The problem
Dashboards show everything, so nobody knows what matters.
Drop-offs are found but not linked to the journey.
Numbers get quoted without sources or definitions.
09
Know where users drop off, and where not to look.
Performance Analyst reads your analytics, maps it to the journey, and writes a measurement plan with a short, evidenced focus list.
Or get all nine agents in the UCD Team bundle.
Dashboards show everything, so nobody knows what matters.
Drop-offs are found but not linked to the journey.
Numbers get quoted without sources or definitions.
Maps every journey stage to a metric and a source
Builds funnels and finds the biggest drop-offs, by device and channel
Reports completion, digital take-up, satisfaction and cost per transaction
Gives a top-three focus list and what to leave alone
Shows the source and date range for every number
Get this on Gumroad and download the zip.
Unzip it into a project folder and open it in your AI tool.
Say `build a measurement plan for <service>`.
It works one stage at a time and stops for your yes.
Journey to metrics
Funnel and drop-off, KPIs
Focus prioritisation, Measurement plan
GOV.UK Service Standard point 10 and the four classic GOV.UK KPIs.
Every claim cites its source. Nothing is written to your tools without your approval.
Service owners, product managers and delivery teams without a performance analyst.
An AI tool that reads project instructions: Claude Code, Cursor, or a tool that reads AGENTS.md. Claude Projects and custom bots work too, with copy-and-paste outputs.
Your own AI subscription or API key. Nothing to host.
Optional connections: Google Analytics 4 MCP, Mixpanel MCP, Amplitude MCP, Python.
| File | What it does |
|---|---|
00-root-instructions.md | Who the agent is, the rules, the gates |
01-map.md | Stage 1: Map |
02-analyse.md | Stage 2: Analyse |
03-plan.md | Stage 3: Plan |
04-tools.md | Connections and permissions |
05-manifest.md | Which skill uses which tool |
06-adapters.md | How to load it in your tool |
LICENCE.md | What you may and may not do |
README.md | Start here: set-up in two minutes |
guardrails.md | Risks, controls and sign-off |
memory.md | What is kept and where |
adapters/AGENTS.md | Adapter for your AI tool |
adapters/CLAUDE.md | Adapter for your AI tool |
adapters/chat-system-prompt.md | Adapter for your AI tool |
adapters/performance-analyst.mdc | Adapter for your AI tool |
evals/eval-suite.md | Tests and pass marks |
samples/01-funnel.csv | Synthetic sample input |
samples/02-channels.csv | Synthetic sample input |
samples/03-brief.md | Synthetic sample input |
Designed before it was built
This agent was designed with the five-question method: purpose, boundaries, what it touches, when a person steps in, and how you know it works. [Read the design](./anatomy).
Designed by Vijay Kandhalu.