Complete AI Training
Sign inGet my AI kit

Your job's AI kit

Get your AI kit

Tell us who you are and what you do. We show you your kit right away and email you the link: skills, prompts, AI agents, MCP servers and courses for your job.

500+ jobs ready, and we make a kit for any other job. No payment needed to look.

Share

AI agent for product owners

Product Analytics Instrumentation Gap Agent

Make sure every shipped feature has working usage tracking that answers the product questions

MonitoringExecutives and strategyProduct development
See how it worksGet access
Product Analytics Instrumentation Gap Agent: what goes in, what the agent does and what you get

What it does

A feature ships and a month later nobody can say how many people used it because no event was added. This agent lists each shipped feature with the questions the team wants answered, for example how many people finish the new setup. It checks event data for the expected events and their properties, and flags missing ones, odd volumes and events that stopped firing. For each gap it drafts a ticket for engineering with the event name, where it should fire and why it matters. After the next release it checks the data again and confirms the event now appears at plausible volume. If it does not, it reopens the ticket. The product leader approves the priority. Edge case: an event firing twice per action is flagged as duplicate.

How it works

Follow the arrows from top to bottom. The orange dashed arrow is the loop: when a check fails, the agent goes back and tries again.

Start and resultWhat it doesA check on its own workWaits for your OKGoes back and retries
Yes, continueApprovedYes, continueNoNo 1 STARTS WHEN Release ships or weekly check runs 2 USES A TOOL List shipped features and their key questions 3 USES A TOOL Query event data for expected events and properties 4 CHECKS THE RESULT Does every expected event appear at plausiblevolume? If not: Classify the gap as missing, broken orduplicate. Back to step 2. 5 DOES Draft an engineering ticket for each gap 6 YOU APPROVE Product leader approves ticket priority 7 USES A TOOL Open approved tickets 8 USES A TOOL Recheck event data after the next release 9 CHECKS THE RESULT Is the event now firing correctly? If not: Reopen the ticket with the new evidence. Back tostep 7. 10 RESULT Instrumentation status report
Read the steps as a list
  1. Release ships or weekly check runs
  2. List shipped features and their key questions
  3. Query event data for expected events and properties
  4. Does every expected event appear at plausible volume?If not: Classify the gap as missing, broken or duplicate. Back to step 2.
  5. Draft an engineering ticket for each gap
  6. Product leader approves ticket priorityThe agent waits here for your OK.
  7. Open approved tickets
  8. Recheck event data after the next release
  9. Is the event now firing correctly?If not: Reopen the ticket with the new evidence. Back to step 7.
  10. Instrumentation status report

How it decides

It marks an event missing when no events arrive 7 days after release, and suspect when volume differs from expected by over 50 percent.

  • Mark an event missing after 7 days with no data
  • Flag volume differing from expectation by more than 50 percent
  • Flag an event that fires twice per action as duplicate
  • Rank tickets by the product question each gap blocks

Make it yours

Every agent is a starting point. You choose these settings for your own situation.

  • Missing event wait time (default 7 days)
  • Volume tolerance (default 50 percent)
  • Features tracked (default last 90 days of releases)
  • Ticket destination (default the engineering queue)

What keeps you in control

It always asks you first

  • Product leader approves ticket priorities

Hard limits

  • Never change tracking code or production data
  • Never open tickets without approval
  • Use aggregate counts, not user-level data

It stops when

  • Done: All key events firing and verified
  • Stop: Event data is unavailable, so hand to the analytics owner

Set it up

We guide you through the set-up, step by step

Members get the full set-up guide for this agent. No technical skills needed: you copy, paste and upload.

10 minto set it up in your AI
5 AIsChatGPT, Claude, Copilot, Gemini, Grok
  • One set of instructions to paste into your AI, with the clicks for ChatGPT, Claude, Microsoft 365 Copilot, Gemini and Grok
  • The agent then walks you through connecting your own data, one source at a time
  • A downloadable copy with the flow chart, the rules and the full guide
Get access to this agent

An example run

What happensAfter a September release the agent finds the setup_complete event has 0 rows and checkout_started has twice the expected volume. It drafts 2 tickets. The leader prioritizes setup_complete. After the next release the event shows 1,840 rows and the check passes. The duplicate is fixed in the same sprint.

More agents for product owners

Automation · Product development

Product Feedback Reproduction Agent

Every piece of product feedback ends as a reproducible issue or a precise request for missing context.

8 steps1 check4 jobs
Communication · Automation

Product Requirement Change Impact Agent

Every requirement change reaches all linked tests, docs and teams.

7 steps1 check3 jobs
Quality control · Creating

Product Specification Example Agent

Specifications whose rules are tested by concrete examples.

7 steps1 check3 jobs
Quality control · IT and development

Ad Hoc Data Request Scoping Agent

A clear request, a first result that ties to known totals, and an analyst-approved answer

10 steps1 check7 jobs