Replacing Excel with software that knows your trade

1. 01

   One person knows how the file works, and everyone waits for them to be in.

2. 02

   There are several versions of the workbook — “v3”, “final”, “final-2” — and nobody is sure which one is right.

3. 03

   The same information is typed several times: in the quote, then the tracker, then the invoice.

4. 04

   A formula broke one day, and it was found at invoicing time.

5. 05

   The file goes out by email for someone to fill in, comes back, and gets merged by hand.

6. 06

   Friday’s figures are no longer true on Monday, because someone corrected a row without saying so.

7. 07

   A simple figure — this month’s margin, the jobs still pending — means reopening several files and recomputing by hand.

8. 08

   Your field teams, or your clients, have no access: everything goes through you.

9. 09

   The workbook has grown heavy and slow, and people hesitate to touch it.

10. 10

    You already tried off-the-shelf software and came back to the spreadsheet because it would not do your calculation.

Is Excel a bad management tool?

No. It is an excellent tool for setting out a calculation, and many companies are right to start there. The problem is not the spreadsheet, it is what it ends up being asked to do: serve as a shared database, as an entry tool for several people and as the company’s memory, all at once. It was designed for none of the three.

How do I know it is time to replace it?

When the file costs more time than it saves: double entry, versions that contradict each other, a broken formula found too late, a simple figure that takes a morning. The ten signs above are the list we go through in a first conversation. If you recognise several, the question is worth asking. If you recognise none, keep your spreadsheet.

Off-the-shelf software or custom software?

It depends on what your workbook contains. If it reproduces a standard process — invoicing, bookkeeping, a calendar — off-the-shelf software will do better, for less, and we will tell you so. If it contains a rule that exists only at your company — a commission computed your way, a margin per job that mixes several sources, a schedule bound by your trade’s constraints — that rule is what costs so much to work around in a configured tool, and that is where custom software earns its place.

What becomes of my Excel file?

It becomes the specification. We read its formulas, translate them into written rules, check them with you on real cases, and they become the core of the software. Your historical data is carried over, cleaned and imported. And Excel stays what it is good at: exporting a table, running a one-off analysis, drafting a calculation not yet settled.

Will I lose my historical data?

No. Carrying the data over is part of the scoping: which sheets, which columns, from which date, and what to do with the inconsistent rows — there always are some. That sorting is often the first time a company looks its data in the face, and it is done with you, not for you.

How long does it take?

We give no duration before scoping, because a ten-column workbook and a ten-tab workbook whose tabs call one another are not the same project. The scoping itself is written: it says what the software will do, in what order, and what it will not do. The scope is cut to deliver early, and a usable version arrives before the end, not at the end.

How much does it cost?

The quote comes after the scoping, on a written scope — never before. What we can say beforehand: custom software is paid for what it does, with no per-user licence; off-the-shelf software is paid in subscriptions, in configuration, then in workarounds. The honest comparison is made over several years, and it depends on how many people use it.

My team is not comfortable with computers. Will it work?

Custom software is drawn for the people who will use it, and for them only: the screens follow your vocabulary and the order you work in, not a generic menu. There is often less to learn than with the spreadsheet it replaces, because there is no longer a formula not to break or a column not to touch.

What if my needs change in a year?

A rule that changes is rewritten; a rule that is added is added. That is precisely what a spreadsheet does well at first and worse and worse afterwards: it accepts everything, until the day nobody dares change it. Software whose rules are written down and tested changes without fear.

Why not move to Airtable, Notion or a no-code tool?

Because it sometimes is the right answer, and we say so when it is. Those tools are excellent as long as your rule fits in their formulas and few people use them at once. They reach their limits in the same place as the spreadsheet: a calculation that crosses several tables, a rule with exceptions, an automation that cannot be tested. At that point you have rebuilt the workbook, with a subscription on top.

Will I depend on you?

No. The code and the data belong to you, the software is built on standard, documented technologies, and every management rule is written out in plain language in the documentation. Another developer can take it over. We would rather be chosen than indispensable.

And what about AI?

After, not before. An assistant answering from your data is only worth something if the data is clean, structured and current — which a shared spreadsheet never guarantees. The management software comes first; AI comes next, when there is something reliable to give it to read.

The work is described without naming clients: the trade, the order of magnitude and what was built — never the identity.