Why trade software fails on site: the van-seat test

Short answer: admin tools are judged on features and used on the passenger seat of a van, in the rain, one-handed, with twenty seconds of patience available. Anything that fails that moment does not get used, however good the feature list is — and unused software is worse than no software, because you are paying for it.

The test

Imagine the real moment. You have finished a job. You are sitting sideways in the driver’s seat with the door open. Your hands are not clean. It is half four and getting dark. Your phone is at 14%. You have a number in your head — the miles you just did, or the price you just agreed — and you have about twenty seconds of willingness before you decide to do it later.

Now ask what the tool requires of you in that moment. Open an app. Wait for it to load. Log in, because it has forgotten you. Find the right screen. Pick the job from a list. Fill four fields, two of which are mandatory and one of which you do not know yet. Tap save.

That is not a hard task. It is just longer than twenty seconds, and so it becomes later, and later becomes Sunday, and Sunday becomes never. The failure is not laziness. It is a mismatch between the design assumption and the moment of use.

Why good software fails this test

Because it was designed where it was built: at a desk, on a large screen, by someone with both hands free and no customer waiting.

Field software has a different and harsher constraint. The cost of capture has to be lower than the cost of remembering. If logging a trip takes longer than holding the number in your head until later, the software loses to your memory — and your memory is unreliable, so the data is lost either way.

This is why feature comparisons mislead so badly in this category. A tool with twice the features and four times the friction captures less, and captured data is the whole point.

The four things the van seat punishes

Install and login. Every app you must install is a decision someone has to make, and every login is a moment to abandon. Both happen before any value is delivered.

Structure demanded up front. Software that insists you create the customer, then the job, then the line item, before you can record anything, is asking you to do data entry when you wanted to record a fact.

Required fields you do not have yet. On site you often know the price but not the start date, or the distance but not which job it should sit against. A form that will not accept partial truth gets abandoned mid-entry.

What passes

Whatever you are already holding, already logged into, and already using forty times a day. For most UK trades that is a messaging app, because that is where the customer conversation lives anyway. The quote was discussed there. The photos went there. The address is in the thread.

That is the whole design argument for doing admin by message, and it is not a feature claim — it is an observation about where attention already is. Sending a line of text takes about four seconds and needs no decision. The structure can be worked out afterwards by the software, which has more patience than you do at half four in the rain.

Free text is ambiguous, so sometimes it has to ask you a follow-up question, and you give up the dense overview that a proper screen gives you. Message-based admin is excellent at capture and worse at browsing. If what you need is a big board showing six engineers’ weeks, this is the wrong shape of tool and a scheduling system is the right one.

Apply it before you buy anything

Take whatever tool you are considering and walk through that twenty seconds out loud. Ask what happens with one hand and no signal. Then ask the honest question: on a Thursday in February, having done nine hours, will I actually do this?

If the answer is no, the features are irrelevant. This is the same reasoning behind why the same-day quote wins work the careful one loses, and the same reason tools without a process do not stick.

What we built from this

GraftG is the argument above turned into a product: quotes, invoices, job tracking, mileage, receipts and a diary over WhatsApp, no app to install. It is in beta, which means rough edges and a fast fix when you report one. There are 21 Founding Tradies places — twelve months free on Pro, no card, then half price for life. Apply the van-seat test to it before you commit to anything; that is what it was designed to pass.

Questions, or want a tool we don't have yet? Email hello@grafto.co.uk — a real person replies.