Unit-of-Measure Conversion Manager

The problem this kills

Somebody received "5 cases" but the system thought they meant 5 eaches. Somebody ordered "10" of a thing that ships 12 to an inner and got buried in stock. A cycle count came out 144 short because the case factor was off by a single digit. These aren't dumb mistakes - they're the inevitable result of unit-of-measure (UoM) conversions living in scattered spreadsheets, item-master notes, and people's heads.

When the eaches-per-inner, inners-per-case, and cases-per-pallet factors aren't stored in one trusted place - and aren't checked for sanity - every receipt, every pick, and every count becomes a chance for a 12x (or 144x) error. And the worst factors are the ones nobody validates: a zero, a blank, a factor someone typed twice, a chain that doesn't multiply up to the pallet quantity printed on the label.

What you'll build

A small, private web app for your team that becomes the single source of truth for UoM:

• Upload your item UoM data (CSV or Google Sheet - no integration needed).
• It stores each SKU's full conversion chain - base unit, eaches per inner, inners per case, cases per pallet - plus weight and volume per unit.
• It validates every chain automatically: it catches zero or negative factors, impossible values, missing base units, broken chains that don't multiply up correctly, and duplicate SKU/UoM rows.
• It shows you exactly which rows are clean and which need a fix, in plain language.
• A data steward reviews proposed UoM changes and approves them - nothing updates the master until a human signs off.
• It exports the approved UoM master as a clean CSV your receiving, picking, and counting tools can rely on.
• It includes a conversion calculator screen so anyone can convert "5 cases of SKU-123" into eaches in one click.

What's inside the Implementation Plan

A complete, paste-and-go runbook for Claude Code. It's written for a non-coder: plain language, one step at a time, with a ready-to-copy prompt at the end of every step.

Most importantly, the plan opens by interviewing you about your business - the units you actually use, your SKU and code conventions, how your factors are named today, your base-unit rules, your typical and peak data volumes, who's allowed to approve a change, and the messy exceptions (catch weights, variable packs, items sold by weight). It reflects a short tailored spec back to you for a thumbs-up, then builds the tool around your units and your rules instead of a generic template.

Inside you'll find: the discovery interview, the data model, the conversion-chain validation logic, the steward approval gate, the CSV export, the conversion calculator, and a verification checklist so you know it actually works.

The governance it includes (this is the point)

This isn't a throwaway script. It's an internal tool with the controls inventory leadership expects:

• Login so only your team can open it.
• Row-level security so each organization only ever sees its own data.
• A full audit trail - who imported what, who approved which factor changes, and when.
• A hard human-in-the-loop approval gate - the tool drafts the UoM changes; a data steward reviews and approves before anything updates the master that receiving and picking depend on. The AI never commits a factor on its own.
• Duplicate guards so the same SKU/UoM combination can't be processed or stored twice.

Who it's for

Inventory control analysts, receiving leads, and purchasing teams who own item-master accuracy - and who are tired of chasing count discrepancies caused by a single wrong conversion factor.

You've got this. Open Claude Code and paste the first prompt.