Set up a weekly report that ties PRs to KPI shifts

A product leader used Claude Code to generate recurring writeups that compare last week’s shipped work with changes in key metrics.

Done with Claude Code· Intermediate

What they did
David Kelly, a product leader, keeps metric tracking in Metabase and wanted a way to connect it with shipping activity. He asked Claude Code to create a workflow that gathers weekly GitHub pull requests, matches them against period-to-period KPI movement, and estimates how strongly each change may have influenced each metric. To reduce blind spots, he wired in Zen MCP tools plus the OpenAI and Gemini APIs so multiple models could review the same analysis.
How it went
He now gets a weekly readout on whether recently shipped features appear to have lifted or weakened KPIs, and he uses that output to keep improving the process. The post does not give any accuracy figure for the match between code changes and metric movement.
Tips
A paid Claude plan is required; he recommends Pro at $20 a month. He also cautions that the API route can become costly fast.

Try it yourself with Claude Code

Each [Monday], review the GitHub pull requests merged in [repository name] over the previous week and pull the latest KPI values from [KPI source]. Compare the shipped changes with week-over-week movement in [KPI names], then draft a one-page summary that says what went out, which items likely pushed each KPI up or down, and what deserves attention next week. Show me the draft first, and do not start any recurring schedule, send anything, or save anything final until I approve.

Fill in the [brackets]. Agents can make mistakes, so review anything before it is sent, paid or deleted.

Read the original on dmkthinks.org ↗

Get 3 agent uses like this every morning

With the day's top AI news. Free, and one click to unsubscribe.