Fashion Re-Engagement Email Tests with marginal
Lapsed fashion subscribers are some of your highest-value contacts — they bought before, then went quiet. A fashion re-engagement email test helps you find the subject line and offer that pulls them back, and marginal lets your AI agent run that experiment end to end.
Why re-engagement testing matters in fashion
Fashion lists churn fast. Seasonal drops, sale fatigue, and crowded inboxes mean shoppers stop opening within weeks of their last purchase. A win-back flow is only as good as the subject line that gets it opened, so testing variants is where the real gains live.
Instead of guessing whether 'We miss you' beats 'Your size is back in stock,' a structured email A/B test gives you measurable lift across opens and clicks for your dormant segment.
- Test offer framing: discount vs. early access to a new drop
- Compare urgency ('last chance') against curiosity ('picked for you')
- Segment by past category — outerwear buyers respond differently than accessories shoppers
- Use within-test lift metrics to confirm a winner before a full send
How marginal runs the test
marginal is a hosted email MCP server that exposes four tools to your AI agent: generate_variants, launch_test, get_results, and recommend_next. You describe the re-engagement campaign in plain language, and the agent drafts subject line variants, launches a managed send, and reports back on what worked.
Because sends and open/click tracking are handled by marginal, there's no separate infrastructure to wire up for a subject line A/B test on your lapsed fashion segment.
- generate_variants — produce several re-engagement subject lines to compare
- launch_test — split your dormant segment and send the managed test
- get_results — pull open and click rates with within-test lift
- recommend_next — get a data-backed suggestion for the next iteration
A practical re-engagement workflow
AI agent email marketing shines when you treat re-engagement as a loop, not a one-off. Start narrow, learn from the lift, then expand the winning angle to the rest of your inactive list.
- Define your dormant segment (e.g. no opens in 90 days)
- Ask your agent to generate_variants for a comeback subject line A/B test
- launch_test to a holdout slice of the segment
- Review get_results, then use recommend_next to plan the follow-up
- Roll the winning variant out to the full lapsed audience
Getting connected
marginal connects to the MCP clients fashion teams already use — Cursor, Claude Desktop, Claude Code, Windsurf, Cline, Continue, Zed, and OpenAI Codex. Point your client at the endpoint at https://marginal.sh/mcp and authenticate with your Bearer API key.
The free tier covers 100 experiments per month, which is plenty to validate a re-engagement subject line strategy before scaling. Full setup details are in the docs at https://marginal.sh/docs/.
- Endpoint: https://marginal.sh/mcp
- Auth: Bearer API key (marg_live_...)
- Hosted — no self-hosted server to maintain
- Free tier: 100 experiments/month
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