Running a Beauty Cross-Sell Email Test
Cross-selling complementary beauty products — a serum for a new cleanser, a brush set for a foundation — lives or dies on the email that pitches it. This guide shows how to design a beauty cross-sell email test with marginal, the hosted email MCP server that lets your AI agent launch and measure experiments directly.
Why cross-sell needs its own A/B test
Beauty shoppers buy in routines, not single SKUs. A customer who just bought a vitamin C serum is a strong candidate for a moisturizer or SPF — but the framing of that recommendation changes everything. "Complete your routine" reads differently from "Customers also loved" or "Pairs perfectly with your serum."
A cross-sell email A/B test isolates which angle actually drives clicks and revenue, instead of guessing. Because beauty purchase cycles are short and replenishment-driven, you can read meaningful results quickly and roll the winner into your post-purchase flow.
- Test routine-completion vs. social-proof framing
- Compare bundling language against single add-on offers
- Match cross-sell suggestions to the original product category
- Measure click-through, not just opens, to gauge purchase intent
Designing the test with marginal
marginal exposes four MCP tools your agent calls in sequence. Start with generate_variants to produce subject lines around your cross-sell hook, then launch_test to ship them to a segment of recent buyers. As recipients engage, get_results returns open/click tracking and within-test lift metrics so you can see which variant pulls ahead.
For a beauty cross-sell email test, a clean structure is two or three subject line variants against the same body, so the subject line A/B test result is unambiguous. Once a leader emerges, recommend_next suggests the follow-up variant to try.
- generate_variants — draft cross-sell subject lines
- launch_test — send the email A/B test to your buyer segment
- get_results — pull opens, clicks, and lift
- recommend_next — get the next variant to iterate on
Cross-sell subject line angles to try
Subject lines carry most of the open-rate signal in beauty cross-sell. Give your agent a few distinct directions so the variants test genuinely different ideas rather than minor word swaps.
- Routine framing: "Your serum has a perfect match"
- Social proof: "The moisturizer beauty editors pair with this"
- Benefit-led: "Lock in that glow — finish your skincare set"
- Urgency-light: "Add the step your routine is missing"
- Personalized: "Because you loved the cleanser…"
Plugging it into your AI agent workflow
marginal is a hosted MCP server, so there is nothing to self-host. Connect it once at https://marginal.sh/mcp with a Bearer API key and your AI agent — in Cursor, Claude Code, Windsurf, Cline, Zed, or any MCP client — can run AI agent email marketing experiments end to end.
The free tier covers 100 experiments per month, enough to test every cross-sell angle in a beauty catalog. See the docs at https://marginal.sh/docs/ for tool schemas and example calls when you wire up your first beauty cross-sell email test.
- Endpoint: https://marginal.sh/mcp
- Auth: Bearer API key (marg_live_...)
- Works with Claude Desktop, OpenAI Codex, Continue, and more
- 100 experiments/month on the free tier
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