Shopify Win-Back Email Test
Lapsed customers are some of the cheapest revenue you can recover — if you can get them to open. This guide shows how to run a Shopify win-back email test with marginal, the hosted email MCP server, so your AI agent can generate variants, launch the send, and measure lift without leaving your editor.
Why win-back deserves its own A/B test
A Shopify win-back flow targets shoppers who bought once or twice and then went quiet. The audience is skeptical, so the subject line carries most of the weight — it decides whether your 'we miss you' message gets seen or buried. A subject line A/B test removes the guesswork about which angle actually re-engages them.
Common win-back angles worth testing against each other:
- Incentive-led: a percentage or dollar discount stated up front
- Curiosity-led: hinting at what's new since their last order
- Loyalty-led: framing the customer as missed or valued
- Urgency-led: a limited reactivation window or expiring perk
How marginal runs the test
marginal is a hosted email MCP server that your AI agent calls directly. You describe the win-back segment and the offer in plain language, and the agent uses marginal's tools to spin up variants, send them, and report back. There's nothing to self-host — you point your client at https://marginal.sh/mcp and authenticate with a Bearer API key.
The four MCP tools map cleanly to a single email A/B test:
- generate_variants — draft several subject lines for the same win-back message
- launch_test — split your lapsed-customer list and send the managed test
- get_results — pull open and click tracking plus within-test lift metrics
- recommend_next — get a data-backed call on the winner and the next experiment
A practical Shopify workflow
Export or sync your dormant Shopify segment — say customers with no order in 90 days — and hand it to your agent as the test audience. From Cursor, Claude Code, Windsurf, or any supported client, prompt the agent to draft four win-back subject lines and launch the split.
Once sends complete, ask for results. marginal reports opens, clicks, and lift per variant so you can see which angle moved the needle for your Shopify shoppers rather than guessing from a single broadcast.
- Define the lapsed segment and the win-back offer once
- Let the AI agent generate and launch the subject line A/B test
- Read get_results to compare variant performance head-to-head
- Use recommend_next to roll the winner into your reactivation flow
AI agent email marketing, on the free tier
Because marginal exposes everything through MCP, AI agent email marketing becomes a natural part of your build loop — no separate dashboard, no copy-pasting between tools. The free tier includes 100 experiments per month, which is plenty to validate your Shopify win-back subject lines before scaling.
Connect a supported client like Claude Desktop, Cline, Continue, or Zed, point it at the endpoint, and start testing. See the docs at https://marginal.sh/docs/ for tool details and setup.
- Endpoint: https://marginal.sh/mcp
- Auth: Bearer API key (marg_live_...)
- Free tier: 100 experiments/month
- Works with Cursor, Claude Code, Windsurf, Cline, Continue, and Zed
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