TL;DR
- Two complaints dominate contractor forums: ads that do not produce work, and calls that go to voicemail while everyone is on a job. They are usually the same problem.
- Your ad platform reports the click and the call as delivered. It has no idea the phone rang out, so the failure never appears in the report you are optimising against.
- Callers rarely leave voicemail. They ring the next contractor. That makes an unanswered call a transfer of your ad budget to a competitor.
- Fixing the answer rate is almost always cheaper than raising the ad budget, and it increases the return on the spend you already have.
- The number to find is not cost per click or even cost per call. It is cost per booked job, and getting it requires joining call data to completed work.
Spend any time reading what contractors say to each other and two complaints come up more than all the others. The first is that advertising does not seem to work. An electrician put eight hundred dollars into Google Ads, got two calls out of it, and no actual work. The second is that the phone rings when nobody can pick it up, and the caller is gone by the time anyone calls back.
These get treated as separate problems. One belongs to marketing, the other to operations. Then an owner wrote a sentence that puts them together, and once you have read it you cannot unsee it: he had been spending two thousand a month on Google Ads while ignoring fifteen thousand dollars worth of phone calls.
He was not running a bad campaign. He was running a campaign into a bucket with a hole in it.
Why the ad report cannot see this
This is the part that keeps the problem hidden for years. From the ad platform's point of view, everything worked. Someone searched, saw your ad, clicked it and pressed call. That is a conversion. It gets reported as a conversion. You paid for a conversion and you received one.
What the platform cannot see is the next four seconds, where the phone rang out because your two technicians were both in attics. It has no idea whether a person picked up, and it certainly has no idea whether a job got booked. So the dashboard shows a working campaign while the business shows a quiet week, and the owner concludes the ads are a scam.
One owner put the frustration precisely: he was trying not to burn money on things that look good in reports but never bring real customers in. That is exactly the right instinct, and the answer is that the report is measuring a different event from the one he cares about.
Nobody leaves a voicemail any more
The reason an unanswered call is so expensive is that it does not wait for you. Contractors describe this the same way over and over. Almost nobody leaves a message; they hang up and call the next name on the search results. One owner said he lost the work even when he rang back within four hours, and heard the same sentence repeatedly: sorry, I have booked elsewhere.
Another described missing two quote calls in a single week and finding both had booked someone else by the time he returned them.
Put that together with the ad spend and the picture is uncomfortable. You paid to put your name in front of somebody with a problem. You paid again for the click. Then the call went unanswered, and that person took the intent you bought and handed it to a competitor who paid nothing for it.
Which one to fix first
Fix the phone. Nearly always, and for a straightforward reason: raising the share of calls you answer is cheaper than raising the ad budget, and it makes every dollar of the spend you already have worth more.
Doubling ad spend to compensate for a leaking answer rate means paying twice to lose twice as many. Whereas if you are currently answering, say, two thirds of the calls that come in and you get that closer to all of them, you have increased your booked work without touching the campaign at all.
There is a harder version of this worth saying plainly. If you genuinely cannot answer the calls you are already generating, buying more of them is not growth. Slowing the ads down while you fix the answer rate is not defeat, it is arithmetic.
How to find your own number
This is worth half an hour and you do not need to buy anything to do it.
- Pull thirty days of call logs from your phone system or call tracking. Count answered against missed and voicemail. That percentage is your answer rate and most owners have never seen it.
- Note when the misses cluster. They usually pile up in the same windows: mid-morning, lunch, and after five. Those windows tell you whether you need coverage or just a different habit.
- Take your average job value and a realistic booking rate, then multiply it against the missed calls. It is a rough number and it will still be large enough to be worth acting on.
- Compare it to your monthly ad spend. For a lot of contractors the value sitting in unanswered calls exceeds what they are paying to generate calls in the first place. That comparison usually settles the argument about what to fix first.
The number worth building towards
Cost per click tells you almost nothing. Cost per call is better and still misleading, because a call you did not answer is not a lead. The number that actually runs a contracting business is cost per booked job, by channel.
Getting there means the call data and the job data have to live in the same place, which for most shops they do not. The ads are in one platform, the calls in another, the completed jobs in the field software, and the invoices in the accounting system. Nobody joins them because joining them by hand takes a Sunday evening, so it never happens consistently.
That is the whole reason the leak survives. Not carelessness, but the fact that the evidence is split across four systems that were never introduced to each other.
Once those numbers sit side by side, the arguments get much shorter. You stop debating whether Google Ads works in general and start seeing that one channel books jobs at a third of the cost of another, and that a measurable slice of what you paid for never got answered at all.