Cost of Poor Quality (COPQ) Tracker
Capture scrap, rework, returns, warranty, and containment costs, split them into internal vs external failure (plus appraisal and prevention), and roll them up so quality can show leadership exactly what defects cost and where to attack first.
A private, team-only web app where you log quality-cost events, categorize them with the standard COPQ buckets, link them to NCRs or complaints, get an AI-built rollup by product/area/period that a manager approves before it's locked, and email a monthly COPQ report with a clean CSV export.
Before you start
- A free Supabase account
- A free Resend account
- A free Vercel account
- A list of your standard cost rates (scrap, labor/hour, sorting, freight) or a CSV/Sheet of past quality-cost events
The problem this kills
Everyone in your plant knows quality problems are expensive. Nobody can say how expensive. The scrap cost lives in one spreadsheet, rework hours live in another, returns and warranty live in the ERP, and the sorting overtime from last month's containment never got captured anywhere. So when you ask leadership to fund an improvement project, you're arguing with a hunch instead of a number.
Meanwhile finance only sees the costs that happen to land in a GL account, and they're scattered across a dozen lines that don't say "quality" anywhere. The true Cost of Poor Quality - the scrap, the rework, the returns, the warranty claims, the people sorting good parts from bad - stays invisible. You can't attack what you can't see, and you can't get the budget to attack it without a dollar figure.
This tool gives you that dollar figure. It captures every quality-cost event, sorts it into the categories finance and leadership recognize, and rolls it up by product, area, and month - so your next improvement pitch leads with "this defect family cost us $48,000 last quarter" instead of "I think this is a problem."
What you'll build
A small, private web app your quality team logs into. People record quality-cost events - the type (scrap, rework, return, warranty, sorting/containment, inspection), the failure category, the product or area, the quantity, and the cost. Costs can be entered directly or calculated from your standard rates (quantity x scrap rate, hours x labor rate, and so on). Each event can link back to the NCR or customer complaint that caused it.
An AI agent rolls everything up using the four standard quality-cost categories - internal failure, external failure, appraisal, and prevention - and breaks the totals down by product, area, and period. If you give it revenue, it trends COPQ as a percent of sales, the number leadership actually benchmarks against. Before any period total becomes the "official" COPQ report, the quality manager reviews and approves it. Then the tool emails a clean monthly report and exports two CSVs: the full cost detail and the category rollup.
What's inside the Implementation Plan
The plan is a single markdown file you paste into Claude Code (a free AI coding tool). It walks the AI through building the whole tool, step by step, with a ready-to-copy prompt at the end of each step.
The first thing it does is interview you about your business - your products and areas, how you name them, the cost rates you use, your typical and peak event volumes, who approves COPQ numbers, and your messy edge cases (warranty that arrives months late, supplier-caused defects, partial recoveries on scrap). It reads back a short tailored spec and waits for your thumbs-up before building a thing. You get a tool shaped around how your plant actually works - not a generic template you have to bend yourself into.
From there it builds the database, the login, the event-capture form, the standard-rate calculator, the AI rollup, the approval gate, the monthly email, and the CSV exports.
The governance it includes (this is the point)
This is a number you'll hand to finance and leadership, so the plan bakes in controls from the start:
- Login so only your team can open the tool.
- Row-level security so each organization only ever sees its own cost data.
- A complete audit trail - who logged each event, who approved each period, and exactly when.
- A human approval gate - the AI drafts the rollup, the quality manager reviews and approves the period total, and only then is it locked into the official COPQ report. The AI never publishes a number on its own.
- Duplicate guards - every cost event carries a unique cost-event-id, so the same scrap ticket or warranty claim can't be counted twice.
Who it's for
Quality managers and continuous-improvement (CI) leads who need to justify improvement projects with real cost data - and who are tired of guessing. If you can fill in a spreadsheet, you can build and run this. No coding experience required.
You've got this - paste the first prompt and let the agent interview you.