Testing Welcome Series Subject Lines in Continue
Continue lets you drive email experiments from your editor's AI assistant. Wire it to marginal's email MCP server and you can generate, launch, and measure welcome series subject lines without leaving your workflow.
Why test welcome series subject lines first
The welcome series is the highest-leverage email a subscriber receives. It's their first impression, opens are typically strong, and small changes to the first subject line compound across every new signup. That makes it the ideal place to start a subject line test.
Instead of guessing, you let an AI agent propose multiple framings — curiosity, benefit-led, personal — and split them across your incoming audience to find what actually earns opens.
- High intent: new subscribers are already engaged
- Steady traffic: signups feed your test continuously
- Clear signal: open and click lift shows up fast
- Reusable winner: the best line keeps working as evergreen
Connecting Continue to marginal
marginal is a hosted email MCP server, so there's nothing to deploy. Point Continue at the endpoint at https://marginal.sh/mcp and authenticate with your Bearer API key (the marg_live_ token from your dashboard). Once registered, Continue can call the marginal tools directly inside a chat session.
Add the server to your Continue MCP configuration and you'll have access to four tools: generate_variants, launch_test, get_results, and recommend_next.
- Registry: sh.marginal/mcp
- Endpoint: https://marginal.sh/mcp
- Auth: Bearer API key (marg_live_...)
- Free tier: 100 experiments/month to get started
Running a welcome series subject line A/B test
Ask Continue to draft variants for your first welcome email, then launch the test against your new subscriber segment. marginal handles the managed send and tracks opens and clicks, reporting within-test lift so you can compare variants on equal footing.
A typical flow in Continue reads like a short conversation that ends with a data-backed recommendation.
- generate_variants — produce several welcome subject lines from your prompt
- launch_test — start the subject line A/B test as a managed send
- get_results — pull open/click metrics and within-test lift
- recommend_next — let the agent suggest the winning line or a follow-up test
Tips for cleaner results
Because welcome emails trigger on signup, results accumulate over days, not seconds. Let the test run until each variant has enough sends to separate real lift from noise before promoting a winner.
Use Continue's AI agent email marketing loop to iterate: take the winning angle, then test a new dimension — length, emoji, personalization — in the next welcome email of the series.
- Change one variable per test so the signal stays interpretable
- Wait for adequate sample size before reading lift
- Promote the winner, then test the second email in the series
- Keep the docs handy at https://marginal.sh/docs/ for tool details
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