Cart Abandonment Subject Lines in Zed
Connect Zed to marginal's email MCP server and let an AI agent draft, test, and refine your cart abandonment subject lines without leaving your editor.
Why test cart abandonment subject lines from Zed
Cart abandonment emails live or die on the subject line — it decides whether a shopper reopens their cart or forgets the purchase entirely. Instead of guessing, you can run a structured subject line test directly from Zed, where your agent already has context about your store and copy.
marginal exposes its tools over MCP, so the same AI agent you use to write code can also generate variants, launch a send, and pull back open and click data for your recovery flow.
- Iterate on subject lines next to your codebase and templates
- Keep cart abandonment copy versioned in plain prompts
- Let the agent compare urgency, discount, and curiosity angles
- Use within-test lift metrics to pick a winner with confidence
Connecting marginal to Zed
marginal is a hosted email MCP server — there is nothing to self-host. Add the endpoint at https://marginal.sh/mcp to Zed's MCP configuration and authenticate with your Bearer API key (it starts with marg_live_).
Once connected, Zed's assistant can call the four marginal tools: generate_variants, launch_test, get_results, and recommend_next. See the docs at https://marginal.sh/docs/ for the latest setup details for Zed.
- Registry: sh.marginal/mcp
- Endpoint: https://marginal.sh/mcp
- Auth: Bearer API key (marg_live_...)
- Free tier covers 100 experiments per month
Running a subject line A/B test for abandoned carts
A typical cart abandonment workflow in Zed starts with a prompt describing your audience and offer. Ask the agent to generate several subject line variants, then launch the A/B test against your abandoned-cart segment.
marginal handles the managed send and tracks opens and clicks, so you get real engagement signals rather than estimates. As results come in, the agent reads them back and recommends the next move.
- generate_variants — produce 'Forgot something?' vs. discount-led lines
- launch_test — send the subject line A/B test to your cart segment
- get_results — review open and click rates plus within-test lift
- recommend_next — get a data-backed suggestion for the next round
Tips for stronger cart recovery subject lines
When you let an AI agent handle email marketing, give it the constraints that matter: product category, average order value, and how aggressive you want to be with discounts. Sharper context produces sharper variants.
Treat each test as one input to a longer cadence. Use recommend_next to carry winning angles into your second and third reminder emails, and keep experiments small enough to stay within the free tier while you learn.
- Mix urgency, social proof, and curiosity across variants
- Avoid leading with a discount in every test — measure incrementality
- Re-run tests by segment if behavior differs by cart value
- Feed winning subject lines back into your reminder sequence
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