From R28 000
Internal tools & dashboards
The spreadsheet six people edit at once, rebuilt as software. From R28 000.
An internal tool is software nobody outside your business ever sees. A dashboard that shows the real numbers, a form that captures a job properly, a screen that replaces the spreadsheet everyone is editing at the same time. I build one, around your process, from R28 000.
- Good fit
- You run something that matters on a spreadsheet, several people edit it, and it has started to break.
- Not a fit
- If the spreadsheet still works, or you want a big off-the-shelf ERP configured rather than a small tool built, I am the wrong person.
No internal tool has shipped yet
Straight up: my record is five website builds. Not one internal tool or dashboard has gone live. If you want to click a working example before you decide, I do not have one.
What carries over is the method. I measure before I design and verify after I build — colours sampled off the rendered page for real contrast ratios, layouts run through headless Chrome at 320, 390, 768 and 1440 to catch overflow. A dashboard lives or dies on that same discipline.
Missing data has to look missing
On the timber decking build I typed Janka hardness as a number or null, because composite boards have no Janka figure and printing "0 lbf" would have been a lie. Same rule on a dashboard. A blank field, a stale feed and a genuine zero are three different things, and the screen must say which one it is.
A number with no provenance is decoration. If your team cannot trust the tile, they go back to the spreadsheet and you have paid for nothing.
What I build with
Astro 5, Tailwind CSS 4, vanilla JavaScript. No React, no component library, no bought dashboard template. Deployed to Google Cloud Run in europe-west1, nginx from a container — the same setup the live sites run on.
Fewer dependencies, fewer things to break.
What you get
- A screen list agreed and priced before I build
- Tables with filter, sort and search on your real records
- Charts and counters wired to live data
- Forms with validated inputs and visible error states
- Separate views per role, where you need them
- Every colour contrast-measured before it ships
- Layout verified at 320, 390, 768 and 1440px
How it runs
- 01 Send me the spreadsheet The file or system you run this on now, and who touches it. The mess is the spec.
- 02 Screen list, then a price I write down every screen, field and permission. The quote comes off that list, not a guess.
- 03 Built screen by screen I build in order of what you use most. You click through it as it goes, not at the end.
- 04 Verify, then deploy Contrast sampled off the rendered screens, overflow checked in headless Chrome at four widths, then it goes live on Cloud Run.
Straight answers
- Have you built one of these before?
- No. Five websites are on record and no internal tool. Judge me on the method and the screen list, not on a portfolio slide I do not have.
- What does R28 000 actually get me?
- It is the floor, not an average. Scope sets the number, and I quote against the agreed screen list so you can see what each screen costs before anything is built.
- Can it read the systems we already use?
- If it has an API or can export a file, usually yes. If it is a closed desktop program with no export, I will tell you that before you pay, not after.
- Will it work on a phone?
- Yes. I check layouts in headless Chrome at 320, 390, 768 and 1440 pixels, so a foreman on a small screen gets the same tool as the office.
Start a job
Need internal tools & dashboards?
You run something that matters on a spreadsheet, several people edit it, and it has started to break.
Pick whichever suits you