Replace a spreadsheet with an app when more than one person depends on it and being wrong has consequences. Until then, leave it alone. Spreadsheets are excellent tools that quietly become infrastructure, and the cost never appears on a budget line — which is exactly why it grows.
What you are actually carrying
Key-person risk. One person understands the formulas. When they are on leave, the process stops. When they resign, it stops permanently and expensively — someone has to reverse-engineer nested lookups, hidden tabs and a macro nobody else has opened, usually in the week a report is due.
No audit trail. Who changed the rate? When? Based on what approval? A spreadsheet cannot tell you, which matters enormously the first time it is questioned — by an auditor, a client disputing an invoice, or a regulator asking how a licence expiry was missed. Version history in OneDrive shows that a file changed, not why or on whose authority.
Silent errors. A dragged formula, a filtered row left hidden, a paste that overwrites a column. Nothing alerts anyone. The number is simply wrong, confidently, for months. Spreadsheets have no concept of a required field or a valid range unless someone builds it, and the validation people do build is easy to paste over.
The version question. Someone emails a copy. Someone edits the copy. Now there are two truths and no way to tell which is current. Co-authoring in SharePoint helps, but only if everyone opens the same file — and the habit of saving a local copy "just to be safe" is hard to break.
| Cost | What it looks like day to day | When you usually find out |
|---|---|---|
| Key-person risk | Questions routed to one person; work waits for them | Leave, illness or resignation |
| No audit trail | "Who changed this?" answered from memory | A dispute, audit or incident review |
| Silent errors | Totals that look plausible | Month-end that does not reconcile |
| The version question | Files named final, final v2, final (Sam) | Two people quote different numbers in one meeting |
Signs it has already become infrastructure
Nobody decides to run a business process on a spreadsheet. It happens one reasonable step at a time. These are the signs it has already happened:
- There is a tab called "Instructions" or "DO NOT EDIT"
- People ask the owner to update it rather than doing it themselves
- It is emailed out on a schedule, as a report
- Other workbooks link to it, so moving or renaming it breaks things
- Someone has written a macro to tidy it up each month
- A client invoice, a payroll run or a compliance record depends on a number in it
Two or three of these and the spreadsheet is no longer a tool someone uses. It is a system the business runs on, without the controls a system should have.
What it costs in time
The risks above are occasional. The time cost is every week. It hides inside other people's jobs, so nobody adds it up.
Example only: one admin plus three project managers. Swap in your own hours.
In this example that is 15 hours a month. At an illustrative $75 an hour, that is $1,125 a month, or $13,500 a year — for one spreadsheet, before counting a single error. Your inputs will differ; the point is to do the sum with real ones. The calculator runs the same arithmetic for a process of your own.
How a change travels today
Most of the cost sits in the steps between people, not in the spreadsheet itself. Here is a typical path for something as simple as updating a supplier rate.
- 01Supplier emails a new rateto whoever they last dealt with
- 02Forwarded to the sheet ownerwaits until they are back
- 03Edited in a local copyapproval, if any, sits in an email
- 04Copy sent aroundother copies now disagree
- 05Same change in an appentered once, approved in Teams, logged with name and date
The app version is not faster because it is cleverer. It is faster because there is one copy, a defined approver, and a record of who did what.
When a spreadsheet is the right answer
Genuinely often:
- One person uses it
- The logic changes constantly
- It is exploratory, not operational
- Getting it wrong costs nothing
Modelling a scenario? Spreadsheet. Running your compliance register? No.
| Keep it in Excel | Move it to an app |
|---|---|
| Budget scenarios and what-if models | Compliance and licence-expiry registers |
| One-off analysis for a board paper | Timesheets, leave and expense capture |
| A personal tracker | Asset and equipment registers used by several people |
| Logic still being worked out | Approvals with a dollar or safety consequence |
Excel also remains the right place to analyse data that an app collects. Moving capture into an app does not mean taking Excel away from the people who are good with it — it means they start from clean data.
The line worth drawing
A spreadsheet becomes a liability when more than one person depends on it and being wrong has consequences. That is the moment to move it — not when it becomes slow or ugly.
Slow and ugly are tolerable. A shared, consequential spreadsheet is a system without any of the controls you would expect of a system: no required fields, no permissions by role, no history of decisions. Naming it that way usually makes the decision easier.
What replacing it involves
Replacing a spreadsheet is less dramatic than it sounds. In most cases the data moves into a SharePoint list your Microsoft 365 licence already covers, a simple Power App or form replaces typing into cells, and a Power Automate flow handles the approvals and reminders that currently live in someone's inbox.
The real work is upstream of the build. Before anything is built, someone has to write down what the spreadsheet actually does — which columns matter, what the rules are, who approves what. That step often exposes rules that exist only in one person's head, which is the key-person risk made visible. We explain why that step comes first in write the process before you automate.
- Name everyone who edits it and everyone who relies on its output
- List the columns that are actually used, and the ones nobody understands
- Write down the rules hidden in formulas, in plain English
- Identify who approves changes today, even informally
- Decide what history needs to be kept, and for how long
- Confirm the replacement can still be exported to Excel for analysis
What replacing it should cost
A focused replacement — a register or capture app on SharePoint lists, with its approvals — is scoped as a single build from $10,000. Smaller replacements are often done inside a retainer, from $3,000 a month, alongside other work rather than quoted on their own. Because it runs on SharePoint lists, there is usually no new licence to pay for; the Power Apps licensing guide explains when that changes.
Weigh that against the cost of one bad month-end, or the week you lose when the owner leaves. If the honest sum says the spreadsheet is cheaper to keep, keep it — not every spreadsheet deserves an app, and we cover the cases where it does not in when not to automate.
The spreadsheet did not cost you anything to build. That is why nobody notices what it costs to keep.
What to do next
Pick the one spreadsheet in your business that people are nervous about — the one that only one person really understands. Ask three questions: who relies on it, what happens if it is wrong, and what happens if its owner leaves next month. If the answers make you uncomfortable, that is your candidate.
Then put a number on it. Count the hours spent feeding and checking it in a typical month, multiply by a loaded hourly rate, and compare the yearly figure with the cost of replacing it. If you would like help scoping the replacement, our Power Apps service sets out how we approach it.
Want the version for your business?
Work out what manual processes are costing you — three minutes, no contact details needed.
Work out the cost