Inventory Forecast Spreadsheet Audits with AI for Amazon Sellers

What they did

Before an AI lunch-and-learn for a tactical gear maker that also sells on Amazon, the presenter gave AI the company's forecasting spreadsheet (dozens of tabs, more than a thousand rows), the planner's walkthrough video and his own notes, and asked it to read everything.

How it went

He reports the AI found a row forecasting negative one knife for months, most vendors missing lead times (so the reorder trigger was buffer plus nothing), excess inventory sitting in parts rather than finished goods, and two years of tabs copied forward with drift.

We only report what the source says. Results are self-reported unless noted.

Worth knowing

His takeaway: build the model in a sheet first, then hand it to AI. 'AI can run the model. A person owns the number.' The author works at Seller Labs, which sells an MCP for reorder math.

Copy-paste prompt

I'm attaching my inventory forecasting spreadsheet. Read every tab before answering. Report: (1) any formulas producing impossible values (negative or blank forecasts), (2) vendors or SKUs with missing lead times, (3) where the reorder point = buffer + lead-time demand is broken, (4) tabs that look copied forward with inconsistent numbers, (5) where excess inventory is sitting (components vs finished goods). Cite the tab and cell for each finding. Don't change anything.

The source didn't publish its exact prompt. We wrote this starter so you can try the same workflow. Replace anything in [brackets].