Guide / Excel stage
An inventory system on Excel: build it right, know the end
Excel can run a small stock operation honestly — if built as movements, not totals. The right build, the habits that keep it true, and the signs that it has become the bottleneck.
The key facts
- The right build is movements-first: a Products sheet, an append-only Movements sheet, quantities derived by formula — never typed.
- Excel's honest ceiling: single-user truth, no guardrails, no scan-to-record, and every management action is a human running a formula.
- The migration signs: the count stops agreeing, two people need the file at once, or a quantity nobody can explain appears.
- The migration is easy when the build was right — products and movements export to any system; BSimple (paid, from $180/mo, trial) imports them as first-class citizens.
- 01The key facts
- 02The build that works
- 03The ceiling, and how it announces itself
- 04Migrating: the reward for building it right
The build that works
Sheet 1 — Products. One row per product: SKU, description, supplier, unit cost, par level. Read-only in spirit; edited deliberately, never in passing.
Sheet 2 — Movements. One row per event: date, SKU, direction (in/out), quantity, reason (received, sold, adjustment), who. Append-only — the discipline that matters more than any feature. The Sheets version is the same build with sharing built in.
Derived quantities. On-hand equals the sum of movements — by formula, never typed into a cell. The day someone "fixes" a quantity directly, the record has started lying; protection and the movements habit are what hold the line.
A monthly count. A real count against a printed list, variances investigated, corrections entered as adjustments with reasons. This is the discipline that keeps any record honest, spreadsheet or not — the count flow formalises it.
Built that way, Excel is a legitimate single-user inventory system, and no honest vendor — including us — will tell a one-person operation it needs software.
The ceiling, and how it announces itself
Excel's limits arrive as symptoms, not abstractions. The two-person problem: the file is on one machine or in one inbox, and the second person's truth is a phone call away. The unexplained quantity: a variance the movements sheet cannot explain, usually a cell edited directly weeks ago. The formula forest: lookups and pivots accreting until only the builder understands the workbook — and the builder is on leave. The scan that cannot be: a scanner types into a cell, but there is no system at the shelf to receive it — the scanning gap is where floor work and the record part ways.
Any one of these is manageable; they arrive together, and they are the system telling you it has become the bottleneck it was meant to fix. The VBA stage and the GitHub escape hatch cover the build-it-yourself responses; the buy response is below.
Migrating: the reward for building it right
A movements-based workbook exports into any real system in an afternoon: products as a table, quantities baselined by one honest stocktake, movements history archived. We build BSimple, so weigh that — import is a first-class feature (your columns, your list), after which the record gains what Excel cannot give: live shared truth, guardrails on bad entries, scan-to-record, reorder drafts, and invoices that push to Xero or MYOB (US integrations rolling out). Plans run $180/$250/$399 per month AUD, and the trial runs your actual workbook beside it before any commitment. The free route stays honest too: the free-options comparison prices the alternatives, including staying.
Frequently Asked Questions
Can Excel work as an inventory management system?
For one person with moderate volume, yes — built as movements with derived quantities and a monthly count. It stops being management when shared truth, guardrails or floor-level scanning enter the requirements.
What is the biggest mistake in Excel inventory systems?
Typing quantities into a totals cell. It feels efficient and destroys everything: no history, no audit, no way to explain a variance later. Movements in, quantities derived — that rule carries the whole build.
How do we share the file without breaking it?
Carefully — shared workbooks and cloud copies resolve conflicts in ways that fork the record. The honest sequence: one owner, one file, or move to a system built for simultaneous users. The Sheets variant is the sharing-aware middle step.
When is Excel the wrong choice outright?
From day one, when the operation is multi-location, batch-tracked, or selling against live stock to walk-in customers — the guardrails those need are not a spreadsheet feature that can be added later.
What does moving to BSimple involve?
Export products, import, one baseline stocktake, movements recorded in the system thereafter — the trial runs the whole thing on your workbook before any money changes hands.
BSimple