Two customers, same inspection, one approves the brakes and one does not.

A digital inspection produces the same recommendations regardless of how the customer found the shop; what varies is whether they say yes.

That variation tracks with why the customer came: somebody drawn by a discounted oil change is making a price decision, and somebody whose car broke down is not.

With declined work in the file, the report gives approval rate and approved value per channel, which explains a shop-wide average repair order that never made sense.

It is also the most actionable output here, because it points at the offer and the service advisor conversation rather than at the channel itself.

A channel with good traffic and poor approval is usually fixable without changing the spend at all.

Oil change and brake specials fill bays quickly and produce a first transaction that is nearly identical across every channel that runs one.

The information appears later, in whether the customer returned for work that was not discounted.

A year of repair orders summed by customer separates the channels that bought a transaction from the ones that produced a relationship.

Discount channels usually lose that comparison by a wide margin, which is not an argument against offers but against judging them on the day they run.

The report keeps first visits and later ones apart so the two effects are never averaged into one misleading figure.

A small fleet is won once, serviced regularly and lost in a single bad experience, and its annual value dwarfs a retail customer's.

Averaged in, one fleet win makes whichever channel produced it look extraordinary on a result that will not repeat.

Read separately, you get an honest retail ranking and an honest commercial one, and the sources feeding each are rarely the same list.

Where the export flags a customer type this is automatic; where it does not, the order frequency usually identifies the accounts quickly.

For a shop trying to build commercial work, that second ranking is the whole reason to run the report.

Drivers with a warning light or a noise call, and so do the ones who saw the sign, the map listing or the coupon.

Without tracked numbers those all arrive identically and the lead file covers only the small minority who used a web form.

Distinct numbers on the website, the map listing, any print and the sign itself produce a lead file with a source on every row.

Until that exists the report states the size of the untracked bucket rather than distributing it across whatever else was running.

It is worth doing before anything else on this page. A shop that tracks its calls can rank every channel it buys; a shop that does not is reading a report about the fraction of customers who happened to use a form, and that fraction is not representative of anything.