site and applied to every email. Teams ship one-off emails, event-driven automations, and transactional email from the same brand system. Live campaign and automation sends require a verified custom domain you own. ## Brew.new developer resources Developers and AI agents can drive the lifecycle without the dashboard. Start at /developers (also /developer-resources) for the Public API, OpenAPI spec, authentication, webhooks, and hosted MCP. The same surfaces are listed in /llms.txt and /index.md. ### Public API and OpenAPI REST API v1 plus the OpenAPI 3.1 document at /openapi.json, /openapi.yaml, and /api/openapi.yaml. The agent guide at /api/v1/llms.txt encodes endpoint order, error codes, idempotency, and rate limits. ### Authentication and webhooks Bearer API keys, brand scoping, and X-Brand-Id. Delivery callbacks live in the events-and-webhooks catalog on docs.brew.new. ### MCP server Streamable HTTP at /api/mcp (POST /mcp rewrites to the same handler). Discover it via /mcp.json and /.well-known/mcp/server.json. ChatGPT, Claude, Cursor, and other agents can generate, send, automate, and analyze email. ## Brew.new features Every plan includes the full product. Plans differ in AI credits, send volume, sending domains, brand workspaces, and whether generated emails carry a watermark. ### AI Email Generation Create complete email campaigns from simple prompts ### Brand Consistency Automatic extraction and application of brand colors, logos, and fonts ### Email Automations Build sophisticated sequences with a single prompt (welcome, onboarding, re-engagement) ### Remix Gallery Browse and customize AI-generated email templates ### Analytics Real-time performance tracking and AI-powered insights ### Deliverability Built-in DKIM, SPF, DMARC authentication ## How teams use Brew.new Marketing and growth teams draft campaigns and event-driven automations in natural language. Typical Brew.new use cases: Marketing Campaigns, Welcome Email Sequences, Onboarding Flows, Post-Purchase Follow-ups, Re-engagement Campaigns, Newsletter Creation, Transactional Emails, Abandoned Cart Reminders, Birthday and Anniversary Emails. Permission-based sending, suppression lists, bounce handling, and inbox-placement testing are built in. Live sends require a verified custom domain you own. Canonical: https://brew.new/ ## Start here - [Pricing](https://brew.new/pricing.md): Free, Growth, Pro, and Enterprise - [Sign up](https://brew.new/sign-up.md): Free plan, no credit card required - [Email templates](https://brew.new/browse/templates.md): Remixable AI-generated designs - [Explore canvas](https://brew.new/explore.md): Pan, zoom, and search the template set - [Integrations](https://brew.new/browse/integrations.md): Connect the tools you already use - [MCP server](https://brew.new/mcp.md): Email marketing for AI agents - [Brew.new developer resources](https://brew.new/developers.md): API, OpenAPI, authentication, webhooks, MCP - [Blog](https://brew.new/blog.md): Product updates and email-marketing notes - [Support](https://brew.new/support.md): Email support and docs ## Links - Website: https://brew.new - Documentation: https://docs.brew.new - Twitter: https://x.com/brewdotnew - LinkedIn: https://www.linkedin.com/company/brewdotnew - Support: support@brew.new --- # Pricing: Brew Transparent pricing for the AI-native ESP. Every plan includes all features. Start free with 500 AI credits and 1,000 sends a month. Canonical: https://brew.new/pricing Free tier available, paid plans from $99/mo (Growth) and $249/mo (Pro). Every plan includes every feature. Plans differ in AI credits, email send volume, sending domains, brand workspaces, and whether generated emails carry the Brew watermark. ## Plans | Plan | Price | AI credits / month | Email sends / month | Sending domains | Brand workspaces | Watermark | | --- | --- | --- | --- | --- | --- | --- | | Free | $0 | 500 | 1,000 | Unlimited | 3 | Yes | | Growth | $99/mo | 10,000 | 50,000 | Unlimited | 10 | No | | Pro | $249/mo | 25,000 | 200,000 | Unlimited | 50 | No | | Enterprise | Custom | Custom | Custom | Unlimited | Unlimited | No | ### Free: $0 All features. No card required. Includes: - 500 AI credits / month - 1K email sends / month - Unlimited sending domains - 3 brand workspaces - Brew watermark on emails ### Growth: $99/mo More volume for growing teams. Includes: - 10K AI credits / month - 50K email sends / month - Unlimited sending domains - 10 brand workspaces - No Brew watermark ### Pro: $249/mo Highest self-serve volume. Includes: - 25K AI credits / month - 200K email sends / month - Unlimited sending domains - 50 brand workspaces - No Brew watermark ### Enterprise: Custom Custom volume and dedicated support. Includes: - Custom AI credit volume - Custom email send volume - Unlimited sending domains - Unlimited brand workspaces - Dedicated account manager Growth is $99 per month. Start on Free and upgrade in the app when you need more volume. ## Self-serve - Start on Free with no credit card required. - Upgrade, downgrade, or cancel anytime from the billing portal. - Monthly AI credits reset on your billing date and do not roll over. - Unlimited team seats on every plan. Brew does not charge per seat. - Live campaign and automation sends require a verified custom domain you own. The Brew default sender is reserved for test sends. - Checkout is self-serve on Stripe. Enterprise volume and a dedicated account manager are available from sales. See the [pricing page](https://brew.new/pricing) to start free or talk to sales. --- # Brew MCP Server: Email Marketing for AI Agents Connect ChatGPT, Claude, Cursor, or any AI agent to Brew over MCP. Agents generate, send, automate, and analyze on-brand email. Canonical: https://brew.new/mcp ## What it is A Model Context Protocol (MCP) server that exposes Brew to AI agents. Connect it in ChatGPT, Claude, Grok, Cursor, Codex, or any MCP client and your agent runs the full email-marketing cycle, backed by the same brand and data as the Brew app. ## What your agent can do - Generate and iterate on on-brand email designs - Send and schedule campaigns and test sends through verified domains - Build automations: wire triggers to multi-step flows, test, publish, and fire events - Manage contacts, custom fields, and saved audiences (segments) - Read delivery, open, click, bounce, and automation analytics - Generate on-brand images and GIFs and render HTML previews ## Connect In ChatGPT, enable Developer mode under Settings → Security and login. Then open Settings → Plugins, select `+`, and create Brew with OAuth using this server URL: ``` https://brew.new/api/mcp ``` Sign in to Brew, select a brand, and approve the disclosed full-brand access. Start a new chat, enable `@Brew` from the Plugins menu, and call `get_brew_capabilities`. If an older cached plugin reports an invalid schema, refresh its actions; remove and re-add Brew only after the corrected endpoint is deployed and only if refresh does not work. For another OAuth-capable MCP client, add the same server URL and complete the same Brew sign-in and brand selection. If your client does not support OAuth connectors yet, add the same URL to its `mcpServers` config and authenticate with a brand-scoped Brew API key from Settings → API instead: ```json { "mcpServers": { "brew": { "url": "https://brew.new/api/mcp", "headers": { "Authorization": "Bearer brew_YOUR_API_KEY" } } } } ``` ## Compose with your CRM Brew is the AI-native ESP: the durable email memory and sending infrastructure for agents. Pair it with a CRM or enrichment MCP (Clay, Attio, and others) to enrich contacts, build intelligent segments, and send personalized email: enrich → segment → design → send → measure → re-segment. ## Workflows The server ships guided prompts and a discovery tool (`get_brew_capabilities`) so agents chain tools in the right order: launch a campaign, build a segment and send, set up an automation, analyze performance, and personalize with a CRM MCP. Learn more: https://brew.new/mcp · Docs: https://docs.brew.new/api-reference/mcp/overview --- # Brew.new developer resources Brew.new developer resources: Public API v1, OpenAPI spec, authentication, webhooks, TypeScript SDK, and hosted MCP server. Canonical: https://brew.new/developers Also published at https://brew.new/developer-resources (redirects here). Predictable URLs for the Public API, OpenAPI spec, authentication, webhooks, and hosted MCP. Use these when an agent searches for Brew.new developer resources. ## Brew.new developer resources - [Brew.new developer resources](https://brew.new/developers): This hub: API, OpenAPI, authentication, webhooks, SDK, and MCP - [Public API v1 introduction](https://docs.brew.new/api-reference/api/api-introduction): Base URL, first request, and cURL examples - [OpenAPI specification (JSON)](https://brew.new/openapi.json): OpenAPI 3.1 document at the well-known /openapi.json path - [OpenAPI specification (YAML)](https://brew.new/api/openapi.yaml): Same spec as YAML at /api/openapi.yaml and /openapi.yaml - [Authentication](https://docs.brew.new/api-reference/api/authentication): Bearer API keys, brand scoping, and X-Brand-Id - [Events and webhooks](https://docs.brew.new/api-reference/api/events-and-webhooks): Delivery events, webhook signatures, and retry behavior - [Brew.new MCP server](https://brew.new/mcp): Human overview of the hosted Model Context Protocol server - [Brew.new MCP Streamable HTTP](https://brew.new/api/mcp): Streamable HTTP MCP endpoint. POST JSON-RPC; OAuth or brew_ API key - [Brew.new MCP server.json](https://brew.new/mcp.json): Official MCP Registry document at /mcp.json and /.well-known/mcp/server.json - [TypeScript SDK](https://docs.brew.new/sdks/overview): `@brew.new/sdk`: typed client generated from the OpenAPI spec - [Public API agent guide](https://brew.new/api/v1/llms.txt): Endpoint order, error codes, idempotency, and rate limits - [Product documentation](https://docs.brew.new): Guides, recipes, and the generated endpoint reference ## Well-known machine-readable files - OpenAPI JSON: https://brew.new/openapi.json - OpenAPI YAML: https://brew.new/openapi.yaml - OpenAPI YAML (API tree): https://brew.new/api/openapi.yaml - Agent index: https://brew.new/llms.txt - Public API agent guide: https://brew.new/api/v1/llms.txt - MCP Streamable HTTP: https://brew.new/api/mcp - MCP server.json: https://brew.new/mcp.json - MCP server.json (well-known): https://brew.new/.well-known/mcp/server.json --- # Brew Support Get help with Brew in ChatGPT, Claude, Grok, Cursor, Codex, other MCP clients, and your email workflows. Canonical: https://brew.new/support For plugin review or connection issues, include the client, affected brand, approximate time, and any visible error message. Never send passwords or access tokens. ## Contact and resources ### Email support Contact the Brew team for account, billing, plugin, or product help. [support@brew.new](mailto:support@brew.new) ### Documentation Read setup guides for the Brew API, MCP server, OAuth, and supported clients. [Open documentation](https://docs.brew.new) ### Privacy and security Review how Brew handles data, access, and its full-brand OAuth permission. [Read privacy policy](https://docs.brew.new/legal/privacy-policy) ## Links - Website: https://brew.new - Documentation: https://docs.brew.new - Support: support@brew.new --- # Explore Email Templates: Infinite Canvas Pan, zoom, and semantically search remixable AI-generated email templates. Click any template to remix it onto your brand. Canonical: https://brew.new/explore The explore canvas is public. No account is required to browse. Remixing a template onto your brand starts a free Brew account. ## Related - [Template gallery](https://brew.new/browse/templates.md) - [Sign up](https://brew.new/sign-up.md) - Website: https://brew.new --- # Email Template Gallery: Brew Browse real AI-generated email designs. Any template can be remixed onto your brand with a single prompt. Canonical: https://brew.new/browse/templates Each design also has a detail page at `https://brew.new/browse/templates/email/{id}` (and `https://brew.new/browse/templates/email/{id}.md`). Open a template to remix it. - [Explore canvas](https://brew.new/explore.md): pan, zoom, and search the full set - [Email flows](https://brew.new/browse/flows): real multi-email sequences from top brands, each at `https://brew.new/browse/flows/{slug}` - [Sign up](https://brew.new/sign-up.md): remix on your brand, no credit card required - Website: https://brew.new --- # Integrations: Connect Your Stack One-click OAuth and API-key integrations for the tools you already use. Pull design files and payments in; push polished emails into your ESP. Canonical: https://brew.new/browse/integrations Brew connects to 20 tools. Data-in providers fire automations from product events or pull designs in. Data-out providers receive finished email designs so you can keep sending from an ESP you already run. ## Data in (9) - [Clerk](https://brew.new/browse/integrations/clerk.md): Auth event triggers - [Figma](https://brew.new/browse/integrations/figma.md): Design-to-email - [Framer](https://brew.new/browse/integrations/framer.md): Forms and signups - [Stripe](https://brew.new/browse/integrations/stripe.md): Customer sync + event triggers - [Stytch](https://brew.new/browse/integrations/stytch.md): Auth event triggers - [Supabase](https://brew.new/browse/integrations/supabase.md): Auth-event triggers - [WorkOS](https://brew.new/browse/integrations/workos.md): Enterprise auth triggers - [Shopify](https://brew.new/browse/integrations/shopify.md): Commerce event triggers - [RevenueCat](https://brew.new/browse/integrations/revenuecat.md): IAP + subscription events ## Data out (11) - [Braze](https://brew.new/browse/integrations/braze.md): Marketing automation export - [Brevo](https://brew.new/browse/integrations/brevo.md): Email marketing & automation - [HubSpot](https://brew.new/browse/integrations/hubspot.md): CRM & marketing automation - [Klaviyo](https://brew.new/browse/integrations/klaviyo.md): Email & SMS marketing - [Iterable](https://brew.new/browse/integrations/iterable.md): Cross-channel marketing - [Postmark](https://brew.new/browse/integrations/postmark.md): Transactional email delivery - [OneSignal](https://brew.new/browse/integrations/onesignal.md): Push & email messaging - [Mailgun](https://brew.new/browse/integrations/mailgun.md): Email delivery infrastructure - [SendGrid](https://brew.new/browse/integrations/sendgrid.md): Email delivery by Twilio - [Mailjet](https://brew.new/browse/integrations/mailjet.md): Email delivery & templates - [Mailchimp](https://brew.new/browse/integrations/mailchimp.md): Email marketing & campaigns - Website: https://brew.new --- # Brew Blog Product updates and notes from the Brew team on email marketing, lifecycle emails, and building with AI agents over MCP. Canonical: https://brew.new/blog ## [Abandoned Cart Emails: How to Recover 10-15% of Lost Sales](https://brew.new/blog/abandoned-cart-emails.md) Master abandoned cart emails with proven templates, timing strategies, and optimization tactics. Learn how to recover lost revenue automatically. *Published: 2025-10-25* ## [What still applies when an agent sends your email](https://brew.new/blog/agent-email-deliverability.md) Agents do not get a separate inbox. Same CAN-SPAM rules, same Gmail thresholds, same need to split marketing from transactional on the wire. *Published: 2026-08-24* ## [How AI Is Changing Email Automation](https://brew.new/blog/ai-email-automation.md) Discover how AI email automation is transforming email marketing. What's real, what's hype, and how to get started. *Published: 2025-10-14* ## [AI Email Marketing Tools: What Actually Works](https://brew.new/blog/ai-email-marketing-tools.md) Discover which AI email marketing tools deliver real results. We separate hype from reality and show you what AI can actually do for your campaigns. *Published: 2025-11-19* ## [B2B Email Marketing: The Complete Strategy Guide](https://brew.new/blog/b2b-email-marketing.md) Master B2B email marketing with proven strategies for lead generation, nurturing, and conversion. Includes templates, sequences, and real examples. *Published: 2025-11-14* ## [10 Best Email Marketing Software in 2026: Complete Comparison Guide](https://brew.new/blog/best-email-marketing-software.md) Find the best email marketing software for your business. We compare 10 top platforms on features, pricing, ease of use, and who each one is best for. *Published: 2025-10-22* ## [Your AI assistant can now run your email marketing](https://brew.new/blog/brew-mcp.md) Connect Brew to ChatGPT, Claude, Cursor, or any AI assistant. It writes on-brand emails, sends and schedules, builds automations, and reports back. *Published: 2026-06-26* ## [Email A/B Testing: The Complete Guide to Better Results](https://brew.new/blog/email-ab-testing.md) Master email A/B testing with proven strategies. Learn what to test, how to run valid tests, and how to interpret results for continuous improvement. *Published: 2025-11-21* ## [The Email Canvas: A Faster Way to Create and Compare Emails](https://brew.new/blog/email-canvas-workflow.md) A canvas-based email workflow lets teams generate, compare, edit, preview, and export campaigns without losing context. *Published: 2026-04-18* ## [Email Deliverability: The Complete Guide to Reaching the Inbox](https://brew.new/blog/email-deliverability.md) Master email deliverability with this comprehensive guide. Learn what affects inbox placement and how to fix common deliverability problems. *Published: 2025-12-10* ## [email-design.md: how we made AI emails actually on-brand](https://brew.new/blog/email-design-md.md) AI agents read CLAUDE.md and llms.txt. So why does your brand live in a database? Why we moved it to email-design.md — a document an LLM can read. *Published: 2026-06-15* ## [Email List Building: 15 Proven Strategies to Grow Your Subscribers](https://brew.new/blog/email-list-building.md) Learn how to build an email list that converts. 15 proven tactics from lead magnets to optimization, with real examples and implementation guides. *Published: 2025-11-03* ## [Email Marketing Automation: The Complete Guide](https://brew.new/blog/email-marketing-automation-guide.md) Master email marketing automation with this comprehensive guide. Learn which automations to build, how to set them up, and strategies that maximize results. *Published: 2025-11-28* ## [10 Email Marketing Best Practices That Still Work](https://brew.new/blog/email-marketing-best-practices.md) Master email marketing best practices that drive results. 10 proven strategies for engagement and revenue in 2025. *Published: 2025-10-09* ## [Email Marketing Platforms Compared: Find Your Perfect Match (2026)](https://brew.new/blog/email-marketing-platforms-compared.md) Side-by-side comparison of leading email marketing platforms. Features, pricing, pros and cons to help you choose the right tool for your business. *Published: 2025-10-29* ## [Email Marketing ROI: How to Calculate and Maximize Your Returns](https://brew.new/blog/email-marketing-roi.md) Learn exactly how to calculate email marketing ROI. Includes benchmarks by industry, strategies to improve returns, and common mistakes that kill ROI. *Published: 2025-11-12* ## [50+ Email Marketing Statistics You Need to Know in 2026](https://brew.new/blog/email-marketing-statistics.md) The latest email marketing statistics and benchmarks for 2025. ROI data, open rates, click rates, and trends backed by current research. *Published: 2025-12-03* ## [Email Marketing Trends 2026: What's Working Now](https://brew.new/blog/email-marketing-trends-2025.md) The email marketing trends shaping 2025. From AI-powered creation to privacy changes, here's what smart marketers are doing differently. *Published: 2025-12-01* ## [Email Marketing vs Social Media: Which Channel Wins?](https://brew.new/blog/email-marketing-vs-social-media.md) Email marketing vs social media compared head-to-head. ROI, reach, engagement, and effectiveness data to help you decide where to invest. *Published: 2025-10-04* ## [Email opens are a weak metric. Measure these instead](https://brew.new/blog/email-metrics-that-matter.md) Apple Mail Privacy Protection inflated opens. Track clicks you trust, replies, conversions, and complaints. Here is how we classify a click. *Published: 2026-08-24* ## [Email Segmentation: The Strategy Guide That Drives 760% More Revenue](https://brew.new/blog/email-segmentation.md) Learn email segmentation strategies that actually work. Practical implementation guide with examples, templates, and step-by-step instructions. *Published: 2025-10-18* ## [Email Subject Lines: 100+ Examples That Actually Get Opens](https://brew.new/blog/email-subject-lines.md) Master email subject lines with proven formulas and 100+ examples. Learn what makes subscribers click and how to write subject lines that convert. *Published: 2025-11-26* ## [Merry Christmas & Happy Holidays from Brew 🎄](https://brew.new/blog/happy-holidays-2025.md) A heartfelt thank you from the Brew team. Reflecting on an incredible year of growth, gratitude, and exciting plans for 2026. *Published: 2025-12-24* ## [How Harmonya turned a monthly newsletter into a lifecycle program](https://brew.new/blog/harmonya-lifecycle-email.md) Noah Adelstein wrote up Cory Nelson at Harmonya. What we saw the morning after he signed up, and what the setup actually depends on. *Published: 2026-08-24* ## [How to brief an email agent](https://brew.new/blog/how-to-brief-an-email-agent.md) A brief an email agent can actually use. Audience, job, offer, proof, and the constraints that stop a fluent draft from shipping the wrong email. *Published: 2026-08-24* ## [How to Write Marketing Emails That Convert (With Templates)](https://brew.new/blog/how-to-write-marketing-emails.md) Learn the fundamentals of writing marketing emails that get opened, read, and clicked. Includes templates, formulas, and real examples. *Published: 2025-12-17* ## [HubSpot vs Mailchimp: Which Email Marketing Platform Is Right for You?](https://brew.new/blog/hubspot-vs-mailchimp.md) HubSpot vs Mailchimp compared head-to-head. We break down pricing, features, automation, ease of use, and which platform wins for different business types. *Published: 2025-10-15* ## [How to Turn Existing Email Designs Into On-Brand Emails With AI](https://brew.new/blog/import-existing-email-designs-ai.md) Teams with existing email designs can use AI to recreate, remix, and adapt their approved layouts without starting from scratch. *Published: 2026-03-21* ## [Introducing Brew: The AI-Native ESP](https://brew.new/blog/introducing-brew.md) Meet Brew—the first email platform where AI creates complete campaigns. Strategy, copy, design, and code in seconds. *Published: 2025-10-07* ## [7 Best Mailchimp Alternatives in 2026: Features, Pricing & Honest Reviews](https://brew.new/blog/mailchimp-alternatives.md) Looking for Mailchimp alternatives? We compare the 7 best options with real pricing, features, pros and cons. Find the right platform for your needs. *Published: 2025-11-07* ## [Newsletter Best Practices: How to Create Emails People Actually Read](https://brew.new/blog/newsletter-best-practices.md) Master newsletter creation with proven best practices. Content strategy, design tips, and growth tactics from successful publishers. *Published: 2025-11-05* ## [Transactional emails: what they are and how to send them](https://brew.new/blog/transactional-emails.md) What counts as transactional under CAN-SPAM, why carts and trial expiry usually do not, and how Brew sends transactional email through automations. *Published: 2026-08-24* ## [Triggered Email Automations: Marketing Flows and Transactional Emails Explained](https://brew.new/blog/triggered-email-automations-guide.md) Triggered email automations include both lifecycle marketing flows and transactional messages. Learn how they work and which ones to build first. *Published: 2026-02-22* ## [Welcome Email Sequence: The Complete Guide (With Templates)](https://brew.new/blog/welcome-email-sequence.md) Learn how to create welcome email sequences that convert. Includes templates, timing strategies, and real examples from high-performing brands. *Published: 2025-10-08* ## [What Is an AI-Native ESP? How Email Platforms Are Changing in 2026](https://brew.new/blog/what-is-an-ai-native-esp.md) AI-native ESPs replace manual email production with agents that understand brand, build campaigns, trigger automations, and connect to your stack. *Published: 2026-05-02* ## [What Is Email Marketing? The Complete Beginner's Guide](https://brew.new/blog/what-is-email-marketing.md) Learn what email marketing is, why it matters, and how to get started. A beginner-friendly guide covering everything you need to know. *Published: 2025-10-11* ## [What we learned from hitting #1 on Product Hunt](https://brew.new/blog/what-we-learned-hitting-product-of-the-week.md) Brew hit #1 Product of the Week on [Product Hunt](https://www.producthunt.com/products/brew-5). The playbook, the surprises, and what I wish I had known. *Published: 2026-06-02* ## [Win-back emails: how to re-engage customers who went quiet](https://brew.new/blog/win-back-emails.md) How to write a win-back sequence. Define quiet in your product, lead with a reason to return, and stop before the list starts complaining. *Published: 2026-08-24* --- # Sign up for Brew Create your free Brew account. Define your brand, then generate on-brand emails, automations, and sends. No credit card required. Canonical: https://brew.new/sign-up The Free plan includes every product feature, with a Brew watermark on generated emails. Upgrade to Growth or Pro in the app when you need more volume. - [Pricing](https://brew.new/pricing.md) - [Sign in](https://brew.new/sign-in.md) - Website: https://brew.new --- # Sign in to Brew Sign in to your Brew account and pick up where you left off with emails, automations, audiences, and analytics. Canonical: https://brew.new/sign-in New to Brew? [Create a free account](https://brew.new/sign-up.md). No credit card required. - Website: https://brew.new --- # Integration details # Braze + Brew Push generated email templates straight into Braze for campaigns and journeys. Canonical: https://brew.new/browse/integrations/braze - Direction: Data out - Connection: API key - Marketing automation export ## Live capabilities - **Export template**: Push a new email template into the provider so campaigns and journeys can use it. ## Related - [All integrations](https://brew.new/browse/integrations.md) - [Sign up](https://brew.new/sign-up.md) - Website: https://brew.new --- # Brevo + Brew Sync templates and transactional sends with Brevo (formerly Sendinblue). Canonical: https://brew.new/browse/integrations/brevo - Direction: Data out - Connection: API key - Email marketing & automation ## Live capabilities - **Export template**: Push a new email template into the provider so campaigns and journeys can use it. - **Verified senders**: Pick a verified sender identity at export time so the provider accepts the message. ## Related - [All integrations](https://brew.new/browse/integrations.md) - [Sign up](https://brew.new/sign-up.md) - Website: https://brew.new --- # Clerk + Brew Trigger Brew automations from Clerk webhook events — user signups, organization changes, paid subscriptions, and waitlist entries. Canonical: https://brew.new/browse/integrations/clerk - Direction: Data in - Connection: Webhook secret - Auth event triggers ## Live capabilities - **Event triggers**: Trigger Brew automations from 13 Clerk webhook events across user, organization, billing, and waitlist categories. ## Related - [All integrations](https://brew.new/browse/integrations.md) - [Sign up](https://brew.new/sign-up.md) - Website: https://brew.new --- # Figma + Brew Turn Figma frames into pixel-accurate, responsive emails — paste a frame URL and Brew handles the rest. Canonical: https://brew.new/browse/integrations/figma - Direction: Data in - Connection: OAuth - Design-to-email ## Live capabilities - **Import design**: Import a Figma frame into Brew as a rendered, editable email draft. ## Related - [All integrations](https://brew.new/browse/integrations.md) - [Sign up](https://brew.new/sign-up.md) - Website: https://brew.new --- # Framer + Brew Add your existing Framer forms to Brew. Submissions add or update contacts and can start automations using each form’s trigger. Canonical: https://brew.new/browse/integrations/framer - Direction: Data in - Connection: Form webhook - Forms and signups ## Live capabilities - **Event triggers**: Each named form collects contacts and fires its own trigger. - **Sync contacts**: Create or update contacts from signed form submissions. ## Related - [All integrations](https://brew.new/browse/integrations.md) - [Sign up](https://brew.new/sign-up.md) - Website: https://brew.new --- # HubSpot + Brew Export emails to HubSpot Marketing Hub as ready-to-send templates for your campaigns and workflows. Canonical: https://brew.new/browse/integrations/hubspot - Direction: Data out - Connection: OAuth - CRM & marketing automation ## Live capabilities - **Export template**: Create a coded email template in Design Manager through the CMS v3 Source Code API. ## Related - [All integrations](https://brew.new/browse/integrations.md) - [Sign up](https://brew.new/sign-up.md) - Website: https://brew.new --- # Klaviyo + Brew Push templates into Klaviyo flows and campaigns in one click via OAuth. Canonical: https://brew.new/browse/integrations/klaviyo - Direction: Data out - Connection: OAuth - Email & SMS marketing ## Live capabilities - **Export template**: Push a new email template into the provider so campaigns and journeys can use it. ## Related - [All integrations](https://brew.new/browse/integrations.md) - [Sign up](https://brew.new/sign-up.md) - Website: https://brew.new --- # Iterable + Brew Ship Brew templates into Iterable as reusable assets for cross-channel campaigns and journeys. Canonical: https://brew.new/browse/integrations/iterable - Direction: Data out - Connection: API key - Cross-channel marketing ## Live capabilities - **Upsert template**: Iterable upsert acts as both export and update — re-running with the same `clientTemplateId` patches the existing template. - **Update template**: Same upsert endpoint as export — `clientTemplateId` keeps it idempotent. ## Related - [All integrations](https://brew.new/browse/integrations.md) - [Sign up](https://brew.new/sign-up.md) - Website: https://brew.new --- # Postmark + Brew Send templates to Postmark for transactional and broadcast email delivery. Canonical: https://brew.new/browse/integrations/postmark - Direction: Data out - Connection: API key - Transactional email delivery ## Live capabilities - **Export template**: Push a new email template into the provider so campaigns and journeys can use it. ## Related - [All integrations](https://brew.new/browse/integrations.md) - [Sign up](https://brew.new/sign-up.md) - Website: https://brew.new --- # OneSignal + Brew Push templates into OneSignal app campaigns and automations. Canonical: https://brew.new/browse/integrations/onesignal - Direction: Data out - Connection: API key - Push & email messaging ## Live capabilities - **Export template**: Push a new email template into the provider so campaigns and journeys can use it. ## Related - [All integrations](https://brew.new/browse/integrations.md) - [Sign up](https://brew.new/sign-up.md) - Website: https://brew.new --- # Mailgun + Brew Save templates to Mailgun for transactional and marketing sends across your domains. Canonical: https://brew.new/browse/integrations/mailgun - Direction: Data out - Connection: API key - Email delivery infrastructure ## Live capabilities - **Export template**: Push a new email template into the provider so campaigns and journeys can use it. ## Related - [All integrations](https://brew.new/browse/integrations.md) - [Sign up](https://brew.new/sign-up.md) - Website: https://brew.new --- # SendGrid + Brew Push templates into Twilio SendGrid as dynamic templates for transactional and marketing sends. Canonical: https://brew.new/browse/integrations/sendgrid - Direction: Data out - Connection: API key - Email delivery by Twilio ## Live capabilities - **Export template**: Push a new email template into the provider so campaigns and journeys can use it. ## Related - [All integrations](https://brew.new/browse/integrations.md) - [Sign up](https://brew.new/sign-up.md) - Website: https://brew.new --- # Mailjet + Brew Save HTML templates to Mailjet for transactional and marketing email campaigns. Canonical: https://brew.new/browse/integrations/mailjet - Direction: Data out - Connection: API key pair - Email delivery & templates ## Live capabilities - **Export template**: Push a new email template into the provider so campaigns and journeys can use it. ## Related - [All integrations](https://brew.new/browse/integrations.md) - [Sign up](https://brew.new/sign-up.md) - Website: https://brew.new --- # Mailchimp + Brew Save AI-generated emails as Mailchimp templates for your campaigns and automations. Canonical: https://brew.new/browse/integrations/mailchimp - Direction: Data out - Connection: OAuth - Email marketing & campaigns ## Live capabilities - **Export template**: Push a new email template into the provider so campaigns and journeys can use it. ## Related - [All integrations](https://brew.new/browse/integrations.md) - [Sign up](https://brew.new/sign-up.md) - Website: https://brew.new --- # Stripe + Brew Sync Stripe customers into the brand audience, trigger automations from 22 Stripe events, and ship pre-built lifecycle emails (welcome, trial-end, failed-payment recovery, win-back). Canonical: https://brew.new/browse/integrations/stripe - Direction: Data in - Connection: OAuth - Customer sync + event triggers ## Live capabilities - **Event triggers**: Trigger Brew automations from 22 Stripe events across subscriptions, invoices, checkouts, quotes, charges, refunds, and payment intents. - **Sync contacts**: Sync Stripe customers into the brand audience with `stripe_*` custom fields available as filterable attributes. - **Bulk customer import**: One-click backfill of every Stripe customer into the active brand. Resumable, durable via Vercel Workflow. - **Lifecycle email templates**: Starter templates for welcome, trial-end, failed-payment recovery, and win-back. ## Related - [All integrations](https://brew.new/browse/integrations.md) - [Sign up](https://brew.new/sign-up.md) - Website: https://brew.new --- # Stytch + Brew Trigger Brew automations from identity-bearing Stytch user and member events. Brew records organization and SCIM connection events without a recipient as unresolved. Only DIRECT-source events are processed. Canonical: https://brew.new/browse/integrations/stytch - Direction: Data in - Connection: Webhook secret - Auth event triggers ## Live capabilities - **Event triggers**: Trigger Brew automations from Stytch Consumer + B2B user, member, organization, and connection events. ## Related - [All integrations](https://brew.new/browse/integrations.md) - [Sign up](https://brew.new/sign-up.md) - Website: https://brew.new --- # Supabase + Brew Trigger Brew automations from Supabase auth.users database webhooks — including derived events like email confirmation, password change, anonymous-to-real upgrade, and ban / unban. Canonical: https://brew.new/browse/integrations/supabase - Direction: Data in - Connection: Webhook secret - Auth-event triggers ## Live capabilities - **Event triggers**: Trigger Brew automations from Supabase auth.users database webhooks — includes synthetic events like email confirmation, password change, and sign-in. ## Related - [All integrations](https://brew.new/browse/integrations.md) - [Sign up](https://brew.new/sign-up.md) - Website: https://brew.new --- # WorkOS + Brew Trigger Brew automations from WorkOS user, authentication, password-reset, organization, directory-sync (SCIM), connection, and invitation events. Canonical: https://brew.new/browse/integrations/workos - Direction: Data in - Connection: Webhook secret - Enterprise auth triggers ## Live capabilities - **Event triggers**: Trigger Brew automations from WorkOS user, authentication, organization, directory sync, and connection events. ## Related - [All integrations](https://brew.new/browse/integrations.md) - [Sign up](https://brew.new/sign-up.md) - Website: https://brew.new --- # Shopify + Brew Trigger Brew automations from Shopify customer, order, checkout, and fulfillment events. Powers welcome, receipt, abandoned-checkout, and shipping flows. Canonical: https://brew.new/browse/integrations/shopify - Direction: Data in - Connection: Webhook secret - Commerce event triggers ## Live capabilities - **Event triggers**: Trigger Brew automations from Shopify customer, order, checkout, and fulfillment events. Powers welcome, receipt, abandoned-checkout, and shipping flows. ## Related - [All integrations](https://brew.new/browse/integrations.md) - [Sign up](https://brew.new/sign-up.md) - Website: https://brew.new --- # RevenueCat + Brew Trigger Brew automations from RevenueCat in-app purchase, subscription, and billing events across Apple, Google, Amazon, Stripe, and Web Billing. Requires the RevenueCat Pro plan or above. Canonical: https://brew.new/browse/integrations/revenuecat - Direction: Data in - Connection: Webhook secret - IAP + subscription events ## Live capabilities - **Event triggers**: Trigger Brew automations from RevenueCat in-app purchase, subscription, and billing events across Apple, Google, Amazon, Stripe, and Web Billing. ## Related - [All integrations](https://brew.new/browse/integrations.md) - [Sign up](https://brew.new/sign-up.md) - Website: https://brew.new --- # Blog posts # Abandoned Cart Emails: How to Recover 10-15% of Lost Sales *Published: 2025-10-25* **Abandoned cart emails** try to recover a checkout the shopper started and left. [Baymard Institute](https://baymard.com/lists/cart-abandonment-rate) has long published cart-abandonment rates near 70% across large checkout studies. Recovery rates vary by offer, timing, and list. Treat any 10-15% recovery claim as a range other operators report, not a guarantee. This is commercial mail. It is not [transactional](/blog/transactional-emails). The shopper never finished the purchase. ## Key takeaways - Name the product they left and make it easy to return to that cart. Vague "you left something" copy wastes the only fact you have. - Abandoned cart is a campaign, not a receipt. [CAN-SPAM](https://www.ftc.gov/business-guidance/resources/can-spam-act-compliance-guide-business) looks at the primary purpose. - Time the first mail while the cart is still true. Then stop. A week of nagging is a [win-back](/blog/win-back-emails) problem, not a cart problem. - Measure recovered checkouts, not opens. See [email metrics that matter](/blog/email-metrics-that-matter). ## Why Carts Get Abandoned Understanding abandonment helps you address objections: | Reason | % of Shoppers | How to Address | | ------------------ | ------------- | -------------------------- | | Unexpected costs | 48% | Show total cost early | | Required account | 24% | Offer guest checkout | | Complex checkout | 18% | Simplify process | | Didn't trust site | 17% | Add trust signals | | Just browsing | 15% | Incentivize first purchase | | Found better price | 8% | Price match or incentive | Your abandoned cart emails should address these objections directly. ## The Ideal Abandoned Cart Sequence ### Email 1: The Reminder (1-2 Hours) **Goal:** Catch people who got distracted **Subject line examples:** - "You left something behind..." - "Your cart is waiting" - "Forget something?" - "Still thinking it over?" **Content elements:** - Show abandoned products with images - Clear CTA back to cart - No discount yet - Helpful tone, not pushy **Why 1-2 hours:** Immediate emails feel aggressive. Waiting too long loses urgency. 1-2 hours is the sweet spot—enough time to return naturally, not so long they've forgotten. ### Email 2: Address Objections (24 Hours) **Goal:** Overcome hesitation **Subject line examples:** - "Still thinking about [Product]?" - "Have questions? We can help" - "Here's what others say about [Product]" **Content elements:** - Customer reviews/testimonials - FAQ answers (shipping, returns) - Product benefits reminder - Support contact info ### Email 3: Create Urgency (48-72 Hours) **Goal:** Push toward decision **Subject line examples:** - "Your cart expires soon" - "Low stock alert: [Product]" - "Last chance for your cart" - "Before it's gone..." **Content elements:** - Urgency messaging (cart expires, low stock) - Small incentive (free shipping) - Strong CTA - Clear deadline ### Email 4: Final Offer (5-7 Days) **Goal:** Last conversion attempt **Subject line examples:** - "10% off to complete your order" - "A special offer just for you" - "We really want you to have this" **Content elements:** - Discount code - Clear expiration - Final call to action - Easy checkout link ## Abandoned Cart Email Templates ### Email 1: The Reminder **Subject:** You left something behind Hi [Name], Looks like you left some items in your cart: [Product Image] **[Product Name]** - $[Price] Your cart is saved, but these items are popular. [Complete Your Order] ← Button Questions? Just reply to this email. [Signature] ### Email 2: Social Proof **Subject:** What customers say about [Product] Hi [Name], Still thinking about [Product]? Here's what other customers are saying: ⭐⭐⭐⭐⭐ "Best purchase I've made this year" — Sarah M. ⭐⭐⭐⭐⭐ "Exceeded my expectations" — Mike T. ⭐⭐⭐⭐⭐ "Worth every penny" — Lisa K. [Product Image] Your cart is still waiting: [Complete Your Order] ← Button [Signature] ### Email 3: Urgency **Subject:** Your cart expires in 24 hours Hi [Name], Your cart will expire tomorrow—we'd hate for you to miss out. [Product Image] **[Product Name]** This is one of our most popular items. Once your cart expires, we can't guarantee availability. [Complete Your Order Now] ← Button [Signature] ### Email 4: Final Discount **Subject:** 10% off to complete your order Hi [Name], We noticed you haven't completed your purchase. Here's a little incentive: **Use code COMEBACK10 for 10% off your order** [Product Image] **[Product Name]** - ~~$[Price]~~ $[Discounted Price] This code expires in 48 hours. [Claim Your Discount] ← Button [Signature] ## Timing Optimization ### Standard E-commerce Timing | Email | Timing | Purpose | | ------- | ----------- | ------------------------- | | Email 1 | 1-2 hours | Catch distracted shoppers | | Email 2 | 24 hours | Address objections | | Email 3 | 48-72 hours | Create urgency | | Email 4 | 5-7 days | Final discount offer | ### High-Value Cart Timing For carts over $200, slow down slightly: | Email | Timing | Purpose | | ------- | --------- | ------------------- | | Email 1 | 2-4 hours | Reminder | | Email 2 | 48 hours | Value reinforcement | | Email 3 | 5 days | Urgency | | Email 4 | 7-10 days | Discount offer | ### Luxury/Considered Purchase Timing For expensive items with longer decision cycles: | Email | Timing | Purpose | | ------- | ---------- | --------------------------- | | Email 1 | 4-6 hours | Helpful reminder | | Email 2 | 3 days | Educational content | | Email 3 | 7 days | Case study/reviews | | Email 4 | 10-14 days | Personal consultation offer | ## Advanced Abandoned Cart Tactics ### Segment by Cart Value Different value = different treatment: **Low value ($0-50):** - Standard sequence - Smaller/no discount - Focus on convenience **Medium value ($50-200):** - Extended sequence - Free shipping incentive - Social proof emphasis **High value ($200+):** - Longer timing - Bigger incentive - Personal touch option ### Segment by Customer Type **New visitors:** - Include trust signals - Add reviews and testimonials - First-purchase incentive **Returning customers:** - Reference past purchases - Loyalty benefits - Familiar, friendly tone **VIP customers:** - Personal attention - Exclusive offers - Priority support ### Dynamic Product Recommendations Add related products to recovery emails: - "You might also like..." - "Customers also bought..." - "Complete the look..." This can increase order value even during recovery. ## Measuring Recovery Success ### Key Metrics | Metric | Target | Excellent | | ------------------ | ------ | --------- | | Recovery rate | 10-15% | 20%+ | | Email 1 open rate | 40%+ | 55%+ | | Email 1 click rate | 10%+ | 15%+ | | Revenue per email | Track | Improve | ### Calculate Your Opportunity ``` Monthly abandoned carts × AOV × Recovery rate = Monthly recovered revenue Example: 1,000 abandoned carts × $100 AOV × 12% recovery = $12,000/month ``` ## Common Abandoned Cart Mistakes ### 1. Waiting Too Long First email should be 1-2 hours, not 24 hours. ### 2. Leading With Discounts Training customers to abandon for discounts destroys margins. **Fix:** Email 1-2 should have no discount. Only offer in Email 3-4. ### 3. Not Showing Products Visual reminders are essential. **Fix:** Always include product images and details. ### 4. Too Many Emails 4 emails is optimal. More becomes annoying. ### 5. No Mobile Optimization Most opens are mobile. **Fix:** Test on mobile before launching. ### 6. Missing Exit Conditions Don't keep emailing after they purchase. **Fix:** Exit automation on purchase. ## Technical Setup ### Required Data To run abandoned cart emails, you need: - Shopper email (capture early in checkout) - Cart contents (products, quantities, prices) - Cart value - Timestamp of abandonment ### Platform Integration Most e-commerce + email combinations work: - Shopify + Klaviyo/Brew - WooCommerce + Various - Custom + API integration See our [email marketing platforms comparison](/blog/email-marketing-platforms-compared) for options. ## Creating Abandoned Cart Sequences With AI Traditional approach: 1. Write 4 emails manually 2. Design each email 3. Set up automation 4. Test and refine With [Brew](/): 1. Describe your sequence 2. AI generates all 4 emails 3. Review and customize 4. Launch **Time saved:** Hours → Minutes ## Ready to Recover Lost Revenue? Abandoned cart automation is the highest-ROI email you can send. Every day without it is money left on the table. Try Brew free and create your abandoned cart sequence in minutes. For sequence design beyond the cart, see [triggered email automations](/blog/triggered-email-automations-guide) and [welcome sequences](/blog/welcome-email-sequence). To contact us about this article, write [Brew support](/support). ## How to use this page **An abandoned cart email** is commercial mail about a checkout the shopper left. **Cart abandonment** is leaving items in checkout without buying. Baymard has long published rates near 70%. **Recovery** is a completed checkout from that cart, not an open. For example, the first mail can wait one to two hours so the shopper can finish on their own. | Move | Why it matters | | --- | --- | | Name the product | Vague 'you left something' wastes the only fact you have | | Stop after the cart is stale | A week of nagging is a win-back problem | US commercial mail still follows the [FTC CAN-SPAM guide](https://www.ftc.gov/business-guidance/resources/can-spam-act-compliance-guide-business) and [16 CFR 316.3](https://www.ecfr.gov/current/title-16/chapter-I/subchapter-C/part-316/section-316.3). Bulk sending still has to meet [Gmail's sender requirements](https://support.google.com/a/answer/81126). To contact us about this article, use the support link above. ## Frequently asked questions ### What is an abandoned cart email? A promotional message sent after someone added items to a cart and left without buying. It asks them to finish the order they started. ### Is abandoned cart email transactional? No. [Klaviyo's docs](https://help.klaviyo.com/hc/en-us/articles/115005082927) and the [FTC primary-purpose test](https://www.ecfr.gov/current/title-16/chapter-I/subchapter-C/part-316/section-316.3) treat it as commercial. Include an unsubscribe. ### How many abandoned cart emails should I send? A short sequence is enough. A reminder, a proof or objection-handler, then a last notice. Pull anyone who buys off the sequence that day. ### What should I measure? Recovered checkouts and complaint rate. Opens are a weak metric after [Mail Privacy Protection](https://support.apple.com/en-us/HT212559). --- # What still applies when an agent sends your email *Published: 2026-08-24* An agent can draft, segment, and send faster than a coordinator. It does not get a courtesy lane at Gmail. Mailbox providers score **domains, IPs, and complaint rates**. They do not score whether a human clicked Send. If the agent ships a promotional body through the transactional IP, you inherit the same problem you would have created by hand, only sooner. This is a companion to the [email deliverability guide](/blog/email-deliverability), not a replacement. That post covers reputation, authentication, and list hygiene. This one is about the mistakes that show up when software is the one hitting send. ## Key takeaways - Mailbox providers score domains, IPs, and complaint rates. They do not care whether a human clicked Send. - [CAN-SPAM](https://www.ftc.gov/business-guidance/resources/can-spam-act-compliance-guide-business) still uses the [primary-purpose test](https://www.ecfr.gov/current/title-16/chapter-I/subchapter-C/part-316/section-316.3). "The model wrote it" is not a defense. - Give the agent a promotional identity for campaigns. Keep [transactional](/blog/transactional-emails) mail on a separate transactional-purpose domain and, if you can, a separate IP. - Read the first unattended campaign against the source brief. Every figure should appear in a doc you own. ![Architecture diagram of an agent reaching Brew over MCP to compose, send, and read results.](/blog/brew-mcp-architecture.png) *The agent is another client on the same sending identity. The inbox still sees your domain.* ## The law does not mention agents [CAN-SPAM](https://www.ftc.gov/business-guidance/resources/can-spam-act-compliance-guide-business) applies to commercial messages. The [primary-purpose test](https://www.ecfr.gov/current/title-16/chapter-I/subchapter-C/part-316/section-316.3) is about the subject line and the body, not about who typed them. If the agent writes a "your invoice is ready" subject and a launch offer in the body, the message is commercial. You still need an honest subject, a physical address, and a working unsubscribe. "The model wrote it" is not a defense. The same is true in reverse. A password reset the agent triggered on a real security event is still transactional, as long as the body stays on that event. ## What Gmail and Yahoo actually require [Gmail's sender requirements](https://support.google.com/a/answer/81126) apply once you are in the bulk-sender bucket. The page's current threshold is **5,000** or more recipients in a day to personal Gmail. Marketing mail then needs: - SPF, DKIM, and alignment with DMARC on a `p=none` or stricter policy. - One-click unsubscribe per [RFC 8058](https://www.rfc-editor.org/rfc/rfc8058.html), `List-Unsubscribe` and `List-Unsubscribe-Post`. - Spam rates under **0.1%**, and never at **0.3%**. Transactional mail is exempt from the one-click rule. It is not exempt from authentication, and it is not exempt from spam complaints if you load the template with a sale. [Yahoo](https://senders.yahooinc.com/best-practices/) asks high-volume senders to put marketing and transactional on **different IPs**, and preferably different domains. An agent that "just uses the default sender" will collapse that split the first week you let it run unattended. ## What to put in the agent's reach Give the agent a **promotional** sending identity for campaigns and automations. The domain you warm, the IP you expect complaints on, the unsubscribe headers you have tested. Keep **transactional** mail on a separate identity. In Brew that identity is a domain with a transactional sending purpose, and the mail itself is a published automation with a trigger. The agent can compose the template. The send path per event should still be `POST /v1/automations/triggers/{triggerEventId}/fire` with a JSON payload, not "send this draft to the list." Do not let the agent pick the From address from a pile of verified domains "to see what lands." That is how you train two reputations at once and understand neither. ## Review before the first unattended send The failure mode is not a typo. It is a fluent paragraph that cites a number you never published, or a subject that looks like a receipt. Before you let an agent send without a queue: 1. Read the first campaign against the source brief. Every figure should appear in a doc you own. 2. Send it to a seed list that includes Gmail, Yahoo, and Outlook. 3. Confirm the one-click header is present on promotional mail, and left off transactional mail on purpose. 4. Confirm the physical address is in the promotional footer. Cory Nelson at Harmonya told [Noah Adelstein](https://thegtmengineer.substack.com/p/how-to-turn-email-marketing-into) he did that first-campaign read before he trusted the loop. That is the correct order. ## The short version Speed is not a new channel. It is the same channel with less time to notice you mixed a sale into a receipt. Authenticate the domain, split the IPs, tell the agent which identity is which, and keep a human on the first send. The brief that stops invented numbers is in [how to brief an email agent](/blog/how-to-brief-an-email-agent). The Harmonya first-campaign read is in [how Harmonya turned a newsletter into a lifecycle program](/blog/harmonya-lifecycle-email). To contact us about this article, write [Brew support](/support). ## How to use this page **Agent-sent email** is still email. Mailbox providers do not give models a courtesy lane. **The primary-purpose test** is how US law decides if the message is commercial. **One-click unsubscribe** is a header requirement for bulk commercial mail, including RFC 8058. For example, a receipt that spends most of the body selling a plan is still a campaign. | Move | Why it matters | | --- | --- | | Split marketing from transactional | Folder names in the ESP do not decide the law | | Authenticate the domain | The agent does not skip SPF, DKIM, or DMARC | US commercial mail still follows the [FTC CAN-SPAM guide](https://www.ftc.gov/business-guidance/resources/can-spam-act-compliance-guide-business) and [16 CFR 316.3](https://www.ecfr.gov/current/title-16/chapter-I/subchapter-C/part-316/section-316.3). Bulk sending still has to meet [Gmail's sender requirements](https://support.google.com/a/answer/81126). To contact us about this article, use the support link above. ## Frequently asked questions ### Do agents get different inbox rules? No. [Gmail](https://support.google.com/a/answer/81126) and [Yahoo](https://senders.yahooinc.com/best-practices/) score the sending identity. An agent is another client on your domain. ### Does transactional mail skip one-click unsubscribe? Gmail's bulk-sender page exempts transactional mail from the one-click header. It does not exempt it from authentication or from spam complaints if you load the template with a sale. One-click itself is [RFC 8058](https://www.rfc-editor.org/rfc/rfc8058.html). ### What should an agent be allowed to send? Promotional campaigns and automations on the warmed marketing identity. Transactional sends through a published automation on a transactional-purpose domain, fired with `POST /v1/automations/triggers/{triggerEventId}/fire`. Do not let it pick From addresses "to see what lands." ### Why review the first unattended send? Fluent copy will cite numbers you never published, or wear a receipt-shaped subject on a sale. Read it against the brief, seed-test Gmail, Yahoo, and Outlook, then confirm the headers. --- # How AI Is Changing Email Automation *Published: 2025-10-14* **AI email automation** is transforming how businesses approach email marketing. The intersection of AI and email creates unprecedented opportunities—but only if you understand what's actually useful versus what's hype. This guide covers how AI is changing [email marketing automation](/blog/email-marketing-automation-guide): from content generation to predictive analytics to smart segmentation. You'll learn what capabilities matter, what the results look like, and how to get started. ## Key takeaways - AI automation goes past autoresponders: send-time prediction, generated copy, live behavior, and campaign tweaks from results. - The useful jobs are content drafts, churn and purchase predictions, [dynamic segments](/blog/email-segmentation), and [subject-line](/blog/email-subject-lines) scoring. - Start with brand examples, then high-impact flows: welcome, cart, post-purchase, and re-engagement. - Marketers still own strategy. AI handles repeat production and suggests the next test. ## The Evolution of Email Automation Email automation has come a long way from simple autoresponders. Today's AI-powered systems can: - **Predict optimal send times** for each individual subscriber - **Generate personalized content** at scale - **Adapt messaging** based on real-time behavior - **Optimize campaigns** automatically based on performance ## Key AI Capabilities in Email Marketing ### 1. Content Generation AI can now create entire email campaigns from a simple prompt: ``` Prompt: "Create a welcome email for new SaaS users that highlights our top 3 features" ``` The AI generates: - Subject line options - Email body copy - Call-to-action text - Design suggestions ### 2. Predictive Analytics AI analyzes historical data to predict: - Which subscribers are likely to churn - Who is ready to make a purchase - Optimal email frequency for each user - Best content types for different segments ### 3. Smart Segmentation Instead of manual segmentation, AI can: - Automatically group subscribers by behavior - Identify hidden patterns in your data - Create dynamic segments that update in real-time - Suggest new segments based on performance data ### 4. Subject Line Optimization AI evaluates subject lines for: - Emotional resonance - Spam score - Predicted open rate - Brand voice alignment ## Real-World Impact Companies using AI-powered email automation are seeing: - **40% higher open rates** through send time optimization - **60% reduction** in email creation time - **3x improvement** in conversion rates - **50% decrease** in unsubscribe rates --- **Ready to try AI email automation?** [See how Brew creates campaigns in 60 seconds →](https://getbrew.ai) --- ## Getting Started with AI Email Automation ### Step 1: Choose the Right Platform Look for email platforms that offer: - Native AI content generation - Predictive analytics - Smart automation builders - Easy integration with your tech stack ### Step 2: Start with Your Brand The best AI email tools learn from your brand: - Upload brand guidelines - Provide example emails - Define your brand voice - Set design preferences ### Step 3: Begin with High-Impact Automations Focus on automations with the biggest ROI: - [Welcome sequences](/blog/welcome-email-sequence) - [Abandoned cart flows](/blog/abandoned-cart-emails) - Post-purchase series - Re-engagement campaigns ### Step 4: Iterate and Improve AI gets better with data: - Monitor performance metrics - Provide feedback on generated content - Test different approaches - Let the AI learn from results ## The Future of AI in Email We're just scratching the surface. Coming innovations include: - **Hyper-personalized journeys**: Every subscriber gets a unique experience - **Predictive content**: AI anticipates what subscribers want before they know it - **Cross-channel intelligence**: Email AI that learns from all customer touchpoints - **Voice and visual AI**: Generate matching assets across all channels ## Conclusion AI is not replacing email marketers—it's empowering them. By handling the repetitive tasks and providing intelligent insights, AI allows marketers to focus on strategy and creativity. Ready to experience AI-powered email automation? [Try Brew free →](https://brew.new) Triggered flows still beat a weekly blast. See [triggered email automations](/blog/triggered-email-automations-guide) for the event-based version of this work. [CAN-SPAM](https://www.ftc.gov/business-guidance/resources/can-spam-act-compliance-guide-business) still applies when the body sells. [Gmail's sender requirements](https://support.google.com/a/answer/81126) still apply when volume climbs. To contact us about this article, write [Brew support](/support). ## How to use this page **AI email automation** is using a model to draft or route mail that a trigger still starts. **A brief** is the proof and constraints the model may not invent past. **A generated cart email** is still a campaign, not a receipt. For example, an agent can write the three cart reminders if you paste the product name and ban extra stats. | Move | Why it matters | | --- | --- | | Keep the trigger | AI does not replace the event that starts the flow | | Review the first send | Fluent copy can still invent a number | US commercial mail still follows the [FTC CAN-SPAM guide](https://www.ftc.gov/business-guidance/resources/can-spam-act-compliance-guide-business) and [16 CFR 316.3](https://www.ecfr.gov/current/title-16/chapter-I/subchapter-C/part-316/section-316.3). Bulk sending still has to meet [Gmail's sender requirements](https://support.google.com/a/answer/81126). To contact us about this article, use the support link above. ## Frequently asked questions ### What is AI email automation? AI email automation is software that writes or times messages, groups people from behavior, and adjusts campaigns from results, instead of only firing a fixed autoresponder. It can draft a welcome email from a prompt, guess a send time, or flag someone likely to churn. A person still sets the goal and approves the send. ### What can AI actually do in an email program? The jobs that exist today are content drafts, predicted send times, live message changes from behavior, automatic campaign tweaks, segmentation from patterns, and subject-line scoring. That is a production and targeting layer on top of the automation guide, not a replacement for a list you own. ### Which automations should you start with? Start with the flows that already pay for themselves: a welcome sequence, abandoned cart mail, post-purchase follow-up, and re-engagement. Feed the tool brand examples first. Then iterate from clicks and unsubscribes, not from a single generated draft you never edit. ### Does AI change CAN-SPAM or inbox rules? No. If the message sells, treat it as commercial under [CAN-SPAM](https://www.ftc.gov/business-guidance/resources/can-spam-act-compliance-guide-business). Authenticate the domain and watch spam complaints the way [Gmail](https://support.google.com/a/answer/81126) asks. A generated cart email is still a campaign. A generated receipt still has to stay a receipt. ### Is AI replacing email marketers? The post's claim is narrower. AI takes repeat production and offers a read on the data. People still pick the offer, the brand voice, and which flow to build. Strategy and taste stay human. The tool fills the blank page and suggests the next test. --- # AI Email Marketing Tools: What Actually Works *Published: 2025-11-19* **AI email marketing tools** promise to revolutionize how you create and send emails. But with every platform slapping "AI-powered" on their features, it's hard to know what's real and what's marketing hype. This guide separates genuine AI capabilities from buzzwords. We'll show you exactly what AI can do for email marketing today, which tools deliver, and how to use them effectively. ## Key takeaways - AI in email is several jobs: copy, full layouts, send-time models, predictions, and [subject-line](/blog/email-subject-lines) scores. "AI-powered" alone means nothing. - Humans still own brand voice, strategy, sensitive copy, and final approval. - Ask what the model does, request samples, count leftover human work, and check whether it learns your brand. - Start with one feature, keep a review step, and measure time saved plus clicks, not vanity labels. - Platform-native AI is usually lighter: subject suggestions, send-time toggles, predicted segments. Dedicated tools go deeper on one job. ## What AI Actually Does in Email Marketing AI in email marketing falls into distinct categories. Understanding these helps you evaluate tools: ### 1. Content Generation AI writes email copy, subject lines, and complete campaigns. **What works well:** - Subject line generation and optimization - First-draft email copy - Variations for [A/B testing](/blog/email-ab-testing) - Product descriptions from data **What still needs humans:** - Brand voice perfection - Strategic messaging decisions - Sensitive or nuanced content - Final approval and editing ### 2. Complete Email Generation More advanced AI creates entire emails—design, copy, and structure. **Current capabilities:** - Full email generation from prompts - Brand-consistent designs - Mobile-responsive layouts - Template-free creation **Leader in this space:** [Brew](/) generates complete, on-brand emails from simple descriptions. ### 3. Send Time Optimization AI analyzes when individual subscribers engage and optimizes delivery. **Typical results:** 10-20% improvement in open rates **How it works:** Machine learning models predict optimal send times per subscriber based on historical engagement. ### 4. Predictive Analytics AI predicts subscriber behavior: who will buy, churn, or engage. **Use cases:** - Target high-value prospects - Prevent churn before it happens - Personalize based on predicted interests - Optimize segment targeting ### 5. Subject Line Optimization AI predicts which subject lines will perform best. **Typical results:** 5-15% open rate improvement **How it works:** Models trained on performance data score subject lines before sending. ## AI Email Marketing Tools Compared ### Brew: Full Email Generation Brew offers the most comprehensive AI approach. Describe what you want, get a complete email. **AI capabilities:** - Complete email generation from prompts - Brand voice learning and consistency - Automatic mobile optimization - [Automation workflows](/blog/email-marketing-automation-guide) from natural language - Design and copy in one step **What makes it different:** Other tools offer AI assistance. Brew offers AI creation—the AI does the work, you review. **Pricing:** - Free: 200 credits - Starter: $49/month (1,000 contacts, 500 credits) - Pro: $199/month (10,000 contacts, 10,000 credits) **Best for:** Teams wanting to eliminate manual email creation entirely. ### Jasper: AI Copywriting Jasper focuses on generating email copy rather than complete designs. **AI capabilities:** - Email copy generation - Subject line creation - Tone adjustment - Multiple content types **Limitations:** Generates text only—you still need to design emails separately. **Best for:** Writers wanting AI assistance with copy, not design. ### Phrasee: Enterprise Subject Lines Phrasee uses AI specifically for subject lines and email copy at enterprise scale. **AI capabilities:** - Subject line optimization - Performance prediction - Language generation at scale - Brand voice learning **Limitations:** Enterprise pricing, focused narrowly on copy optimization. **Best for:** Large enterprises with high email volumes and big budgets. ### Seventh Sense: Send Time AI Seventh Sense optimizes when to send emails to each subscriber. **AI capabilities:** - Individual send time optimization - Engagement prediction - Email fatigue detection - Throttling to prevent overload **Limitations:** Only optimizes timing—doesn't help with content. **Best for:** Companies with large lists wanting better timing. ### Platform-Native AI (Mailchimp, Klaviyo, etc.) Most major platforms now include AI features: **Typical capabilities:** - Subject line suggestions - Send time optimization - Basic content suggestions - Predictive segmentation **Limitations:** Usually surface-level AI, not transformative. ## What AI Features Actually Matter Based on real-world impact, here's what moves the needle: ### High Impact 1. **Complete email generation** — Saves 80%+ of creation time 2. **Subject line optimization** — 10-30% open rate improvement 3. **Send time optimization** — 10-20% engagement improvement ### Medium Impact 1. **Copy suggestions** — Good starting points, needs editing 2. **Predictive segmentation** — Better targeting 3. **A/B test analysis** — Faster optimization cycles ### Lower Impact (Currently) 1. **Content personalization AI** — Still developing 2. **Fully autonomous campaigns** — Needs human oversight 3. **Creative optimization** — Inconsistent results ## How to Evaluate AI Email Tools When evaluating AI claims, ask: ### 1. What Specifically Does the AI Do? "AI-powered" is meaningless. Ask: Does it write? Design? Optimize timing? Predict behavior? ### 2. Can You See Examples? Request sample outputs. Quality varies dramatically between tools. ### 3. How Much Human Work Remains? AI that saves 20% of time is different from AI that saves 90%. ### 4. Does It Learn From Your Data? Generic AI vs. AI trained on your brand and audience—significant difference. ### 5. What's the Quality Level? Some AI output needs heavy editing. Some is ready to send. Understand which you're getting. ## Implementing AI in Your Email Marketing ### Step 1: Identify Your Biggest Time Sink Where do you spend the most time? - **Writing copy** → Jasper or Brew - **Designing emails** → Brew - **Optimizing send times** → Seventh Sense - **Everything** → Brew ### Step 2: Start With One AI Feature Don't try everything at once. Master one capability, measure results, then expand. ### Step 3: Maintain Quality Control AI isn't perfect. Build review processes: - Human review before sending - Brand voice checklist - Error catch workflow ### Step 4: Measure the Impact Track: - Time saved per email - Performance changes (opens, clicks, conversions) - Total campaigns sent (velocity often increases) ## The Future of AI in Email Marketing We're seeing rapid advancement: - **Visual generation**: AI creating custom images and graphics - **Hyper-personalization**: Unique content for each subscriber - **Autonomous optimization**: Self-improving campaigns - **Cross-channel AI**: Unified intelligence across email, SMS, ads Within 2-3 years, manual email creation will feel as outdated as hand-coding HTML emails does today. ## Getting Started With AI Email Marketing ### If You're New to AI Start with Brew's free tier—experience full email generation without commitment. Create a few emails, evaluate quality, decide if AI fits your workflow. ### If You Already Use an ESP Most platforms offer AI features. Enable what's available, measure impact. If results are limited, consider dedicated AI tools. ### If You're Ready for Full AI Brew offers the most complete AI email experience. Describe what you want, get professional emails in seconds. ## Ready to Try AI-Powered Email? The gap between AI-created and manually-created emails has closed. What took hours now takes seconds—without sacrificing quality. Try Brew free and create your first AI-generated email in under 60 seconds. A [brief](/blog/how-to-brief-an-email-agent) is how you stop invented claims. [MCP](https://modelcontextprotocol.io) is how another assistant can drive the same brand. See [Brew MCP](/blog/brew-mcp). Live commercial mail still needs [CAN-SPAM](https://www.ftc.gov/business-guidance/resources/can-spam-act-compliance-guide-business) and [Gmail](https://support.google.com/a/answer/81126) authentication. To contact us about this article, write [Brew support](/support). ## How to use this page **An AI email marketing tool** is software that drafts or routes mail, not a magic inbox pass. **A brief** is how you stop the model from inventing a logo or a statistic. **Generated promotional mail** is still commercial mail. For example, a subject-line generator is a helper. An agent that sends still needs authentication. | Move | Why it matters | | --- | --- | | Ask what it writes | Copy help is not the same as a sending platform | | Keep the legal split | A receipt that sells is still a campaign | US commercial mail still follows the [FTC CAN-SPAM guide](https://www.ftc.gov/business-guidance/resources/can-spam-act-compliance-guide-business) and [16 CFR 316.3](https://www.ecfr.gov/current/title-16/chapter-I/subchapter-C/part-316/section-316.3). Bulk sending still has to meet [Gmail's sender requirements](https://support.google.com/a/answer/81126). To contact us about this article, use the support link above. ## Frequently asked questions ### What do AI email marketing tools actually do? They split into copy drafts, full email generation, send-time models, predicted behavior, and subject-line scores. Platform-native features are usually lighter: a suggested subject, a send-time toggle, a predicted segment. Ask which of those jobs the product performs. "AI-powered" is not an answer. ### Which AI email features have the most impact? This post ranks complete email generation, subject-line scoring, and send-time models as high impact. Copy suggestions, predicted segments, and faster A/B test reads sit in the middle. Fully autonomous campaigns and creative optimization still need a person. Impact here means time saved or a clearer next test, not a promised lift. ### How should you evaluate an AI email tool? Ask what the model does in concrete terms. Request sample outputs. Count how much editing remains. Check whether it learns your brand and audience or only a generic voice. Quality ranges from a usable draft to a sendable email. Walk through that gap before you move a live list. ### Should you replace your ESP with an AI tool? Not always. If you already have an ESP, turn on the native features and measure. If those stay shallow, add a dedicated generator or timing tool. If creation time is the whole bottleneck, a full-generation product is the one to trial. Keep a human review before send either way. ### Does AI change compliance or inbox rules? No. Generated promotional mail is still commercial under [CAN-SPAM](https://www.ftc.gov/business-guidance/resources/can-spam-act-compliance-guide-business). Bulk sending still has to authenticate and stay under [Gmail's](https://support.google.com/a/answer/81126) spam-rate ceilings. A brief should mark promotional versus [transactional](/blog/transactional-emails) so the model does not mix a receipt with a sale. --- # B2B Email Marketing: The Complete Strategy Guide *Published: 2025-11-14* **B2B email marketing** operates differently than B2C. Longer sales cycles, multiple stakeholders, and higher stakes mean your strategy needs more sophistication—but the fundamentals still apply. This guide covers everything for B2B: strategy, sequences, templates, and optimization tactics that work for complex sales. ## Key takeaways - B2B cycles run weeks to months, with several buyers. Lead with education, not a same-day sale. - The funnel is awareness newsletters, consideration proof, then decision mail: trials, demos, proposals. - Build the list from gated work and events. Buying or scraping a list wrecks [deliverability](/blog/email-deliverability). - Core sequences are [welcome](/blog/welcome-email-sequence), long nurture, product education, events, case studies, and sales follow-up. - Measure replies, meetings, and pipeline, not only opens. ## B2B vs B2C Email Marketing | Factor | B2B | B2C | | --------------- | --------------------------- | ------------------- | | Sales cycle | Weeks to months | Minutes to days | | Decision makers | Multiple stakeholders | Usually one | | Email tone | Professional, value-focused | Casual, emotional | | Primary goal | Lead nurturing | Direct sales | | List building | Quality over quantity | Scale matters | | Content depth | In-depth, educational | Quick, visual | | Relationship | Long-term partnership | Transaction-focused | ## The B2B Email Funnel ### Top of Funnel (Awareness) **Goal:** Establish expertise, build trust **Email types:** - Newsletter content - Industry insights - Educational resources - Thought leadership - Trend analysis **Frequency:** Weekly or bi-weekly ### Middle of Funnel (Consideration) **Goal:** Demonstrate value, address objections **Email types:** - Case studies - Product comparisons - Webinar invitations - Demo offers - ROI calculators **Frequency:** 2-3 emails per week during active nurture ### Bottom of Funnel (Decision) **Goal:** Convert to customer **Email types:** - Free trial offers - Consultation requests - Pricing information - Implementation guides - Proposal follow-ups **Frequency:** Based on sales stage ## Essential B2B Email Types ### 1. Welcome Sequence New subscribers need context and value immediately. See our complete welcome email sequence guide with B2B-specific timing. ### 2. Nurture Campaigns Long-term relationship building: - Educational content series - Industry news digests - Expert interviews - Best practice guides ### 3. Product Education Demonstrate value: - Feature spotlights - Use case examples - Customer success stories - Comparison content ### 4. Event Invitations Drive engagement: - Webinar invitations - Conference meetups - Product launches - Exclusive briefings ### 5. Case Studies Social proof for B2B: - Customer success stories - [ROI data](/blog/email-marketing-roi) - Implementation examples - Industry-specific results ### 6. Sales Sequences Convert warm leads: - Meeting requests - Demo offers - Proposal follow-ups - Contract renewals ## Building Your B2B Email List ### High-Quality Lead Generation **Effective tactics:** - Gated content (whitepapers, reports, tools) - Webinar registrations - Free tool access - Industry research - Demo requests - Newsletter signups **Avoid:** - Buying lists (destroys deliverability and conflicts with [CAN-SPAM](https://www.ftc.gov/business-guidance/resources/can-spam-act-compliance-guide-business)) - Scraping (violates platform terms) - Cold outreach without permission ### Lead Magnet Ideas for B2B | Lead Magnet | Conversion Rate | Lead Quality | | --------------- | --------------- | ------------ | | Industry report | 15-25% | High | | ROI calculator | 10-20% | Very High | | Template bundle | 20-35% | Medium | | Free tool | 15-30% | High | | Webinar | 20-40% | High | | Checklist | 25-40% | Medium | See our [email list building guide](/blog/email-list-building) for more tactics. ## B2B Email Sequences ### Welcome Sequence (B2B Timing) **Email 1** (Immediate): Deliver lead magnet, company introduction **Email 2** (Day 3): Company story, mission, unique value **Email 3** (Day 7): Best educational content **Email 4** (Day 10): Case study relevant to their industry **Email 5** (Day 14): Soft offer (consultation, demo) ### Lead Nurture Sequence For leads not ready to buy: **Monthly rhythm:** - Week 1: Educational content - Week 2: Industry news/insights - Week 3: Case study or customer story - Week 4: Product-related (soft touch) ### Re-engagement Sequence For leads gone cold: **Email 1** (Day 0): "We haven't heard from you" **Email 2** (Day 3): Valuable resource (no ask) **Email 3** (Day 7): Direct question about interest **Email 4** (Day 14): Last chance before removal ## B2B Email Templates ### Cold Outreach Template **Subject:** [Specific observation about their company] Hi [Name], I noticed [specific observation—recent news, content they published, company milestone]. At [Your Company], we help [similar companies] [achieve specific outcome]. For example, we recently helped [Company] [specific result with numbers]. Would a 15-minute call to explore whether we could help [Company] make sense? [Signature] ### Follow-Up Template **Subject:** Following up on [topic] Hi [Name], I wanted to follow up on my previous email about [topic]. I came across this [resource/article/data] that seemed relevant to [their initiative]: [Link or key insight] Would you be open to a brief conversation about [specific value prop]? [Signature] ### Case Study Email **Subject:** How [Company] achieved [specific result] Hi [Name], Thought you'd find this relevant—[Company] was facing [problem you solve]. Here's what they did: • [Action 1] • [Action 2] • [Action 3] The result: [Specific outcome with numbers] [Link to full case study] Seeing similar challenges at [Their Company]? Happy to discuss how we might help. [Signature] ### Webinar Invitation **Subject:** You're invited: [Webinar title] Hi [Name], On [Date], we're hosting a live session on [topic]: **[Webinar Title]** [Date] | [Time] | [Duration] You'll learn: • [Takeaway 1] • [Takeaway 2] • [Takeaway 3] [Register Now Button] Can't make it live? Register anyway—we'll send the recording. [Signature] ## B2B Email Best Practices ### Personalization That Works Go beyond first name: - Company name and context - Industry-specific challenges - Reference their content or news - Acknowledge their role - Mention mutual connections ### Optimal Send Times for B2B | Day | Performance | Best Times | | --------- | ----------- | --------------- | | Tuesday | Best | 8-10 AM | | Wednesday | Great | 8-10 AM, 2-4 PM | | Thursday | Good | 8-10 AM | | Monday | OK | Avoid early | | Friday | Lower | Morning only | Always test with your specific audience. [A/B testing](/blog/email-ab-testing) on [subject lines](/blog/email-subject-lines) is the cheap way to check a weekday guess. ### Subject Line Tips B2B subject lines should be: - Professional (minimal emojis) - Specific about value - Relevant to their business - Curiosity-driving but not clickbait Examples: - "[Name], quick question about [topic]" - "How [Similar Company] solved [problem]" - "[Industry] trends for [Year]" - "Idea for [Their Company]" See our email subject lines guide for more. ## Measuring B2B Email Success ### Key Metrics | Metric | Good | Excellent | | ------------------ | ----- | --------- | | Open rate | 20%+ | 30%+ | | Click rate | 2.5%+ | 5%+ | | Reply rate | 1%+ | 3%+ | | Meeting conversion | 2%+ | 5%+ | ### Beyond Opens and Clicks Track business impact: - Leads generated - Sales conversations started - Pipeline influenced - Revenue attributed - Customer acquisition cost ### Segment Performance Compare metrics across: - Industry segments - Company size - Lead source - Funnel stage ## Common B2B Email Mistakes ### 1. Selling Too Soon B2B buyers need education before sales pitches. **Fix:** Lead with value, earn the right to sell. ### 2. Too Formal Professional doesn't mean robotic. **Fix:** Write like you talk. People buy from people. ### 3. Wall of Text Busy professionals skim emails. **Fix:** Short paragraphs, bullets, clear structure. ### 4. No Clear CTA Every email needs one obvious next step. **Fix:** One CTA per email, prominently placed. ### 5. Inconsistent Sending Sporadic emails lose mindshare. **Fix:** Regular cadence builds trust and recognition. ## B2B Email Tools For creating B2B emails efficiently, [Brew](/) generates professional emails while maintaining your brand voice. **Why Brew for B2B:** - Generate nurture sequences quickly - Maintain consistent professional tone - Create personalized emails at scale - [Automate workflows](/blog/email-marketing-automation-guide) from simple prompts **Pricing:** - Free: 200 credits - Starter: $49/month (1,000 contacts, 500 credits) - Pro: $199/month (10,000 contacts, 10,000 credits) ## Getting Started ### Week 1: Foundation - Audit your current B2B email performance - Define your segments - Map your buyer journey ### Week 2: Build Core Sequences - Create welcome sequence - Plan nurture content - Draft key templates ### Week 3: Launch and Monitor - Activate automations - Send first campaigns - Track initial metrics ### Week 4: Optimize - Review performance - A/B test subject lines - Refine based on data Ready to upgrade your B2B email marketing? Try Brew free and create professional sequences in minutes. Authenticate before you scale a cold sequence. [Gmail's sender requirements](https://support.google.com/a/answer/81126) still apply when volume climbs. [Win-back](/blog/win-back-emails) is a different job from first-touch nurture. To contact us about this article, write [Brew support](/support). ## How to use this page **B2B email marketing** is permission-based mail to people who buy for a company. **A buying committee** is more than one person, so one blast rarely fits all of them. **A meeting** is often a better conversion than a click. For example, a category manager and a VP of procurement need different proof in the same account. | Move | Why it matters | | --- | --- | | Write for the role | Industry plus title beats 'newsletter list' | | Track replies and pipeline | Opens will not close a six-month deal | US commercial mail still follows the [FTC CAN-SPAM guide](https://www.ftc.gov/business-guidance/resources/can-spam-act-compliance-guide-business) and [16 CFR 316.3](https://www.ecfr.gov/current/title-16/chapter-I/subchapter-C/part-316/section-316.3). Bulk sending still has to meet [Gmail's sender requirements](https://support.google.com/a/answer/81126). To contact us about this article, use the support link above. ## Frequently asked questions ### How is B2B email marketing different from B2C? B2B cycles last weeks to months and several people share the decision. Tone stays professional and value-focused. The job is nurture, not a same-day checkout. You want fewer, better contacts. Content runs longer and more educational. B2C is faster, more emotional, and more visual. ### What emails belong in a B2B program? A welcome sequence, long nurture, product education, event invites, case studies, and sales follow-up. Top-of-funnel mail teaches. Middle-of-funnel mail proves the work. Bottom-of-funnel mail asks for a trial, demo, or proposal. Frequency follows the stage, not a fixed daily cadence. ### How should a B2B team build an email list? Use gated reports, webinars, tools, demos, and a newsletter people asked for. Skip bought lists and scraped addresses. Those wreck deliverability and fail the permission test in [CAN-SPAM](https://www.ftc.gov/business-guidance/resources/can-spam-act-compliance-guide-business). Quality beats a large cold file. List building has more tactics. ### What B2B email mistakes should you avoid? Selling in the first note, writing like a press release, sending a wall of text, hiding the next step, and sending whenever you remember. Lead with a useful point. Write the way you talk. Use short paragraphs. Put one CTA on the page. Keep a cadence so people still recognize the name. ### Which B2B email metrics matter? Opens and clicks are a start. Replies, meetings, pipeline, attributed revenue, and acquisition cost are the business read. Compare those by industry, company size, lead source, and funnel stage. [Gmail](https://support.google.com/a/answer/81126) still publishes spam-rate ceilings, so complaint rate belongs on the same dashboard. --- # 10 Best Email Marketing Software in 2026: Complete Comparison Guide *Published: 2025-10-22* Choosing the **best email marketing software** shouldn't require weeks of research. Yet with dozens of platforms claiming to be #1, finding the right fit feels overwhelming. This guide cuts through the noise. We've tested the 10 best email marketing platforms and break down exactly who each one serves best. Whether you're a startup, e-commerce store, or enterprise, you'll find your match here. ## Key takeaways - Score tools on ease of use, features, price, [deliverability](/blog/email-deliverability), and support, not on a single "best" badge. - Match the job: AI drafts, e-commerce revenue, creator simplicity, heavy [automation](/blog/email-marketing-automation-guide), or an all-in-one CRM. - [Welcome](/blog/welcome-email-sequence) and [abandoned cart](/blog/abandoned-cart-emails) automations drive a large share of email revenue when the software can actually send them. - Try two or three tools on a real campaign before you commit. Switching later is painful. ## How We Evaluated Email Marketing Software We assessed each platform on five criteria: 1. **Ease of use**: How quickly can you create and send emails? 2. **Features**: Automation, segmentation, templates, analytics 3. **Pricing**: Value for money at different scales 4. **Deliverability**: Do emails actually reach inboxes? Check [Gmail's sender requirements](https://support.google.com/a/answer/81126). 5. **Support**: Quality of help when you need it ## The 10 Best Email Marketing Platforms ### 1. Brew — Best for AI-Powered Email Creation Brew reimagines email marketing. Instead of manually building emails, you describe what you want and AI generates complete, on-brand emails in seconds. **Key Features:** - AI generates emails from simple prompts - Automatic brand consistency - Automation workflows from natural language - A/B testing and analytics - Basic CRM (similar to Mailchimp's) **Pricing:** - **Free**: 200 credits, preview only - **Starter**: $49/month (1,000 contacts, 500 credits) - **Pro**: $199/month (10,000 contacts, 10,000 credits) **Best for**: Teams who want professional emails without the time investment. Ideal if design and copywriting aren't your strengths. **Verdict**: The fastest path from idea to sent email. Try it free. ### 2. Mailchimp — Best Known Platform The most recognized name in email marketing with a comprehensive feature set. **Key Features:** - Drag-and-drop builder - Marketing CRM - Landing pages - 300+ integrations - Social posting **Pricing:** Free (500 contacts), paid from $13/month **Best for**: Small businesses wanting an established, trusted platform with extensive resources. ### 3. Klaviyo — Best for E-commerce Klaviyo owns e-commerce email with deep Shopify and WooCommerce integration. **Key Features:** - E-commerce data sync - Predictive analytics - Revenue attribution - SMS marketing - [Advanced segmentation](/blog/email-segmentation) **Pricing:** Free (250 contacts), paid from $20/month **Best for**: Online stores serious about email-driven revenue. ### 4. ConvertKit — Best for Creators Built specifically for bloggers, YouTubers, and course creators. **Key Features:** - Visual automation builder - Landing pages - Paid newsletter support - Creator network - Simple interface **Pricing:** Free (1,000 subscribers), paid from $15/month **Best for**: Individual creators who prioritize simplicity. ### 5. ActiveCampaign — Best Automation The most powerful automation builder available. **Key Features:** - Advanced workflow automation - Built-in CRM - Site tracking - Predictive sending - Machine learning features **Pricing:** Starting at $29/month (no free tier) **Best for**: [B2B companies](/blog/b2b-email-marketing) with complex nurture sequences. ### 6. HubSpot — Best All-in-One A complete marketing suite with email as one component. **Key Features:** - Full CRM (free) - Marketing automation - Sales tools - Service hub - Extensive training **Pricing:** Free tools, Marketing Hub from $20/month **Best for**: Growing companies wanting an all-in-one platform. See our [HubSpot vs Mailchimp](/blog/hubspot-vs-mailchimp) comparison. ### 7. Brevo — Best Budget Option Solid features at competitive prices (formerly Sendinblue). **Key Features:** - Pay-per-email option - SMS marketing - Transactional emails - Chat widget - GDPR-focused **Pricing:** Free (300 emails/day), paid from $25/month **Best for**: Budget-conscious businesses needing email + SMS. ### 8. MailerLite — Best Simple Platform Clean, modern, and easy to use. **Key Features:** - Intuitive builder - Website builder - Automation - Paid newsletters - Generous free tier **Pricing:** Free (1,000 subscribers), paid from $10/month **Best for**: Small businesses wanting simplicity at low cost. ### 9. Constant Contact — Best for Events Strong event marketing alongside email. **Key Features:** - Event management - Social marketing - Surveys - Long track record - Phone support **Pricing:** Starting at $12/month **Best for**: Organizations running events and local businesses. ### 10. Campaign Monitor — Best Templates Beautiful, professionally designed templates. **Key Features:** - Premium template library - Agency features - Strong deliverability - Clean analytics - Easy branding **Pricing:** Starting at $11/month **Best for**: Design-conscious brands wanting polished emails fast. ## Email Marketing Software Comparison Table | Platform | AI Features | Free Plan | Starting Price | Best For | | ---------------- | ----------- | --------- | -------------- | ----------- | | Brew | ✅ Full | ✅ | $49/mo | AI creation | | Mailchimp | Limited | ✅ | $13/mo | General use | | Klaviyo | Limited | ✅ | $20/mo | E-commerce | | ConvertKit | ❌ | ✅ | $15/mo | Creators | | ActiveCampaign | Limited | ❌ | $29/mo | Automation | | HubSpot | Limited | ✅ | $20/mo | All-in-one | | Brevo | Limited | ✅ | $25/mo | Budget | | MailerLite | Limited | ✅ | $10/mo | Simplicity | | Constant Contact | ❌ | ❌ | $12/mo | Events | | Campaign Monitor | ❌ | ❌ | $11/mo | Templates | ## How to Choose the Best Email Marketing Software ### Step 1: Define Your Primary Need - **Speed**: Brew (AI creates emails in seconds) - **E-commerce**: Klaviyo (best revenue tracking) - **Simplicity**: MailerLite or ConvertKit - **Automation**: ActiveCampaign - **All-in-one**: HubSpot - **Budget**: Brevo ### Step 2: Consider Your Technical Ability | Technical Level | Best Platforms | | --------------- | ---------------------------------- | | Beginner | Brew, MailerLite, ConvertKit | | Intermediate | Mailchimp, Brevo, Campaign Monitor | | Advanced | ActiveCampaign, Klaviyo, HubSpot | ### Step 3: Factor in Growth Choose software that scales. Switching platforms is painful—pick one that works at 10x your current list size. ### Step 4: Try Before You Buy Most platforms offer free trials or free tiers. Test your top 2-3 choices with a real campaign before committing. ## What About Email Marketing ROI? Email marketing delivers [$36-42 for every $1 spent](/blog/email-marketing-roi)—the highest ROI of any marketing channel. The right software helps you capture this potential. Key ROI factors: - **Automation**: Welcome sequences and abandoned cart emails drive disproportionate revenue - **Deliverability**: Emails must reach inboxes to convert. [CAN-SPAM](https://www.ftc.gov/business-guidance/resources/can-spam-act-compliance-guide-business) still wants an honest subject and a working unsubscribe. - **Efficiency**: Time saved on creation = more campaigns sent ## Our Top Recommendations **For speed and efficiency**: Brew — AI eliminates manual email creation **For e-commerce**: Klaviyo — Best revenue attribution and data integration **For creators**: ConvertKit — Purpose-built simplicity **For complex B2B**: ActiveCampaign — Unmatched automation power **For tight budgets**: MailerLite or Brevo — Most value per dollar ## Ready to Get Started? The best email marketing software is the one you'll actually use. If you want professional results without spending hours on design and copywriting, try Brew free. Create your first AI-powered email in under 60 seconds. Compare HubSpot and Mailchimp in more depth in HubSpot vs Mailchimp. [Email A/B testing](/blog/email-ab-testing) is how you prove the tool's editor actually improves results. To contact us about this article, write [Brew support](/support). ## How to use this page **Email marketing software** is the stack that holds the list, sends the mail, and shows the result. **The right tool** is the one that matches sending volume, automation, and who will actually use it. **Deliverability** is a domain problem as much as a vendor logo. For example, a cheap editor that cannot authenticate your domain will fail the first bulk send. | Move | Why it matters | | --- | --- | | Score sending first | Pretty templates do not reach the inbox by themselves | | Check export and auth | You should leave with the list and a working domain | US commercial mail still follows the [FTC CAN-SPAM guide](https://www.ftc.gov/business-guidance/resources/can-spam-act-compliance-guide-business) and [16 CFR 316.3](https://www.ecfr.gov/current/title-16/chapter-I/subchapter-C/part-316/section-316.3). Bulk sending still has to meet [Gmail's sender requirements](https://support.google.com/a/answer/81126). To contact us about this article, use the support link above. ## Frequently asked questions ### What is the best email marketing software? There is no single winner. This guide scores ten platforms by job: AI creation, e-commerce, creators, automation, all-in-one CRM, budget, simplicity, events, and templates. Pick the job first. Then compare price at your list size and whether you can send a real campaign during a trial. ### How did you evaluate email marketing platforms? Five criteria: how fast you can create and send, automation and segmentation features, price at different scales, whether mail reaches the inbox, and support quality. Deliverability is a sending-domain problem as much as a vendor logo. [Gmail](https://support.google.com/a/answer/81126) still sets the bulk-sender floor. ### Which software is best for e-commerce versus B2B? Klaviyo is the e-commerce pick here: store data, revenue attribution, SMS, and cart flows. ActiveCampaign is the B2B pick for long nurture. HubSpot is the all-in-one if you want CRM plus email. See B2B email marketing if the sales cycle is measured in weeks. ### Do I need AI features in an email platform? Only if creation time is the bottleneck. Several tools add subject suggestions. A few generate the whole email. If you already have designers and a template system, native AI is optional. If a blank canvas is why you send fewer campaigns, generation is the feature to try first. ### Will the right software guarantee email ROI? No. Software helps you capture return. It does not create a list or a permission path. Welcome sequences and abandoned cart emails do more of the revenue work than the editor skin. [CAN-SPAM](https://www.ftc.gov/business-guidance/resources/can-spam-act-compliance-guide-business) still applies to the commercial sends those flows produce. --- # Your AI assistant can now run your email marketing *Published: 2026-06-26* If you run email, you've probably noticed how much of your day now happens inside an AI assistant. You're drafting copy in Claude, brainstorming subject lines in ChatGPT, maybe asking it to pull together a quick report. The work keeps drifting into the chat window. So we made Brew work there too. ## Key takeaways - Brew speaks [MCP](https://modelcontextprotocol.io). An assistant can compose, send, automate, and read results on a brand you already set up. - The inbox still sees your domain. Agents do not get a courtesy lane. Read [what still applies when an agent sends](/blog/agent-email-deliverability). - Brief the agent the way you would brief a designer. See [how to brief an email agent](/blog/how-to-brief-an-email-agent). - Connection URL is `https://brew.new/api/mcp`. OAuth, or a brand-scoped API key from Settings, then API. Brew now speaks [MCP](https://modelcontextprotocol.io), the Model Context Protocol. In plain terms, that means your AI assistant can do real work in Brew instead of just talking about it. Connect it once and you can ask your assistant to write a campaign, send it to the right people, set up an automation, or pull last week's numbers. It's the same brand, the same lists, and the same emails you'd build in the app. You just don't have to be the one clicking around to make it happen. ## What it can actually do This isn't a chatbot handing you advice. It's your assistant doing the job. Once Brew is connected, you can ask it to: - Write on-brand emails from a quick description, then keep tweaking until they're right - Send campaigns and test sends through your verified domain, now or scheduled for later - Build automations like a welcome series or a win-back flow, and turn them on - Add contacts, manage custom fields, and build a segment just by describing who you want to reach - Pull the numbers that actually matter, like opens, clicks, bounces, and how your automations are doing - Generate on-brand images and GIFs for your emails It all stays on-brand because it's running on the same Brew that already learned your colors, fonts, and voice. An email your assistant sends looks exactly like one you'd build by hand, and it still goes out through your verified domain and lands in the same reports you already check. ![A diagram showing Claude, Cursor, Viktor, and any AI assistant connecting through one Brew connection to the full email toolkit.](/blog/brew-mcp-architecture.png) _One connection, and your assistant gets the whole Brew toolkit._ ## Getting connected In ChatGPT, first enable Developer mode under Settings → Security and login. Then open Settings → Plugins, select `+`, and create Brew with server URL `https://brew.new/api/mcp` and OAuth authentication. Sign in to Brew, select a brand, approve the disclosed full-brand access, start a new chat, enable `@Brew`, and call `get_brew_capabilities`. If an older ChatGPT app version reports an invalid schema, refresh its actions. Remove and re-add Brew only after the corrected endpoint is deployed and only if refresh does not work. In Claude, Grok, Cursor, Codex, or another assistant that supports MCP OAuth, add the same server URL, sign in with Brew, select a brand, and call `get_brew_capabilities` to verify the connection. If your assistant doesn't support OAuth connectors yet, drop the same URL into its config and authenticate with a brand-scoped API key from Settings, then API instead: ```json { "mcpServers": { "brew": { "url": "https://brew.new/api/mcp", "headers": { "Authorization": "Bearer brew_YOUR_API_KEY" } } } } ``` Either way, it works with ChatGPT, Claude, Grok, Cursor, Codex, and other clients that support remote Streamable HTTP MCP servers. ## How marketers are using it The most common thing we see is people running campaigns in conversation. You type something like "write a launch email for the new feature, match our brand, and schedule it for Tuesday at 9am to everyone on a trial." Your assistant drafts it in Brew, shows you a preview, and schedules the send. You never leave the chat. The next step is letting it run on its own. Some teams hand the keys to an AI "employee" like [Viktor](https://viktor.com), which lives in Slack and does real work, and give it a standing job: watch for new signups, drop them into the welcome series, and post the open rates to the channel every Monday. Brew handles the email, Viktor keeps an eye on things, and the team just reviews what came back. You're still the one calling the shots on strategy and creative. You're just not stuck doing the repetitive parts anymore. ## Even better with your CRM Brew is good at one thing: being the memory and the muscle behind your email program. It gets a lot more useful when your assistant runs it next to the other tools you already have. Connect Brew to a CRM or enrichment tool like [Clay](https://clay.com) or [Attio](https://attio.com), and your assistant can handle the whole loop without you copying anything between tabs. It pulls fresh data on a contact, saves that context to their Brew profile, builds a segment around it, sends a campaign written for that group, then checks how it did and adjusts for next time. That's real personalization, and you didn't have to babysit any of it. ![A diagram showing an AI assistant using Brew, Clay, and Attio together to enrich, segment, design, send, and measure.](/blog/brew-mcp-compose.png) _Your assistant runs Brew next to your CRM: enrich, segment, design, send, measure, repeat._ ## Give it a try The Brew MCP is live today. Grab a key in Settings, then API, connect it to your assistant, and ask it to send your next campaign. I think you'll be surprised how much of your week it can take off your plate. - [Explore the Brew MCP](/mcp) - [Read the setup docs](https://docs.brew.new/api-reference/mcp/overview) Give it a shot and let me know how it goes. Thomas The Harmonya loop that uses this connection is in [how Harmonya turned a newsletter into a lifecycle program](/blog/harmonya-lifecycle-email). To contact us about this article, write [Brew support](/support). ## How to use this page **The Brew MCP** is a hosted Model Context Protocol server that lets an assistant drive one brand. **A brand-scoped connection** is the rule that the assistant cannot see another customer's list. **A verified sending domain** is still required for live campaigns. For example, Claude can ask Brew to draft a welcome series, then you confirm the from-address before it sends. | Move | Why it matters | | --- | --- | | Connect one brand | Do not hand an assistant the whole workspace | | Confirm the send identity | MCP does not create a courtesy inbox lane | US commercial mail still follows the [FTC CAN-SPAM guide](https://www.ftc.gov/business-guidance/resources/can-spam-act-compliance-guide-business) and [16 CFR 316.3](https://www.ecfr.gov/current/title-16/chapter-I/subchapter-C/part-316/section-316.3). Bulk sending still has to meet [Gmail's sender requirements](https://support.google.com/a/answer/81126). To contact us about this article, use the support link above. ## Frequently asked questions ### What is the Brew MCP? A hosted [Model Context Protocol](https://modelcontextprotocol.io) server at `https://brew.new/api/mcp`. It lets ChatGPT, Claude, Cursor, and other MCP clients run Brew on one brand. ### Does the assistant send email itself? It calls Brew. Sends still go through your verified domain and the same reports you already check. Live campaigns need a domain you own. ### Can it build automations? Yes. Welcome series, [win-back](/blog/win-back-emails) flows, segments, and analytics readback are in the same connection. Confirm the send identity before you let it run unattended. ### How do I connect it? In ChatGPT, enable Developer mode, add the server URL with OAuth, pick a brand, then call `get_brew_capabilities`. In Claude, Cursor, or another MCP client, use the same URL. Docs live at [docs.brew.new](https://docs.brew.new/api-reference/mcp/overview). --- # Email A/B Testing: The Complete Guide to Better Results *Published: 2025-11-21* **Email A/B testing** transforms opinions into data. Instead of guessing what works, you let your audience tell you through their actions. This guide covers everything from basic split testing to advanced multivariate strategies. By the end, you'll know how to run tests that drive real improvements. ## Key takeaways - Change one element per test. A subject plus a send time in the same split tells you nothing. - Test [subject lines](/blog/email-subject-lines), send time, and from name first. Content, CTAs, and images come next. - Pick the metric before you send: opens for subjects, clicks for body and CTA, both for send time. - Wait for enough recipients and enough time. Peeking early and tiny lists produce fake winners. - Document the winner and the next test. Small improvements compound when you keep a log. ## What Is Email A/B Testing? A/B testing (split testing) sends two versions of an email to see which performs better. You change one element, measure results, and apply the winner. **Simple example:** - Version A: "50% off everything today" - Version B: "Half off everything today" - Send each to 50% of your list - Measure opens - Winner becomes your template ## What to A/B Test (Priority Order) ### High Impact: Test These First **Subject Lines** - Biggest impact on opens - Easiest to test - Quick results - Compounds over time **Send Time** - When does your audience engage? - Test days and times - Consider time zones - Platform-specific optimization **From Name** - Company name vs. person - Different team members - Descriptive vs. simple - Major impact on trust ### Medium Impact **Email Content** - Short vs. long - Formal vs. casual - Story vs. direct - Educational vs. promotional **CTA (Call to Action)** - Button text - Button color - Placement (top vs. bottom) - Single vs. multiple CTAs **Images** - With images vs. text-only - Product vs. lifestyle photos - Image placement - Number of images ### Lower Impact (But Worth Testing) - Preview text - Personalization depth - Social proof placement - Footer content - Sender email address ## How to Run Valid A/B Tests ### Step 1: Test One Variable Only Change only one element per test. If you test subject line AND send time, you won't know which caused the difference. **Good test:** Subject A vs. Subject B (same everything else) **Bad test:** Subject A at 9am vs. Subject B at 2pm ### Step 2: Determine Sample Size Statistical significance requires adequate recipients: | Total List Size | Per Variant | Confidence Level | | --------------- | ----------- | ---------------- | | 1,000 | 300+ | Moderate | | 5,000 | 500+ | Good | | 10,000+ | 1,000+ | High | **Rule of thumb:** 1,000+ per variant for reliable results. ### Step 3: Define Your Success Metric What defines "winning"? | Test Type | Primary Metric | | ------------ | ------------------------ | | Subject line | Open rate | | Content | Click rate | | CTA | Click rate or conversion | | Send time | Open rate + click rate | Treat subject-line opens as noisy after [Apple Mail Privacy Protection](https://support.apple.com/en-us/HT212559) prefetches images. Confirm with clicks when the change is in the body. ### Step 4: Run the Test - Send variants simultaneously (controls for time) - Wait for adequate data (2-4 hours minimum for opens, 24 hours for clicks) - Don't peek and call early ### Step 5: Analyze Results Is the difference statistically significant? **Quick rule:** If winner is 5%+ better with 1,000+ recipients per variant, it's likely significant. For precise analysis, use a statistical significance calculator. Pair the result with [metrics that still matter](/blog/email-metrics-that-matter) after prefetch clients inflate opens. ### Step 6: Apply and Document - Use the winner going forward - Document what you learned - Plan the next test - Build a testing knowledge base ## Subject Line A/B Test Ideas ### Test Ideas with Examples **1. Personalization** - A: "Your weekly update" - B: "[Name], your weekly update" **2. Curiosity vs. Clarity** - A: "The #1 mistake marketers make" - B: "Avoid this common email mistake" **3. Numbers** - A: "Tips to improve your emails" - B: "5 tips to improve your emails" **4. Emoji** - A: "New products just dropped" - B: "🎉 New products just dropped" **5. Length** - A: "Sale" - B: "Our biggest sale of the year starts now" **6. Question vs. Statement** - A: "Ready for better email results?" - B: "Get better email results today" ## CTA A/B Test Ideas ### Button Text Tests | Version A | Version B | | ------------------ | ---------------------- | | "Buy Now" | "Get Yours" | | "Learn More" | "See How It Works" | | "Start Free Trial" | "Try Free for 14 Days" | | "Download" | "Get Your Free Copy" | ### Button Placement Tests - Above the fold only - Multiple buttons (top and bottom) - Single button at bottom - Inline link vs. button ### Button Design Tests - Brand color vs. contrasting color - Large vs. standard size - With arrow icon vs. without ## Content A/B Test Ideas ### Length Tests - Short (100 words) vs. long (500+ words) - Often depends on offer complexity - B2B may prefer longer; B2C shorter ### Format Tests - Single column vs. multi-column - Text-heavy vs. image-heavy - Bullet points vs. paragraphs - Numbered steps vs. prose ### Tone Tests - Formal vs. conversational - First person vs. second person - Emotional vs. logical ## Common A/B Testing Mistakes ### 1. Testing Too Many Variables One change at a time. Multiple changes = meaningless results. ### 2. Ending Tests Too Early At least 2-4 hours for opens, 24 hours for clicks. B2B may need 48 hours. ### 3. Declaring Winners Prematurely Statistical significance requires sample size. Small lists mean less certainty. ### 4. Not Documenting Results Create a testing log. Patterns emerge over time. ### 5. Testing Trivial Things Focus on high-impact elements first. Button shade differences rarely matter. ### 6. Never Testing at All Some testing beats no testing. Start somewhere. ## Building a Testing Calendar ### Weekly Tests - Subject line variations - Send time optimization ### Monthly Tests - CTA optimization - Content format ### Quarterly Tests - Major design changes - From name/sender - Segment-specific messaging ## Measuring Test Impact Track improvements over time: | Metric | Baseline | After 3 Months | Improvement | | ---------- | -------- | -------------- | ----------- | | Open rate | 20% | 25% | +25% | | Click rate | 2% | 3% | +50% | | Conversion | 1% | 1.5% | +50% | Small improvements compound. 10% better opens × 10% better clicks = 21% more conversions. ## Advanced Testing Strategies ### Multivariate Testing Test multiple variables simultaneously: - Subject line × Send time - CTA text × CTA color **Requirements:** Large lists and statistical software. ### Holdout Testing Keep a percentage that never gets optimization. Compare long-term performance to measure cumulative impact. ### Sequential Testing Build improvements incrementally: - Week 1: Optimize subject line → Winner - Week 2: Optimize CTA → Winner - Week 3: Optimize content → Winner ## A/B Testing With AI [Brew](/) makes testing easier: - AI generates subject line variants automatically - Quick iteration on content options - Consistent quality across variants Instead of spending hours writing test variants, describe what you want and AI creates options. ## Getting Started 1. **Pick your first test** — Subject line is easiest 2. **Set up in your platform** — Most ESPs support A/B testing 3. **Run the test** — Send to equal groups 4. **Analyze results** — Statistical significance check 5. **Apply and repeat** — Use winners, test new elements Ready to optimize your emails? Try Brew free and create test variants in seconds with AI. A winning variant still has to reach the inbox. [Email deliverability](/blog/email-deliverability) and [Gmail's sender requirements](https://support.google.com/a/answer/81126) do not pause for a test. Commercial tests still need a working unsubscribe under [CAN-SPAM](https://www.ftc.gov/business-guidance/resources/can-spam-act-compliance-guide-business). To contact us about this article, write [Brew support](/support). ## How to use this page **Email A/B testing** is sending two versions to a sample and keeping the one that wins on a named metric. **A valid test** is one change, a large enough sample, and a metric that is not a prefetch pixel. **A winner** still has to reach the inbox. For example, a subject-line test should not also change the hero image. | Move | Why it matters | | --- | --- | | Change one thing | Mixed tests cannot tell you what worked | | Confirm with clicks | Opens moved after Mail Privacy Protection | US commercial mail still follows the [FTC CAN-SPAM guide](https://www.ftc.gov/business-guidance/resources/can-spam-act-compliance-guide-business) and [16 CFR 316.3](https://www.ecfr.gov/current/title-16/chapter-I/subchapter-C/part-316/section-316.3). Bulk sending still has to meet [Gmail's sender requirements](https://support.google.com/a/answer/81126). To contact us about this article, use the support link above. ## Frequently asked questions ### What is email A/B testing? A/B testing sends two versions of an email to see which performs better. You change one element, measure the result, and keep the winner. A simple case is two subject lines to equal halves of the list, then you count opens. The same method works for send time, from name, body, and CTA once you pick the matching metric. ### What should you A/B test first? Start with subject lines, then send time, then from name. Those change opens and trust with the least production work. After that, test length, tone, CTA text and placement, and images. Preview text and footer tweaks are lower impact. One change per test, or you cannot name the cause. ### How do you run a valid email A/B test? Test one variable. Send both variants at the same time. Size the groups so each variant has enough recipients. Define the win metric before you send. Wait at least a few hours for opens and a day for clicks. Do not call a winner from a peek. Document the result and plan the next test. ### Why can open rate mislead a subject-line test? [Apple Mail Privacy Protection](https://support.apple.com/en-us/HT212559) prefetches images, including tracking pixels, when the message arrives. That can count as an open whether or not anyone read the mail. Use opens as a directional signal. Confirm with clicks when the test is about the body, and read email metrics that matter. ### What mistakes ruin A/B tests? Testing two things at once, stopping after a couple of hours, declaring a winner on a tiny list, skipping the log, and arguing over button shade are the usual failures. Some testing still beats none. Keep the calendar simple: subjects weekly, CTAs monthly, design quarterly. --- # The Email Canvas: A Faster Way to Create and Compare Emails *Published: 2026-04-18* Most email tools still treat email creation like editing a document. You open one template, make one set of changes, preview it, and move on. That workflow breaks down when you use AI. AI makes it easy to generate multiple directions at once. A document-style editor hides those options. A canvas makes them visible. ## Key takeaways - An email canvas shows generated campaigns side by side so you compare directions instead of editing one file at a time. - The useful loop is generate, compare, pick, edit, version, then send or export. - Canvas view is for creative comparison. List view is for finding an old campaign in a large library. - Cheap variants change behavior. Teams test more [subject lines](/blog/email-subject-lines) and layouts when each option does not need a rebuild. - From the canvas you can test-send, schedule, export HTML, or push to a connected ESP without leaving the comparison view. ## What is an email canvas? An **email canvas** is a visual workspace where generated emails appear side by side. Instead of working through one email at a time, you can see the whole set of ideas in one place. For example, you can ask for: > Create three [welcome email](/blog/welcome-email-sequence) directions. One should feel polished and enterprise, one should feel warm and founder-led, and one should feel bold and product-forward. A canvas lets you compare all three immediately. ## Why canvas workflows work better with AI AI changes the bottleneck. The hard part is no longer getting a first draft. The hard part is choosing the best direction and refining it quickly. A canvas helps because it supports the new workflow: 1. Generate multiple options 2. Compare them visually 3. Pick a winner 4. Edit manually or with AI 5. Save versions 6. Send, schedule, or export This is more natural than creating one email, closing it, creating another, and trying to remember what changed. ## What teams can do on the canvas In Brew, each generated campaign appears on the canvas. From there, teams can: - Select an email and edit it with AI - Click into text, links, buttons, and images to edit manually - Generate variants without losing the original - Resize previews for mobile, tablet, and desktop - Send a test preview - Send now or schedule for later - Export to connected ESPs like HubSpot, Klaviyo, Mailchimp, or Brevo - Copy, download, or export HTML - Add an email to templates - Set an email as the reference design for future generations The canvas becomes the production space, not just a preview. ## Canvas view vs list view A canvas is best for creative work. It helps when you need to compare directions, spot design issues, and work with multiple versions. A list view is best for operations. It helps when you need to scan many emails, find a specific campaign, or manage a larger library. Most teams need both: | Task | Best view | | --------------------------- | --------- | | Generate new campaigns | Canvas | | Compare creative directions | Canvas | | Edit copy and layout | Canvas | | Review many emails quickly | List view | | Find a specific old email | List view | | Manage a large library | List view | ## Why this matters for marketing velocity The best email teams run more experiments. They test angles, subject lines, formats, and CTAs. [A/B testing](/blog/email-ab-testing) is cheaper when the variants already sit on one canvas. Traditional editors make that expensive because each variation requires production time. A canvas makes variation cheap. You can create three options, refine the best one, and send it without rebuilding anything. That speed changes behavior. Teams stop waiting for the perfect brief and start testing more ideas. ## Where Brew fits [Brew](/) uses a canvas because AI email creation should not feel like filling out a template. You describe what you want, Brew generates the emails, and the canvas becomes the place where you compare, edit, send, and export. If your team is using AI to create emails, a canvas is not just a nicer interface. It is the workflow that makes AI useful. A short [brief](/blog/how-to-brief-an-email-agent) still beats a vague prompt. Commercial sends still need an honest subject and a working unsubscribe under [CAN-SPAM](https://www.ftc.gov/business-guidance/resources/can-spam-act-compliance-guide-business). Inbox placement still follows [Gmail's bulk-sender rules](https://support.google.com/a/answer/81126). To contact us about this article, write [Brew support](/support). ## Frequently asked questions ### What is an email canvas? An email canvas is a visual workspace where generated emails sit side by side. You see the whole set of ideas in one place instead of opening one template, previewing it, and closing it. That is the layout that matches AI, because the model can draft several directions at once and you need to compare them. ### Why does a canvas work better with AI email? The bottleneck moved. First drafts are cheap. Choosing a direction and refining it is the work. A canvas supports generate, compare, pick, edit, save versions, then send or export. A document editor hides the other options, so you rebuild the same email to try a second angle. ### When should you use canvas view versus list view? Use the canvas when you are generating campaigns, comparing creative directions, or editing copy and layout. Use list view when you need to scan many emails, find a specific old campaign, or manage a large library. Most teams need both. Creative work and operations are different jobs. ### What can you do from the canvas after an email is generated? You can edit with AI or by clicking text, links, buttons, and images. You can generate variants, resize previews, send a test, send or schedule, export HTML, or push to a connected ESP. You can also save a template or set one email as the reference design for later generations. ### Does a canvas replace a brief or inbox rules? No. A canvas makes variants cheap. It does not invent strategy. Write a brief before you generate. Treat live commercial mail as commercial under [CAN-SPAM](https://www.ftc.gov/business-guidance/resources/can-spam-act-compliance-guide-business). A test send still has to authenticate the way [Gmail](https://support.google.com/a/answer/81126) expects. --- # Email Deliverability: The Complete Guide to Reaching the Inbox *Published: 2025-12-10* **Email deliverability** is the foundation of email marketing. If your emails don't reach the inbox, nothing else matters—your brilliant copy, beautiful design, and perfect timing are all wasted. This guide explains what affects deliverability, how to diagnose problems, and how to fix them. ## Key takeaways - Delivery rate is whether the receiving server accepted the message. Inbox placement is whether it landed in the inbox. Those are different numbers. - Authentication is SPF, DKIM, and DMARC alignment. [Gmail's bulk-sender page](https://support.google.com/a/answer/81126) requires that once you send 5,000 or more messages a day to personal Gmail. - Keep spam reports under 0.1%, and never hit 0.3%. Split marketing and [transactional](/blog/transactional-emails) mail the way [Yahoo](https://senders.yahooinc.com/best-practices/) asks, on different IPs if you can. - List hygiene and complaint rate beat clever subject lines. Agents do not get a courtesy lane. See [what still applies when an agent sends](/blog/agent-email-deliverability). ## What Is Email Deliverability? Deliverability measures whether your emails reach subscribers' inboxes versus spam folders or being blocked entirely. **Critical distinction:** - **Delivery rate** = emails accepted by receiving servers (usually 95%+) - **Inbox placement** = emails that actually reach the inbox (often much lower) You can have 99% delivery rate but terrible inbox placement. The emails "delivered" might all be in spam. ## What Affects Deliverability ### 1. Sender Reputation Your sender reputation is like a credit score for email. Mailbox providers track: | Factor | Impact | | -------------------------- | --------------- | | Spam complaints | High negative | | Bounce rates | High negative | | Engagement (opens, clicks) | High positive | | Spam trap hits | Severe negative | | Sending patterns | Medium | | List quality | High | Reputation is built over time and can take weeks to recover if damaged. ### 2. Authentication (SPF, DKIM, DMARC) Three protocols verify you're a legitimate sender: **SPF (Sender Policy Framework):** - Authorizes which servers can send for your domain - DNS TXT record listing approved senders - Prevents spoofing of your domain **DKIM (DomainKeys Identified Mail):** - Digital signature proving email wasn't modified - Cryptographic authentication - Builds sender reputation **DMARC (Domain-based Message Authentication):** - Policy for handling failed SPF/DKIM - Reporting mechanism for authentication failures - **Required** on your sending domain for bulk senders (~5,000+ messages/day to personal Gmail/Yahoo accounts) since February 1, 2024 — a `p=none` policy is the accepted floor. A missing DMARC record draws Gmail error `4.7.31`. ### 3. Content Quality Spam filters analyze your email content: | Factor | Watch Out For | | ------------------- | ----------------------------------- | | Spam trigger words | "Free," "act now," excessive claims | | Image-to-text ratio | Too many images, too little text | | Link quality | Shortened URLs, suspicious domains | | HTML quality | Broken code, hidden text | | Subject line | ALL CAPS, excessive punctuation | ### 4. List Quality Your list directly impacts reputation: | Metric | Target | Problem Zone | | --------------- | ------ | ------------ | | Bounce rate | <2% | >5% | | Spam complaints | <0.1% | >0.3% | | Engagement rate | >15% | <5% | | List age | Fresh | Years old | ## Google & Yahoo Bulk Sender Requirements (2024–2026) Since February 1, 2024, Google and Yahoo enforce a shared set of requirements. They apply to **all senders**, with stricter rules for **bulk senders** — anyone sending close to **5,000 messages or more per day to personal Gmail accounts** (counted across your whole primary domain, including subdomains; the status never expires once you cross it). **All senders must:** - Authenticate with **SPF or DKIM** - Have valid **forward and reverse DNS (PTR)** on the sending IP - Transmit over a **TLS** connection - Keep the [Postmaster Tools](https://postmaster.google.com) spam rate **below 0.3%** (aim for under 0.1%) - Format messages to **RFC 5322**, with a valid `Message-ID` and a single `From:` address that honestly identifies the sender (no deceptive `Re:`/`Fwd:` or emoji-as-graphics) **Bulk senders (~5,000+/day) must additionally:** - Set up **both SPF and DKIM** - Publish a **DMARC** record (`p=none` is an acceptable floor) - **Align** the `From:` domain with the SPF or DKIM domain - Support **one-click unsubscribe** — see below ### One-Click Unsubscribe (RFC 8058) Marketing and subscribed messages must include one-click unsubscribe headers so recipients can opt out from within Gmail/Yahoo in a single click: ``` List-Unsubscribe: List-Unsubscribe-Post: List-Unsubscribe=One-Click ``` - The unsubscribe URL **must accept a POST** request. A `mailto:`-only or plain-link unsubscribe does **not** satisfy the requirement. - A **clearly visible unsubscribe link must also exist in the message body**. - Honor unsubscribe requests **within 48 hours (2 days)**. - **Transactional messages** (password resets, receipts, reservation confirmations) are **exempt**. ### Enforcement is now permanent Through 2024–2025 non-compliant mail was mostly _deferred_ with temporary `4xx` errors. **As of November 2025, Google escalated to permanent `5xx` (550) rejections** — non-compliant messages are refused outright and not retried. Enhanced status codes tell you what failed: `5.7.25` (PTR), `5.7.27` (SPF), `5.7.29` (TLS), `5.7.30` (DKIM), `4.7.31` (missing DMARC), `4.7.32` (From misalignment), and `4.7.28` (rate limited — stop sending for at least 10 minutes). > **On Brew:** authentication, TLS, PTR, one-click unsubscribe headers, and body-link enforcement are handled for you on every marketing send. Publish your DNS records (including DMARC), keep your list clean, and watch your spam rate. ## Diagnosing Deliverability Problems ### Warning Signs 🚩 Open rates declining across all [segments](/blog/email-segmentation) 🚩 Bounce rate consistently above 2% 🚩 Spam complaints above 0.1% 🚩 Test emails landing in spam 🚩 Sudden engagement drops 🚩 Subscribers saying they didn't receive emails ### Diagnostic Tools **Google Postmaster Tools:** - Gmail-specific reputation data - Spam rate monitoring - Authentication status - Free to set up **Microsoft SNDS:** - Outlook/Microsoft reputation - IP-level data - Spam trap hits **MXToolbox:** - Authentication verification - Blacklist checking - DNS record validation **Mail Tester:** - Send test email, get score - Identifies specific issues - Quick sanity check ### Checking Authentication Run these checks: 1. SPF: Does your DNS have the correct TXT record? 2. DKIM: Is the signature valid and matching? (Gmail requires a key of 1024 bits or longer — 2048 recommended.) 3. DMARC: Do you have a policy? Is it enforced? 4. Reverse DNS (PTR): Does your sending IP resolve to a hostname, and does that hostname resolve back to the same IP? (Gmail requires matching forward + reverse DNS — error `5.7.25` when missing.) 5. TLS: Is mail transmitted over a TLS connection? (Required for all senders — error `5.7.29` when absent.) > On a managed platform like Brew, SPF and DKIM are generated for you, TLS and PTR are handled by the sending infrastructure, and DMARC is provided as a record to add. Your job is to publish the records and keep your spam rate low. ## Fixing Deliverability Problems ### Step 1: Fix Authentication **SPF Setup:** ``` v=spf1 include:youresp.com ~all ``` Replace with your email platform's SPF include. **DKIM Setup:** Follow your email platform's instructions to add the DKIM record. **DMARC Setup:** Start with monitoring: ``` v=DMARC1; p=none; rua=mailto:dmarc@yourdomain.com ``` Progress to enforcement over time: ``` v=DMARC1; p=quarantine; rua=mailto:dmarc@yourdomain.com ``` ### Step 2: Clean Your List Remove: - Hard bounces (immediately, automatically) - Soft bounces (after 3-5 consecutive) - Inactive subscribers (no engagement 6+ months) - Role addresses (info@, support@) - Known spam traps (purchased lists) - Obvious typos (gmial.com, yaho.com) ### Step 3: Improve Engagement Higher engagement = better reputation: - Segment and send relevant content - Optimize send times for your audience - [Write better subject lines](/blog/email-subject-lines) - Make unsubscribing easy (reduces complaints) - Send consistently (not erratically) ### Step 4: Warm Up New IPs/Domains If starting fresh or recovering: | Week | Daily Volume | Target | | ---- | ------------- | ------------------- | | 1 | 100-500 | Most engaged only | | 2 | 500-1,000 | Engaged subscribers | | 3 | 1,000-5,000 | Broader engaged | | 4 | 5,000-10,000 | Most active | | 5+ | Normal volume | Full list (engaged) | Monitor reputation closely throughout. ## Deliverability Best Practices ### List Building ✅ Double opt-in for best quality ✅ Clear expectations at signup ✅ Immediate [welcome email](/blog/welcome-email-sequence) ✅ Regular list hygiene ❌ Never buy lists ❌ Don't add people without consent ❌ Avoid aggressive scraped contacts ❌ Don't use old, cold lists ### Sending Practices ✅ Consistent sending schedule ✅ Segment for relevance ✅ Honor unsubscribes immediately ✅ Monitor engagement metrics ✅ Regular list cleaning ❌ Don't send to cold lists suddenly ❌ Avoid erratic volume spikes ❌ Never ignore bounce handling ❌ Don't send to entire list always ### Content Best Practices ✅ Clean, valid HTML ✅ Plain text version included ✅ Balanced text-to-image ratio ✅ Valid, reputable links ✅ Clear sender identity ❌ Don't use excessive link shorteners ❌ Avoid spam trigger words in excess ❌ Don't hide text (tiny fonts, white on white) ❌ Never mislead with subject lines ## Platform-Specific Tips ### Gmail - Sender reputation is domain-based - Engagement heavily weighted - Tabs (Promotions, Updates) affect visibility - Enable BIMI for brand visibility ### Microsoft/Outlook - IP reputation matters more - Use Microsoft SNDS for monitoring - Junk mail feedback loop available - Can be stricter than Gmail ### Yahoo/AOL - Slower to update reputation - Consistent sending helps - Authentication is critical - Complaint feedback loop available ## Recovering From Deliverability Issues ### If You're Hitting Spam 1. **Pause sending** — Stop making it worse 2. **Identify the cause** — Check reputation, authentication, content 3. **Fix root issues** — Clean list, improve content, verify authentication 4. **Warm up again** — Start with most engaged subscribers only 5. **Monitor closely** — Check placement with every send ### Recovery Timeline | Severity | Recovery Time | | ---------------------------- | ------------------------ | | Minor (temporary dip) | 1-2 weeks | | Moderate (reputation damage) | 4-6 weeks | | Severe (blacklisted) | 2-3 months or new domain | ## Maintaining Good Deliverability ### Weekly Checks - [ ] Review bounce rates - [ ] Check spam complaints - [ ] Monitor open rate trends - [ ] Verify sends are going out ### Monthly Checks - [ ] Review authentication records - [ ] Check Google Postmaster data - [ ] Clean inactive subscribers - [ ] Audit content practices ### Quarterly Checks - [ ] Full list hygiene pass - [ ] Re-engagement campaigns - [ ] Platform/IP reputation review - [ ] Authentication audit ## Deliverability and Email Platforms Different platforms have different deliverability characteristics: **Shared IP platforms** (most small business tools): - Reputation shared with other senders - Less control but easier setup - Quality varies by platform **Dedicated IP** (enterprise/high volume): - Your reputation only - More control but more responsibility - Requires warm-up [Brew](/) handles deliverability infrastructure for you, with best practices built into the platform. ## The Bottom Line Deliverability isn't glamorous, but it's essential. The best email in the world is worthless if it never reaches the inbox. Focus on: 1. **Proper authentication** (SPF, DKIM, DMARC) 2. **Clean, engaged lists** 3. **Relevant, wanted content** 4. **Consistent sending practices** Ready to send emails that actually arrive? Try Brew for email marketing with deliverability best practices built in. For the legal split, read transactional emails. For opens after Apple prefetch, read [email metrics that matter](/blog/email-metrics-that-matter). To contact us about this article, write [Brew support](/support). ## How to use this page **Email deliverability** is whether the message reaches the inbox, not whether the ESP says sent. **Authentication** is SPF, DKIM, and DMARC on the domain you send from. **A complaint rate** is the share of recipients who mark the mail as spam. For example, a bought list can look large in the ESP and still fail Gmail's spam-rate ceiling. | Move | Why it matters | | --- | --- | | Authenticate first | Volume without SPF, DKIM, and DMARC is a fast way to the spam folder | | Clean quiet names | Dead addresses teach mailbox providers the wrong lesson | US commercial mail still follows the [FTC CAN-SPAM guide](https://www.ftc.gov/business-guidance/resources/can-spam-act-compliance-guide-business) and [16 CFR 316.3](https://www.ecfr.gov/current/title-16/chapter-I/subchapter-C/part-316/section-316.3). Bulk sending still has to meet [Gmail's sender requirements](https://support.google.com/a/answer/81126). To contact us about this article, use the support link above. ## Frequently asked questions ### What is email deliverability? Deliverability is whether your mail reaches the inbox instead of spam or a block. A high delivery rate can still mean most accepted mail landed in junk. ### What does Gmail require from bulk senders? [Gmail's sender requirements](https://support.google.com/a/answer/81126) apply at 5,000 or more recipients a day to personal Gmail. Authenticate with SPF, DKIM, and DMARC. Put one-click unsubscribe on marketing mail. Stay under a 0.1% spam rate. ### Should transactional and marketing share an IP? [Yahoo](https://senders.yahooinc.com/best-practices/) tells high-volume senders to split them. A cart campaign that shares an IP with password resets puts receipts in the same reputation bucket as the campaign. ### What is one-click unsubscribe? [RFC 8058](https://www.rfc-editor.org/rfc/rfc8058.html). You need `List-Unsubscribe` plus `List-Unsubscribe-Post: List-Unsubscribe=One-Click`. A footer link that dumps someone on a preferences page is not enough. --- # email-design.md: how we made AI emails actually on-brand *Published: 2026-06-15* Your coding agent reads a `CLAUDE.md` before it touches a line of your code. Your docs site probably ships an `llms.txt` so models can find their way around. Half the repositories I cloned this year had an `AGENTS.md` sitting in the root, quietly briefing whatever agent showed up on how the project actually works. So here's the question that nagged at me for months: if every serious AI system now gets handed a markdown file before it does anything, why was the most important context we had — a customer's **brand** — still trapped in a database? This is the story of how we stopped storing brands as objects and started writing them as documents. We call the format `email-design.md`, and it's the single biggest reason Brew's emails actually look on-brand now. ## Key takeaways - Brand context for an email agent works better as a markdown document than as a hundred-field object. - `email-design.md` names signature moves, email-safe fonts, dark-mode palettes, and voice in prose a model can read. - Uploaded brand guides outrank guesses from the live site when the two disagree. - The measured win was the floor: fewer invented logos and product names, not a jump in the average score. - The same file can travel through [MCP](https://modelcontextprotocol.io) so another agent can edit the spec and ask for a new email. ## Markdown quietly became the language AI speaks I didn't notice this trend happening until it had already happened. In August 2025, OpenAI, Google, Cursor, and a handful of others agreed on [AGENTS.md](https://agents.md/) — one markdown file at the root of your repo that tells any coding agent how to build, test, and not break your project. It's now stewarded by the Linux Foundation and lives in north of 60,000 repositories. A year before that, Jeremy Howard proposed [llms.txt](https://llmstxt.org/): a markdown index that points a model at the parts of your site worth reading. Stripe ships one. Anthropic ships one. Cursor ships one — six hundred sites and counting. And Claude Code, the tool half my team lives in, reads a `CLAUDE.md` before every single task. Different files, same instinct. When you want to brief an AI, you hand it a markdown document. There are good reasons it shook out this way, and they're worth saying out loud because they're exactly the reasons it works for brand too. Models were trained on a web that is overwhelmingly markdown — every README, every Stack Overflow answer, every doc page. It's their native dialect. The headings give a model hierarchy for free: this is the main idea, this is a sub-point, this is a list of rules. It's readable by a human and parseable by a machine at the same time, so there's exactly one source of truth instead of a document and its database shadow. It's dense — fewer tokens than the equivalent JSON. And it never blows up on a missing bracket. A half-truncated markdown file is still useful; a half-truncated JSON object is garbage. Anthropic has a good phrase for the discipline of deciding what an agent should actually see — [context engineering](https://www.anthropic.com/engineering/effective-context-engineering-for-ai-agents). Once you start thinking that way, the format of your context stops being an implementation detail and becomes the whole game. We learned that the hard way. ## How we used to capture a brand: the brand object When we started Brew, we did the obvious engineer thing. A brand is structured data, right? So we modeled it. We had a table — call it the brand object. Dozens of design tokens for color and type, a tree of assets and links, and a sprawling style profile with — I am not exaggerating — well over a hundred fields, one for every visual dimension we could think of: palette, lighting direction, even camera angle, each decomposed into a named property with a type. Something would crawl a website, fill in the boxes, and then at generation time we'd read the boxes back out and reassemble them into something the model could use. On paper it was beautiful. Typed. Validated. Predictable. Every field had a home. In practice, the emails came out _fine_. The right blue. The right font. AA contrast, every time. And completely generic — a template with the brand's colors swapped in. They were technically on-brand and recognizably no one's. It took us embarrassingly long to understand why, and the answer is simple once you see it. A hex code tells you a brand uses `#2E5BBA`. It does not tell you that the brand puts a tiny mono eyebrow label above every headline, that it draws its dividers as 1px hairlines and never as boxes, that it writes its CTAs like terminal commands. That second category — the _signature moves_ — is the entire reason an email looks like it belongs to somebody. And there was no field for it. There was never going to be a field for it, because the next brand's signature move would be something we hadn't thought to add a column for. The schema captured **what** a brand looked like. It had no way to hold **why**. And anything it couldn't hold, it quietly threw away. And it went deeper than any one brand. To have a field, you need a strict definition for it — and then every brand you onboard, across every industry, size, and company type, has to fit that same set of fields. A law firm, a streetwear label, a Series B fintech, a neighborhood bakery, all squeezed into one rigid shape. But brands aren't shaped like a schema. They're fluid, a little contradictory, and gloriously specific. Forcing each one into identical boxes was fitting a round peg into a square hole — and the parts that wouldn't fit, usually the exact parts that made the brand _that brand_, got shaved off until they did. It got worse from there. Every new idea meant a migration. The nested JSON we shipped into the model's context was enormous and mostly noise — we were spending the agent's attention on field names and braces for a job that is fundamentally about taste and prose. We had built a very precise way to describe a brand that lost the brand in the process. ## What if the brand spec was just a document? The unlock, when it came, was almost too obvious. When you hire a senior brand designer, they don't hand you a JSON blob. They hand you a brand guide. A document — prose, with exact values where exact values matter, and the reasoning sitting in between. The hex codes are in there, but so is the _intent_: use the accent sparingly, let the photography breathe, never center the logo. So recently we stopped asking our extraction agent to fill in fields. We asked it to write the guide. We rebuilt brand extraction from scratch. Now an agent researches a brand the way a person would — reads the site, studies the screenshots, looks at the real emails — and then it does what a designer does at the end of all that research. It writes it down. The output isn't a row in a table. It's a document called `email-design.md`. ![Brew's brand extraction in progress: loading logos and colors, exploring 67 website pages, researching the company, and scanning real page content](/blog/email-design-md-extraction.png) *The extraction agent doing its homework — logos, colors, links, company research, and a sweep of real pages — before it writes a single line.* ## What `email-design.md` actually is It's exactly what it sounds like: a markdown file that describes how this one brand does email. Here's a trimmed version of what the agent writes: ```markdown # Acme — Email Design ## Brand Essence Acme is for engineers who are tired of marketing fluff. The emails read like good documentation: direct, confident, a little dry. ## Signature Moves - A mono eyebrow label in the accent color above every headline - Hairline dividers — never boxes, never drop shadows - Pill CTA, medium weight, no uppercase - One product screenshot per email, full-bleed, no frame # ...color system, typography, layout, voice, and component recipes follow ``` ![The Brew app showing a rendered email-design.md — the 'Brew — Email Design System' page with brand essence and a color system — beside the chat that generated the brand](/blog/email-design-md-editor.png) *The same kind of file, rendered in Brew — in fact, Brew's own. The agent wrote it from the site; edit any line and the next email follows.* Three things matter about this file, and all three are things the old object couldn't do. It's **agent-written**. The extraction agent authors it from research, the same way a designer would write up a brand after a discovery call. You don't fill out a form. It's **human-editable**. It's just markdown. If the agent gets your voice slightly wrong, you open the file and fix the line. You don't file a ticket to add a database column — you edit a sentence. And it's the **source of truth**. When Brew generates an email, this document goes into the agent's context as the thing to follow — exact hex values, exact fallback font stacks, the signature moves, the voice. Not a reconstruction assembled from tokens. The actual document, read the way it was written. And it's email-safe by construction. Everything written into it is something that actually survives the inbox — which, as anyone who has shipped an email knows, is a far shorter list than what works on the web. Which gets at the thing I most want you to take from this. The _format_ is the easy part — it's just markdown; you could sketch one on a napkin. The hard part, the part that took us a year, is reading a brand well enough to name its signature moves, then turning them into email that renders right in every inbox on earth. The document is just where that work lands. Doing the work is the whole job — and it's what Brew is. ## Why `email-design.md`, not `design.md` You might be wondering why it's `email-design.md` and not just `design.md`. Fair question — and the answer is most of the reason this works at all. A brand's design system and a brand's _email_ design system overlap, but they are not the same thing, and the gap between them is exactly where generic AI emails go to die. Email is a hostile medium. It's arguably the worst rendering target in modern software. Your email doesn't run in a browser; it runs in a few dozen of them at once — [Gmail](https://support.google.com/a/answer/81126), Apple Mail, Outlook (still rendering with Word's engine in places), [Yahoo](https://senders.yahooinc.com/best-practices/), Superhuman, the native mobile clients — each in light mode _and_ dark mode, each with its own bugs. No flexbox. No grid. Web fonts that may never load. Layouts built out of nested tables like it's 2005. Styles inlined because `