Most small businesses run on at least one critical spreadsheet. The one that calculates the staff schedule. The one that holds the price list. The one that the bookkeeper opens on Monday morning to reconcile last week. Spreadsheets are wonderful, they are the cheapest, fastest, most flexible piece of software ever invented, and most of these will live forever and never need to change.

But some of them age into liabilities. This is a guide to telling the difference, and to what replacing the bad ones with a small custom tool actually looks like.

When a spreadsheet is genuinely the right tool

Spreadsheets are great when:

  • One person owns the file and is the only one editing it.
  • The logic is simple enough that anybody who opens it can read it.
  • Errors are recoverable and visible: if the formula is wrong, you notice the wrong number on screen.
  • The shape of the work doesn't change much month to month.

If your critical spreadsheet has those properties, leave it alone. Replacing it with a custom tool would be a downgrade, not an upgrade.

The warning signs

A spreadsheet has aged into a liability when one or more of the following is true.

It gets emailed back and forth. Every email creates a new version. After three weeks there are versions named prices_v3_FINAL.xlsx, prices_v3_FINAL_marko.xlsx, and prices_v3_FINAL_marko_corrected.xlsx, and nobody is sure which is the actual current price list. The cost of being wrong is a discount given to the wrong customer, paid forever.

The formula tower is unreadable. A cell that calculates the final number does so through six nested IFs and a VLOOKUP into a sheet whose tab is hidden. The original author knows what it does. They left the company two years ago. Nobody has touched it since because nobody is sure what would break.

Errors are caught after the fact. Last quarter you discovered that one column was sorted wrong in the May export, and the wrong numbers had been quietly published to clients for three weeks. Nothing in the spreadsheet warned you. The error was caught when a customer noticed.

The same operation happens every week. Somebody opens the same file every Monday, applies the same five sort-and-filter steps, copies the result to a new tab, and pastes it into an email template. Every Monday. For two years. That is custom software wearing a spreadsheet costume.

Multiple people need it, but only one can have it open. The other three are waiting, or working from a stale copy, or maintaining their own version that has to be reconciled later.

The file is the source of truth, but the truth needs to live somewhere else. The website's prices come from this file. The accountant's report comes from this file. The team's targets come from this file. Every place that reads it copies a snapshot, and the snapshots drift.

If three or more of those are true for one of your spreadsheets, the spreadsheet has aged into a liability. Replacing it pays back faster than most owners expect.

The intermediate steps people try first

Before commissioning a custom tool, most owners try cheaper escapes. They are worth knowing about because each one buys time, and sometimes time is enough.

Macros. Recording the weekly sort-and-filter sequence as a macro turns a 20-minute task into a button click. Worth doing, almost always. Caps out when the rules get complex enough that the macro stops working on edge-case rows.

Power Query. Excel's built-in ETL layer can absorb messy input and produce a clean output. Genuinely useful, especially when the input is consistent. Caps out when the input itself starts to vary, or when the output needs to be consumed by another system.

A shared Google Sheet. Solves the multi-user problem cleanly. Doesn't solve any of the others. The price list still has six nested IFs.

A no-code tool. Airtable, Notion databases, similar. Can replace the spreadsheet with something more structured. Worth trying if the workflow is generic. Becomes its own version of the same problem when the workflow is specifically yours, see Why Internal Tools Beat Another SaaS Subscription.

If you have already tried the relevant ones and the symptoms are still showing up, the spreadsheet has earned its replacement.

What replacing it looks like

The fear most owners have is that "custom software" means six months and €30,000. For a single-spreadsheet workflow, that is wildly wrong.

A typical replacement build for a critical spreadsheet looks like this:

  1. Read the existing spreadsheet directly. No new template, no fresh data entry, no asking the team to learn a new format. The tool absorbs the file as it currently exists, with all its idiosyncrasies.
  2. Encode the rules in code. Whatever the formulas were doing, whatever the manual sort-and-filter steps were, whatever the "and then I copy this to that tab" did, becomes versioned, readable code.
  3. Produce the output the team already uses. Another spreadsheet, a PDF, an email, whatever the next step in the workflow expects. No new format to learn downstream.
  4. Ship as a button. A .bat file on Windows, a small private web page, or a scheduled job, depending on how the team works. Double-click, drop the file in, get the result. No terminal. No install. No cloud login.

That kind of build runs 2 to 4 weeks and costs €500 to €2,500, depending on how complex the rules are. The numbers come from the transparent breakdown.

A worked example

The Start-List Generator on the portfolio is exactly this shape. Before: an evening of manual sorting in Excel before every tournament, applying a federation rulebook to two hundred athletes, getting the heats right, separating clubs across lanes, double-checking every row. After: drop the registration file on a button, get a sorted heat list out 30 seconds later, no errors, no late-night work.

The build was three weeks. The replacement isn't dramatic. The federation's file still goes in. A heat list still comes out. The difference is that the rules are now in versioned code instead of in someone's head, applied identically every time, in seconds instead of an evening.

What stays in spreadsheets

Even after a replacement, plenty of work stays in Excel. Quick exploratory analysis. One-off calculations. Data that one person is mostly using by themselves and that doesn't need to feed any other system. Spreadsheets are still the right answer for almost everything they were the right answer for last year. The category that needs to leave is the narrow slice that has aged into a liability, the ones with the warning signs above.

Work with me

If one of your spreadsheets is showing the warning signs, a free 30-minute call is the cheapest way to find out whether it has actually aged out. I will read the file, ask three questions, and give you an honest read on whether a custom replacement would pay back. Email info@tonibarisic.com or use the contact form. For the umbrella overview of the category, see Internal Tools for Small Businesses, A Practical Guide. For the build-vs-rent question, see Why Internal Tools Beat Another SaaS Subscription.