Testing Review Request Subject Lines in Cursor
Review request emails live or die on the subject line. With marginal connected to Cursor, you can draft, test, and refine review request subject lines without leaving your editor — driving the opens that turn into ratings and testimonials.
Why subject lines decide your review reply rate
A review request is one of the highest-intent emails you send: the customer already bought, and you're asking for thirty seconds of their time. But that ask never registers if the subject line gets skipped in a crowded inbox. Small wording shifts — naming the product, adding a time estimate, or leading with gratitude — change open rates more than the body copy ever will.
Rather than guessing which framing works for your audience, you can run a structured subject line test and let the open and click data settle it. That's exactly what marginal's email MCP server is built for, and Cursor gives you a natural place to iterate.
- Test gratitude-led vs. ask-led openers for review requests
- Compare named-product subjects against generic ones
- Measure whether a time cue ("30 seconds?") lifts opens
Connecting marginal to Cursor
marginal is a hosted MCP server, so there's nothing to deploy or self-host. Add the endpoint to Cursor's MCP configuration with your Bearer API key, and the four tools — generate_variants, launch_test, get_results, and recommend_next — become available to the agent.
Point Cursor at https://marginal.sh/mcp and authenticate with your marg_live_ key. Once it's wired up, you can describe a review request campaign in plain language and let the agent handle variant generation and launch.
- Endpoint: https://marginal.sh/mcp
- Auth: Bearer API key (marg_live_...)
- Registry: sh.marginal/mcp
- Setup details: https://marginal.sh/docs/
Running a subject line A/B test for review requests
A typical workflow inside Cursor looks like a short conversation. Ask the agent to generate a handful of review request subject lines, launch them as a split test, and report back once opens accumulate. The AI agent email marketing loop stays inside your editor while marginal manages the sends and tracking.
Because marginal reports within-test lift, you don't have to eyeball raw open counts. The agent can call recommend_next and tell you which variant is pulling ahead and why, so your follow-up send uses the winner.
- generate_variants — produce several review request subject lines to compare
- launch_test — send the subject line A/B test to your audience split
- get_results — pull open/click tracking and within-test lift metrics
- recommend_next — get the leading variant for the next send
Tips for review request subject line tests
Review requests tend to perform best when the subject feels personal and low-effort. Build your variants around a single hypothesis at a time so the lift is attributable, and keep the audience split balanced for cleaner reads.
- Vary one element per variant — tone, length, or a specific cue
- Send review requests soon after delivery while the experience is fresh
- Use the free tier's 100 experiments/month to test across product lines
- Re-run a subject line test seasonally as inbox behavior shifts
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