Short answer: keep the spreadsheet for analysis and one-off calculations, and replace it when more than one person edits it, when it holds the record of what happened, or when someone rebuilds the same report every week. That is the point where a custom dashboard or management system pays for itself, and the test below tells you whether you have reached it.
What spreadsheets are good at
Spreadsheets are the best tool ever made for one person thinking about numbers. Modelling a price change, checking a quote, planning a budget: nothing beats them. The problems start when a spreadsheet stops being a thinking tool and becomes the system of record, the thing the business runs on. It was never built for that, and it shows in six ways.
Six signs the spreadsheet has become the business
1. More than one person edits it. Version conflicts, overwritten rows, a file called "FINAL v3 USE THIS". Shared cloud sheets help, but they still have no idea of who may change what.
2. Everyone can see everything. A spreadsheet has no concept of roles. The person updating stock can see the margins; the part-timer can see the salaries. Most owners discover this the week they hire someone new.
3. It holds the history. If the spreadsheet is the only record of which customer was told what and when, a deleted row is a lost fact. There is no audit trail and no undo past the last save.
4. Someone rebuilds the same report every week. Monday morning starts with an hour of copying, filtering and emailing. The report is a job, not a glance.
5. Rules live in cell colours and memory. "Yellow means waiting for the client." "Don't touch column K." The logic that runs the business is stored in people, which fails on holiday.
6. It is slow or scary. Thousands of rows, formulas nobody dares edit, a tab called "old do not delete".
Three or more of these and the spreadsheet is costing you more than a system would. One or two, and you can probably stay a while.
The cost test
Write down, for each person who touches the spreadsheet, the hours per week spent maintaining it: entering, fixing, reconciling, rebuilding reports, answering "which version is right". Multiply by their hourly cost and by 50 weeks. Then add the cost of the last mistake it caused (a missed renewal, a double order, a wrong invoice).
Compare that figure with the quote for a system that removes those hours. If the system pays back inside a year, build it. If it takes three, stay on the spreadsheet and fix the worst sign instead. Ask the developer to show which of the six signs the system removes and which it does not; that is the honest version of a return calculation.
What a small business management system includes
A management system does not need to be large. It needs to fix the six signs:
- Records with proper fields and statuses, so "waiting for client" is a status, not a colour.
- Roles: who can see and change what.
- An audit trail: every change with a who and a when.
- A dashboard that is a glance, not a job: the numbers the owner opens every morning, filtered by location or team.
- Reports that are live views, not weekly rebuilds.
- Imports from the spreadsheet you have today, so day one is not an empty screen.
Spinfluence is an example of the dashboard end of this: a multi-location owner view with metrics, reviews and billing per restaurant, and per-location logins for managers (/work/spinfluence). Scriva's creator dashboard is another shape of the same idea (/work/scriva). Reasy's admin side shows the records-and-roles end: users, properties and alerts managed from one place (/work/reasy).
The middle path
You do not have to go from a spreadsheet to a coded system in one step. Airtable or a similar tool fixes signs 1 and 4 quickly. A Bubble app fixes 1 to 5 for a small team. A coded system on Next.js and Postgres is the right call when roles, audit and integrations have to be exact, or when the data is sensitive. I build the last two and will say which fits.
Whichever you choose, write the one-page brief from post 06 first: the problem, the users, the three outcomes, the rules you already know. It is the fastest way to get an accurate quote.
Which of the six signs do you recognise?
Send me a description of the spreadsheet (what it holds, who edits it, what the Monday report looks like) and I will reply with what I would replace first, whether a tool or a build fits, and roughly what that involves. Form at /contact, or see /services/management-systems-and-dashboards.