I spent a week looking at the distribution of scores across 400+ emails in the Kopi public gallery at trykopi.ai/emails, trying to understand what the automated critique was actually penalizing.
The gap between 70 and 93 is almost never the copy. It's three things: CTA contrast below the threshold where mobile Gmail renders it as a gray button instead of a colored one, layout hierarchy that frontloads secondary content above the fold, and line-height on body text that renders fine on desktop but collapses to unreadable on a 375px viewport.
The problem with every email tool I'd used before is that none of them tell you any of this before you send. You get a blank canvas, a template, and a "preview on mobile" button. The preview shows you what it looks like — it doesn't tell you whether it's broken.
Kopi runs every generated email through a scorer that checks 12+ design criteria and returns a 0–100 score with specific callouts: which element failed, why, and what it would take to fix it. Not a generic "your CTA could be stronger" — a specific flag that your button's foreground-to-background contrast is 3.2:1 when it needs to be above 4.5:1 to survive dark mode rendering on iOS Mail.
You start with a text prompt. Under 5 minutes, you have a complete HTML email. The score comes with it. If it's a 74, you can see exactly where the 26 points went.
It also integrates with Klaviyo for direct export and runs as an MCP server inside Claude Code, Cursor, and ChatGPT if you'd rather work from a terminal or editor.
For anyone building email campaigns without a dedicated designer — or with one who's stretched thin — the score isn't a nice-to-have. It's the only way to know if you've shipped something broken before your open rate tells you.
Top comments (0)