Running a SaaS Win-Back Email Test
Churned and dormant users are the cheapest pipeline a SaaS team has — they already know your product. A SaaS win-back email test helps you find the message that actually pulls them back, and marginal lets your AI agent run that test end to end through a single MCP endpoint.
Why win-back deserves its own A/B test
Win-back audiences behave differently from active users. They've already stopped opening, so a generic re-engagement blast usually lands in the void. The variable that moves the needle most is the subject line — it decides whether the email gets opened at all before any offer or copy matters.
A focused subject line A/B test on your lapsed segment tells you which angle re-earns attention: a feature you've shipped since they left, a discount, a 'we miss you' nudge, or a pointed question about why they churned.
- Test re-engagement hooks against your dormant SaaS segment
- Measure opens and clicks, not just sends
- Compare within-test lift so you keep the winning angle
- Keep tests small and frequent instead of one big risky send
How marginal runs the test
marginal is a hosted email MCP server. Your AI agent connects to https://marginal.sh/mcp with a Bearer API key and calls four tools to drive the full email A/B test loop — no separate dashboard juggling required.
For a win-back campaign, the flow maps cleanly to each tool below.
- generate_variants — draft several win-back subject lines for the same email body
- launch_test — send the subject line A/B test to your lapsed-user segment with tracking on
- get_results — pull open/click rates and within-test lift between variants
- recommend_next — get a data-backed call on which subject to roll out
Letting an AI agent own re-engagement
Because marginal exposes these tools over MCP, AI agent email marketing becomes practical rather than theoretical. Inside Cursor, Claude Code, Claude Desktop, Windsurf, Cline, Continue, Zed, or OpenAI Codex, your agent can generate win-back variants, launch the test, read results, and propose the next iteration in one conversation.
That tight loop is ideal for win-back, where you want to iterate on tone and offer over several rounds rather than guessing once. Each cycle feeds the next, so your re-engagement subject lines sharpen against real opens from your own SaaS list.
- Works with the MCP clients your team already uses
- Free tier covers 100 experiments per month to get started
- Hosted — nothing to deploy or maintain
A practical win-back test plan
Keep the first test deliberately simple so the lift signal is clean. Hold the email body and CTA constant and vary only the subject line; that isolates attention as the variable and tells you what re-opens the relationship.
- Define the dormant segment: e.g. no login in 60+ days
- Generate 3–4 subject lines across distinct angles (value, discount, curiosity, miss-you)
- Launch the email A/B test and let tracking collect open/click data
- Promote the winner with recommend_next, then test the offer in the next round
- Read the docs at https://marginal.sh/docs/ to wire up your client
Get started with marginal
marginal is a hosted email marketing MCP server at marginal.sh. Sign up free, create an API key, and connect https://marginal.sh/mcp from Cursor, Codex, or Claude Desktop.
- 100 experiments/month on the free tier
- Four MCP tools: generate_variants, launch_test, get_results, recommend_next
- Listed in the MCP Registry as sh.marginal/mcp