We spend a day with the people who would use it, in the truck, on the site, on the warehouse floor. What they do with their hands, in gloves, in the rain, decides the design.
Bring in the business
Mobile apps, only when the case is real
Most app requests that reach us are answered better by a web page that works properly on a phone, and we will say so on the first call, before you have paid us anything.
When an app is right, it is usually about money that is stuck. A signed delivery note sitting in a driver's cab until Thursday is an invoice you cannot send until Thursday. Capture it at the gate and the invoice goes out that same night, on every load, every week. It is quoted per job, and you see the number before anything starts.
01
An app earns its place when
- It has to work with no signalFarm gates, basements, mine sites, the last thirty kilometres of a rural route. A web page cannot do this. An app can.
- The phone's hardware is the pointCamera, barcode scanner, GPS trail, a signature, a photo tied to a place and a time that nobody can edit afterwards.
- The same people use it every dayTwenty drivers, twelve technicians, a warehouse crew on two shifts. Daily use by a fixed team is what pays the build back.
- A notification has to reach someone who is not looking at a screenA load reassigned, a delivery gone wrong, a shift starting in ten minutes.
- The app is the productCustomers pay for access to it, and it is the thing you sell.
02
What we usually build instead
| What people ask for | What usually works better | Why |
|---|---|---|
| An app so customers can browse our catalogue | A fast website that works on a phone | Nobody installs an app to look at a catalogue once |
| An app so staff can see their jobs | A web page saved to the home screen | Looks like an app on the screen, costs a fraction, updates the same day |
| An app for loyalty cards | A pass in Apple Wallet or Google Wallet | Already on every phone, nothing to install |
| An app so customers can book | A booking page and a reminder message | The booking happens once. An install is a bigger ask than the booking is. |
| An app to replace our paperwork | A form on a phone, connected to where the work is recorded | Same result, live in weeks, no store review |
03
What it costs to build, and what it costs to keep
| What you are building | Cost |
|---|---|
| A web page saved to the home screen, no store involved | Quoted per job, and you see the number before anything starts |
| One platform, staff only, offline capture, installed straight onto company phones | Quoted per job, and you see the number before anything starts |
| Both platforms, customer facing, published in the stores | Quoted per job, and you see the number before anything starts |
| Keeping it alive each year | A care plan from R7,000 a year, quoted for your app, plus about R2,000 a year to Apple and a once-off of roughly R450 to Google |
What moves the price: whether it is a web page saved to the home screen or a real installed app, one platform or two, staff only or customer facing, whether it has to work offline, and whether it goes into the app stores. Exceptions are the real cost. A perfect delivery is easy. A short delivery, a refused load and a damaged pallet are three more paths to build and test.
"An app is never finished" is not a sales line, it is a budget line. Phones and operating systems change every year, and a version left alone for two years stops installing. We give you the yearly number with the quote, not in month four.
04
How an app project runs
We agree the one job the app must do well
Everything else waits. Apps get abandoned because version one tried to do six things and did none of them fast.
We put a working version on real phones early, with real thumbs and real signal, long before it is pretty.
We connect it to the system the office already uses, so nobody in the office gains a second place to look.
We publish it, or install it straight onto company phones
if it is staff only, which is faster, cheaper, and lets us push a fix the same day.
We stay on for the first three months
People find shortcuts, and an app after three months of real use is a different app to the one you signed off.
Who is in the room, and who holds the keys
We work over screen share by default, and for most projects that is genuinely enough. An app is the exception. Somebody has to stand where the work happens for a day, and that travel is written into the quote before you sign, not invoiced afterwards.
What we need access to:
- A few real phones, the actual models your team carries, not the newest one in the shop.
- Your app store accounts, opened in your company's name with us added as a developer you can remove.
- The system the office records work in, on a login your team creates.
If the app runs on personal devices, we say exactly what it can see, which is the job and nothing else.
On the question every owner should ask: there are two of us. The code lives in a repository in your company's name, the store accounts are yours, and a written handover document exists from the first month rather than the last. If neither of us were here, another developer clones it and carries on. Nothing about your invoicing depends on us being reachable.
06
Where AI fits in an app like this
Usually not on the phone. Behind it.
The driver photographs a signed delivery note the way they always are, at an angle and in bad light. The reference number and the quantities are read off it, the load is matched and closed, and nobody in the office types anything. The driver did nothing extra. Somebody in the office stopped doing an hour a day.
That is the real shape of AI in operations: invisible to the person in the field, and visible in the admin bill. Reading documents at volume is priced separately from the app, and it drops sharply as the count goes up. Past about a thousand documents a month we price the whole job as one monthly figure rather than per document, and we set that out under document automation.
The numbers, in one place
- Quoted per job, and you see the number before anything starts.
- What moves the price: whether it is a web page saved to the home screen or a real installed app, one platform or two, staff only or customer facing, whether it has to work offline, and whether it goes into the app stores.
Straight answers
Questions owners ask us
If it is for your staff on company phones, one platform is often enough and it cuts the build cost noticeably. If it is for customers, you need both.
Yes, and if you need offline then an app is very likely the right call. It holds what was captured and sends it the moment there is signal, without the person having to remember to press anything.
Not for a staff app. It installs straight onto company devices, which skips store review and lets us push a change on the same morning you ask for it.
That is the risk worth taking seriously. A driver goes back to paper on day two if the app cost him thirty seconds once. It is why we design in the yard rather than the boardroom, and why the first version does one job.
Almost always, and that is what we would suggest. A form on a phone, connected properly, tells you inside a month whether the daily habit sticks. If it does, that is the strongest case anyone can make for building the real thing.
The next step
Send us a photo of the paper the person in the field fills in today, creased and all, on WhatsApp to +27 73 932 6852.
We come back with whether that needs an app or a page, and what each one would cost. Email [email protected] works the same way, and twenty minutes on a call is usually enough to settle it.
Or email [email protected]. No slides, and no invoice for the conversation.