Free tool
Does your count match the system?
Upload your BioTrack export and whatever you counted on the floor. Every identifier gets matched by ID — not just compared row by row — and sorted into what agrees, what doesn't, and what's missing from either side.
This is an exact comparison, not a guess: the same record classified the same way every time, so you can hand the report to whoever needs to see it.
Neither file is uploaded
Both files are read and compared by JavaScript running on your own machine. Nothing is sent to us, nothing is stored, and closing the tab clears everything. Check it yourself: open developer tools, watch the Network tab, and load a file.
The AI assistant below is different: it's a separate, optional tool, and what you say or type into it does reach our server to reach Claude. It never touches either file above.
No package tags handy? Read off what you counted
Reconciliation below matches by package ID, so a spoken count needs one extra step: dictate what's on the shelf to the assistant, export it as Excel or CSV, then drop that file into the Physical count box below like any other file. It won't replace a real scan-based count for something you'd hand to an inspector — there are no package IDs in a spoken list — but it's a fast way to get a rough one on record.
Ask or do anything to help your business
AI assistant — free, no accountEverything ChatGPT or Claude does — writing, drafting, explaining, working through a problem. This one already knows the NM migration, and gets more useful the more it knows about your operation.
Unlike the analyzer, what you type or say here is sent to our server to reach Claude — telling it what's on your shelves, one item at a time, is exactly what the mic and the spreadsheet export are for. A full inventory export or state-system report is a different thing; keep that in the analyzer, which never leaves your browser. This can be wrong; check anything that matters against the CCD bulletin. Not legal advice.
Try one
System export
Your BioTrack Current Inventory export — what the system says you have.
Or drop it here. CSV, TSV or Excel.
Physical count
Whatever you counted on the floor — one identifier and one quantity per row.
Or drop it here. CSV, TSV or Excel.
How matching works
What it does
- Matches by identifier, not row position — column order or extra rows in either file don't matter.
- Ignores stray spaces inside an ID, the same way the readiness analyzer does, so a typo in one file doesn't hide a real match.
- Sums quantities when the same identifier appears more than once in a file — two rows for one batch are treated as one total.
- Every classification is deterministic — the same two files always produce the same report.
What it doesn't do
- It doesn't guess. If a quantity is blank on one side, that record is reported as counted with an unconfirmed amount — never as a shortage that isn't actually in the data.
- It isn't a compliance check. For that, use the readiness analyzer.
- It doesn't change either file, and it doesn't remember them after you close the tab.
Nothing on this page is saved to an account either — not the reconciliation report, not an exported inventory list; export what you need before you close the tab. What an account does hold is a dated history of your readiness analyzer score. Sign in — no password, just an email link.