Google advertising for a car repair shop on the day of the breakdown
In the morning someone rings and asks only one thing — do you have room today. The car has stopped outside their building, they are already late, and by the end of the day the question will be settled one way or another. If the phone was busy, or rang with nobody picking up, the decision was made without you.
The search starts from a breakdown, not from a wish
Most advertising works with demand that can be put off. A person buys furniture this spring or this autumn. A car repair shop sits at the other end of the scale. Nobody wakes up wanting a repair. The demand is created by a physical event: a warning light on the dashboard, a noise from underneath, smoke, a car that will not start.
The event has a beginning and almost always an end within the same day. The person searches, rings a few places, gets an answer and the matter is closed. The enquiry does not sit in a basket, does not come back later, does not mature. By evening it no longer exists.
Everything else follows from this. The whole path from stranger to client happens between the moment of the breakdown and the moment the car has to be moved. Every setting in the account either falls inside that window or misses it. There is no middle ground.
A click costs something only while there is someone to answer the phone
The bidding system optimises against what it can measure. It sees impressions, clicks and recorded conversions. It does not see whether the phone at the garage was busy, whether you were under a car with oil on your hands, whether anyone picked up at all. To the algorithm, the hour when calls start coming in looks like a good hour, regardless of what happened to those calls.
This is exactly where the link breaks. A missed call is not a separate line in the account. It is either not counted at all, or it enters the report only as a started call, and the difference between a phone that rang and one that was answered lives in the phone, not in Google. So the garage pays steadily for hours in which it never had a chance to take on the work, and nothing in the report hints at it.
Ad schedule hours are not a way to save money. They are a way to make spend match your capacity to answer. When I take over a garage’s account, the distribution by hour is among the first things I look at, and it almost always sits as the campaign launch left it. The garage’s capacity changes during the day — the morning rush, lunchtime, the last hour before closing — while the ad schedule stays fixed over all of it. The same money buys a different amount of work at different hours.
Urgent and planned work are two different businesses in one account
Two different searches come into the same garage. The urgent one sounds like a symptom: will not start, smoking, knocking from underneath, a light on. The planned one sounds like a service: tyre change, oil, annual inspection, seasonal preparation. The words are different, the hours are different, and so is the distance a person is willing to travel. After closing time, the same word is typed by a different person — one who is no longer choosing where to go right now, but arranging tomorrow.
When the two sit in one campaign, bidding averages them out. Planned searches are more numerous and cheaper, conversions from them come more easily, and gradually they begin to teach the system what a good click is. The seasonal peak for tyres swallows the budget exactly when urgent breakdowns are also at their highest, because the cold brings out both at once.
The split is not a cosmetic change to the structure. It gives each type its own hours, its own radius, its own copy and its own conversion. I separate the two before I touch the budget, because after that every setting has an address.
The radius is measured by towing, not by kilometres
The circle around the pin on the map assumes that the car can come. With an urgent breakdown, that is exactly what is not true. The distance that matters is how far a person is willing to pay for towing, and that is less than the distance they would travel themselves, and it is not the same in every direction.
The geography of the city cuts the circle unevenly. A river, a ring road, a traffic jam in the direction people commute to work. A person on the other side of the city is the same number of kilometres away on the map and will never come to you, because their route passes other garages. The circle includes them and you pay for them. When I draw the geography for a garage, I start from the addresses of people who have already come, not from a circle around the pin.
The location setting is separate. By default, an ad may also be shown to people who show interest in the area without being in it: someone from another city who is reading, or who typed the city name into the query. For planned work this sometimes makes sense. For a car that will not start, only physical presence matters.
The radius also has an hour. In the morning the car stops outside the building, at lunchtime in the car park at work, in the evening on the way home. The same garage is close to different people at different times of day, and the person searches from where they are standing, not from where they live. That is why radius and hours are not two separate settings but one.
The copy answers the question that has already been asked
An enquiry for an urgent breakdown carries the breakdown itself. The person has typed the symptom, or typed a garage and the city. At that moment they are not comparing offers but filtering. They go through the results and remove everything that does not answer their single question — can the car be taken in today. Whatever is left, they call.
That is why experience, equipment, the list of makes and years on the market do not do the job at that moment. They answer a question the person has not yet asked. They will ask it when choosing where to take the car for a planned repair. In the urgent ad, the same place should be taken by availability, opening hours and a phone number visible on screen without scrolling.
The copy also does another job that is rarely noticed. It turns people away. If you do not take heavy vehicles, do not work on certain systems, do not accept anyone without a prior call, this must be in the ad. A click costs the same whether or not the call after it fits your profile, and a wrong call eats up the same resource as a right one — the person who has to take it.
The account counts calls, not repaired cars
The conversion you have set determines what the system learns. If it is a tap on the phone number on the mobile page, the account counts intent, not work, and learns to find people who tap the number rather than people who reach the garage. The two groups overlap only partly.
The difference in value also disappears. An urgent call and a call about an oil change are counted as the same conversion. The system directs the budget to where conversions come more easily, and that is almost always the small planned work. The garage sees a growing number of conversions and shrinking revenue, with nothing in the report looking broken.
That is why I insist that the conversion records a call that turned into a conversation, and that the urgent and planned campaigns optimise towards different goals. Otherwise everything described so far — the hours, the radius, the copy — is tuned according to the wrong feedback.
One thing you can check today
Open the Google Ads account and pull the report by hour of day for a longer period back, with clicks and conversions by hour. Then take the phone that receives the calls from the ads and open the call log for the same days. Do not touch anything, just put the two side by side.
You are looking for the overlap between the hours in which the account pays and the hours with unanswered calls in the log. That is the part of the spend that never had a chance to become work. In the report it simply looks like clicks without conversions and is explained away with all sorts of reasons. In the phone log it looks exactly like what it is.
If you see numbers that repeat at the same hours and were never answered, those are the people who went to the next garage. The check takes as long as it takes to open two screens and does not require you to change a single setting. Whatever you decide after it — hours, radius or copy — will rest on something you have seen, not on an assumption.
Frequently Asked Questions
Should I switch the ads off outside working hours?
It depends on whether there is someone to answer, and it is decided separately for the two types of search, not for the account as a whole. The urgent campaign buys clicks that have nowhere to go once the phone is left unattended. The planned one can tolerate a delayed reply, because that person is not heading to a garage this evening anyway. If you pause the whole account, you also pause the second campaign, and it is the only one with something to do in those hours.
How large should the radius around the garage be?
It is not measured in kilometres on the map but in reachability: for an urgent breakdown the boundary lies where a person is willing to pay for towing. That is why the radius for urgent and planned work does not match — for a planned repair the car travels on its own and reaches further. If you also offer transport, the boundary shifts, because it is no longer determined by the person’s willingness to find someone to tow them.
Which works better for a car repair shop, Google or Facebook?
The two work in different ways. Google catches demand that already exists, because the person has described their own problem. Facebook interrupts someone who is not thinking about their car at that moment. With an urgent breakdown, the interruption has no time to take effect, because the choice is made within the day. For planned work, and for making the garage familiar before a breakdown happens, the second makes sense.
Related: Advertising for car repair shops · The calls your service garage’s ad account does not see · Google advertising
