What the symptom check cannot know

Six limits

• It reads ten answers about a symptom. That is a starting point for a conversation, not a diagnosis, and it is not a survey of the vehicle.
• It never names a fault. Everything it produces is a list of what would be MEASURED and in what order, because a fault named from a description is a guess wearing a result screen.
• There is no dataset behind it. It does not know what usually goes wrong on your make and model, because we have not measured a population of them and would not present it as evidence if we had.
• It does not score, rank, grade or rate. A number would read as a measurement and there is nothing here to measure against.
• The figures it prints are the published hourly rate multiplied by a range of hours for a named piece of work. That is what that first piece of work costs, not an estimate for putting right a vehicle nobody has seen.
• Two of the twelve questions are chosen from your earlier answers, so two people will not necessarily see the same ten.

Nothing has been measured yet. This is an indicative scope, not a quotation and not a diagnosis. It reads a description of what the machine is doing and names the circuits that would be tested first — it has not seen the vehicle, it cannot know what is wrong with it, and no fault named from a questionnaire is a fault anybody has found.

The one that matters most

It never names a fault. Everything it produces is a list of what would be MEASURED and in what order, because a fault named from six answers is a diagnosis from a questionnaire — and this whole website is an argument against exactly that, made by a business whose customers keep arriving having had four parts fitted on the strength of a guess.

So a result screen will say that the first hour would go on the starting circuit, and what would be measured on it, and what those measurements separate. It will not say that your starter has failed. It cannot know that, and neither can anybody who has not put a meter on it.

Why it prints a figure at all

Because the alternative is worse. A tool that describes work and then hides what the work costs is doing the thing this site exists to argue against. What it prints is the published hourly rate multiplied by a range of hours for a named piece of work — arithmetic anybody can check against the rates page.

What it never prints is a single figure. That would read as a quotation for a repair on a vehicle nobody has seen, and the difference between those two things is the whole of the honesty argument here.

Every figure on this site is an hourly rate multiplied by a number of hours, and both numbers are printed. A workshop hour is $145, an on-site hour is $185, a bench hour is $125 and an after-hours on-site hour is $265. Parts are at cost plus twenty per cent with both figures on the invoice. There is nothing else: no callout fee on top, no shop supplies percentage, no environmental levy.

What it is good for

Arriving at a telephone call with the right question already asked. The most useful thing it does is not the result screen at all — it is the questions, which are the ones we would ask anyway, in the order we would ask them. People who have been through it tell us more in the first minute than people who have not tell us in five.

Go and do it