“It Makes a Weird Noise”: Better Service Intake for Independent Garages

At 7:45 on a Monday, a customer drops off their car. "It makes a weird noise when I brake, sometimes. Oh, and a light came on last week." The service advisor, with three more customers in line, writes on the repair order: Noise when braking. Check engine light.

The technician takes the car for a test drive and hears nothing. Forty minutes later, after a second drive and a call to the customer that goes to voicemail, he finds out what the customer would have said if anyone had asked: the noise only happens in the morning, when turning right into their driveway, and it started after the tyres were changed at another shop last month. The check engine light, it turns out, is a separate issue entirely.

None of this is anyone's fault. It's what happens when a complex, intermittent problem is summarised in one line at a busy counter. And it's a problem AI can help with before the car even arrives, simply by asking the questions a good service advisor would ask if they had the time.

Why intake matters more every year

Older cars need more repairs, and today's older cars are full of electronics. At the same time, there aren't enough technicians. That combination makes a technician's hour the most valuable thing in the shop, and diagnosis is where most of those hours go. Every minute spent re-interviewing a customer or trying to reproduce a symptom nobody described properly is a minute that isn't billed or isn't spent on the next car.

One line versus a real concern

The auto repair trade has a good framework for this already: the three Cs on a repair order, concern, cause, correction. The technician writes the cause and correction. The concern is supposed to capture the problem as the customer experiences it. In practice, it's usually the weakest of the three.

The right column isn't the diagnosis. It's the information that lets a technician start diagnosing in the right place: brakes on the front right, check the work done during the tyre change, and treat the warning light as a separate job with its own scan.

The questions that matter

Good service advisors ask these instinctively. Under pressure at the counter, they get skipped. A structured interview makes sure they're asked every time, in words customers understand.

The last row is the one that saves the most time. "Started after the tyre change" or "since the battery was replaced" points straight at the likely cause.

Intake before the car arrives

The video deserves special mention. For intermittent noises, a fifteen-second phone recording made by the customer at the moment the noise happens is worth more than any description. It lets the technician hear the squeal before the test drive, and sometimes it answers the question outright.

What the customer sees

The questionnaire should feel like a conversation with a helpful person, not a form. For the braking noise, it might go like this:

What's the problem? "Weird noise when I brake, sometimes."

What does it sound like? Squeal · Grinding · Clicking · Clunk · Something else → Squeal

When do you notice it most? First drive of the day · After driving a while · Always → First drive of the day

Does anything make it more likely? Turning left · Turning right · Braking gently · Braking hard · Going over bumps → Turning right, braking gently

Has any work been done on the car recently, here or anywhere else? → "New tyres about a month ago at the tyre place."

If you can, record a few seconds of the noise next time you hear it. → Video uploaded

Six taps and one sentence of typing. Every answer narrows the search, and none of them requires the customer to know anything about cars. The model's job is to pick the next question based on the previous answer and to turn the result into the concern the technician reads.

What it saves

Some of that time was never billed in the first place, because no customer wants to pay for a technician trying to guess what they meant. Freeing it either brings in more work with the same team or lets the team finish on time.

After the diagnosis

The same approach helps on the other side of the job. Once the technician has found the cause, the customer needs an explanation they can understand before they approve the repair. A draft written from the technician's notes and inspection photos saves the advisor ten minutes per car:

The squeal comes from the front right brake. The pad's wear indicator is touching the disc, and the pads are close to the minimum. We'd recommend replacing the front pads and checking the discs, which look fine for now. Photos attached. The check engine light is a separate issue: the scan shows a sensor reading we'd like to test further before recommending anything.

The advisor checks it, adjusts the wording, adds the price, and sends it. Customers who understand what they're paying for approve faster and trust the shop more.

Where the line is

I'd hold that line firmly. A language model can produce a very convincing explanation of what "probably" causes a noise, and customers will remember it when the invoice says something else. The shop's credibility comes from the technician's diagnosis, not from a chatbot's guess.

The intake checklist

Tools that fit

Most independent shops use shop management software such as Tekmetric, Shopmonkey, Shop-Ware, or Mitchell 1 Manager SE, often with a digital vehicle inspection tool. The intake plugs in at booking: it creates or updates the appointment with a structured concern, the customer's attachments, and the vehicle history. The technician reads it where they already read everything else.

Customer answers and videos are personal data. Keep them with the repair order, use them for the repair, and follow your normal retention rules. In the EU, customers should be told when an automated assistant is asking the questions.

Questions shop owners ask

Won't customers find the questions annoying?

Short, relevant questions that adapt to their problem aren't annoying, especially if you explain the reason: "so we can find the problem faster and you pay for less diagnosis time". What annoys customers is being called back three times.

Can it read warning lights from a dashboard photo?

It can usually recognise the common symbols and colours from a clear photo. It can't read fault codes; that needs a scan tool in the shop. Treat the photo as context, not as data.

Does this replace the service advisor?

No. It gives the advisor a head start and a structured note to confirm at the counter. The conversation, the authorisation, and the customer relationship stay with the advisor.

What about walk-ins with no booking?

Hand them a tablet or a QR code at the counter and let them answer while they wait. Even two minutes of structured questions beats a one-line note written under pressure.

The rule for every repair order

Write down the concern the way the car behaves, not the way the customer guesses. Ask the same good questions every time, and let the technician start in the right place.

If vague repair orders are costing your technicians time, tell me which shop software you use and how bookings reach you. I'll suggest how a structured intake could work in your shop. The same "prepare before the visit" idea helps appliance repair technicians bring the right part, and good intake matters just as much for bike shops in the spring rush.