runbookify
← All plans
Procurement & Purchasing / Contracts & Pricing Agreements

Contracted-Price Compliance Checker: Stop Quietly Overpaying Suppliers

Compare every PO and invoice line against your negotiated price lists, flag overcharges, expired prices, and unit-of-measure mismatches, then let a buyer approve each claim before any email goes to the supplier.

IntermediateA weekendBuilds onNext.jsSupabaseResend
What you'll build

A web tool where you import your contracted price lists and PO/invoice lines, AI matches each line to the right contract price by item and date validity, flags overcharges, expired-price usage, and UoM mismatches with the overcharge dollars computed, a buyer reviews each flag and chooses claim / dispute / accept, approved claim emails go to suppliers, and a recovery tracker plus CSV export show what you've clawed back.

Gated download

Enter your email — the plan downloads instantly and a copy lands in your inbox.

By submitting your email you'll also receive the weekly runbookify newsletter. You can unsubscribe at any time.

Before you start

  • A Supabase account (free)
  • A Vercel account (free)
  • A Resend account (free)
  • Your contracted price lists (supplier, item/SKU, price, UoM, valid dates)
  • A PO / invoice line export (item, price, UoM, date)
  • Claude Code or any AI coding agent

The problem this kills

You negotiated hard. You have signed price lists, agreed discounts, and effective dates for every item from every key supplier. And then the invoices come in — and nobody actually checks them line by line against the contract.

So the leaks start. A supplier bills last quarter's higher price on an item whose contract price dropped. An expired promo rate keeps getting charged months after it lapsed. A price agreed per case shows up billed per each, and the unit-of-measure mismatch hides a 12x overcharge in plain sight. Each one is small enough that AP pays it without blinking, and category managers are too busy negotiating the next deal to audit the last one.

Multiply a handful of cents and a few dollars across thousands of lines a month and you're quietly handing back the savings you fought for. Worse — when you do finally catch one, you have no clean record of how long it's been happening or how much you're owed, so the credit claim is a guess.

What you'll build

A private web tool for your procurement and AP team that does the line-by-line checking nobody has time to do:

  • Import your contracted price lists (supplier, item/SKU, price, unit of measure, valid-from/valid-to dates) and your PO or invoice line data.
  • Match every transaction line to the correct contract price for that supplier + item, honoring the date the line was billed against the contract's effective dates.
  • Flag three kinds of leak: price variances (billed above contract beyond your rounding tolerance), expired-price usage (no valid contract price on that date), and unit-of-measure mismatches (billed per each vs. contracted per case).
  • Compute the overcharge dollars per line, with currency and UoM conversions handled.
  • Hand each flag to a buyer, who reviews it and chooses an action — claim a credit, dispute, or accept — before anything leaves the building.
  • Send approved claim emails to the supplier and track recovery by supplier and claim, with a CSV export for your finance system.

What's inside the Implementation Plan

A step-by-step runbook you paste straight into an AI coding agent (like Claude Code). You don't write code — you answer questions and paste prompts.

It starts by interviewing you about your business. Before a single line of code, the plan has the agent ask how your price lists and PO/invoice data are actually shaped — your SKU and supplier-code conventions, whether prices are effective-dated, which currencies and units of measure you use, your rounding tolerance, your approval rules, and the messy exceptions (volume breaks, freight, surcharges). It reflects a short tailored spec back to you for a thumbs-up, then builds the data model, the matching logic, and every screen around your answers — not a generic template.

Inside you'll find: the full discovery interview, the exact stack, an architecture diagram, and copy-paste prompts for each build step — import, matching engine, variance/expired/UoM flagging, overcharge math, the buyer review queue, the approval-gated claim emails, the recovery tracker, and the verification checklist. Plus a no-API fallback so you can build the whole thing today from Sheets/CSV with no integration to your ERP.

The governance it includes (this is the point)

This is not a script that fires off accusatory emails to your suppliers. Governance is built into every layer:

  • Login so only your team can open the tool.
  • Row-level security so each organization only ever sees its own contracts and transactions.
  • A complete audit trail — who reviewed which flag, what they decided, and when.
  • A hard human-in-the-loop approval gate — the AI flags the overcharge and drafts the claim, but a buyer reviews and approves before any email reaches the supplier. The tool flags; the human acts.
  • Duplicate guards so the same invoice line can't generate two variance records and the same claim can't be double-counted in recovery.

Who it's for

Category managers, procurement leads, and accounts-payable teams who negotiate prices but have never had a practical way to confirm suppliers actually charge them what was agreed. If you've ever found one overcharge by accident and wondered how many you've missed, this is for you.

You've got this — paste the first prompt and let the agent interview you.

Gated download

Enter your email — the plan downloads instantly and a copy lands in your inbox.

By submitting your email you'll also receive the weekly runbookify newsletter. You can unsubscribe at any time.