Your marketing spend leaks at the phone, not the ad

You paid for that call. Maybe it came from a search ad, a Local Services listing, a directory profile or the magnet on a truck that cost you forty bucks and a Saturday. The homeowner saw it, trusted it enough to dial, and the phone rang in your shop while everyone was on a job. Nobody picked up. Your ad report still counts that as a win. It says “call,” not “call answered.” Most owners who feel like their marketing stopped working are looking at the wrong end of it. The ad did its job. The money leaked out after the phone started ringing.

01The report that stops at the ring

Open whatever dashboard your ads come with and look at what it counts. Depending on the platform, a “call” might be a tap on your number, a call that connected for a few seconds, or a call that lasted past some minimum length. What almost none of them can tell you is whether a person on your end answered, whether they got the address, and whether that call ever turned into a job.

That is not the ad company hiding anything. They only see their side of the wire. Their side ends when your phone starts ringing. Everything after that happens inside your business, and it is yours to measure.

So here is the uncomfortable version. You can have an ad that is working well, bringing in exactly the people you want, at a fair price per call, and still lose money on it every month. The leak is not in the targeting or the headline. It is in the twenty seconds after the ring, and in the week after that.

I see owners respond to a slow month by changing the ad. New photos, new offer, a bigger budget, a different agency. Sometimes that helps. Often it just pours more paid calls into the same hole. If three in ten calls go unanswered today, doubling the budget doubles the calls you never hear about.

02Five places a paid call goes missing

The leak is rarely one thing. When I listen to how a small shop actually handles its phone, the paid calls tend to fall out in a few predictable spots.

The first is the obvious one. The phone rings and nobody gets to it. The tech is under a house, the office person is on the other line, the owner is driving. The caller hears four rings and a greeting, and most of them hang up before the beep. I wrote about how much of this happens in the middle of a normal workday in the daytime missed call nobody talks about.

The second is the call after hours. Ads do not switch off at 5pm unless you tell them to, and plenty of people search for a plumber or an electrician at night. If the line goes to voicemail, you paid full price for a call that had almost no chance of becoming a job.

The third is the call that gets answered badly. Someone grabs it in a hurry, says “can I call you back,” and does not write down the number or what the problem was. The caller hangs up thinking they have been helped. They have not. Nobody calls back, or someone calls back two days later and the customer has already hired the shop that picked up.

The fourth is the price question. A customer calls, asks roughly what it costs, gets a sensible “we’d need to see it,” and says they will think about it. Then nothing. They were not saying no. They were busy, or comparing, or waiting for a nudge that never came.

The fifth is the text. More people text a business number than most owners realize, especially from a listing that shows a mobile number. A text that sits unread for a day is the same as a missed call, except it is easier to forget because it never rang.

Every one of those started with money you spent. The ad worked. The phone did not.

03Do the math with your own numbers

I am not going to tell you what a lead costs in your trade or your town, because I do not have a real number for that and it varies too much to guess. You do have the number. It is on your last ad invoice. Use it.

Here is an example, so you can see the shape of it. These are made up for the arithmetic, not a benchmark.

Say you spend $2,000 a month on ads and they bring in 40 calls. That is $50 a call. Now say your call log shows 12 of those 40 went unanswered, which is 30%. You paid $600 a month for calls that reached nobody, or $7,200 a year.

That is just the spend. The bigger number is the work. If one in three of those 12 callers would have booked if someone had picked up, that is 4 jobs a month. At a $600 average ticket, that is $2,400 a month in jobs, or $28,800 a year. That is revenue, not profit, and your close rate and ticket will be different from mine. Put in your own and see what comes out. For the longer version of this, there is how much money a missed call actually costs a home services business.

To get your missed calls, you do not need a vendor. Pull last month’s call log from your phone provider, line it up against the calls your ad report says it sent, and count the ones with no answer or a very short duration. I walked through the method in how to count the calls your business is missing. Most owners I talk to are surprised by what they find. Some are not, and those are usually the ones who already knew and have been meaning to fix it.

04Why a better ad usually isn’t the fix

When a paid channel looks weak, the instinct is to blame the channel. Sometimes that is right. Some lead sources really do send junk, and you should drop those. But before you fire an ad, check whether it was given a fair chance.

Speed of response matters a lot, and there is research on it, though not from phone calls. Harvard Business Review audited 2,241 US companies and timed how long they took to answer web inquiries. The average first response took 42 hours. Only 37% answered within an hour, and 23% never answered at all. Companies that responded within an hour were about seven times more likely to have a real conversation with a decision maker than those that waited even one more hour. That study was about web forms, not phones, so take it as background rather than a measurement of your line. But a phone caller is usually less patient than someone filling out a form, not more. If typed inquiries go stale that fast, a call that rang out has an even shorter shelf life.

Here is the part that matters for your ad budget. When the leak is at the phone, every improvement you make to the ad gets taxed by it. A better ad that brings 20% more calls still loses the same share of them. A cheaper cost per call still lands the same share on voicemail. You can spend months tuning the top of the funnel and never see the bottom move.

Fix the phone first, and every dollar you were already spending starts working harder. That is the cheapest improvement most shops will ever make to their marketing, because it does not cost a single extra dollar of ad spend.

05What to check before you raise the budget

None of this needs special software. It needs about an hour and a little honesty.

Call your own number from a phone your shop does not recognize, at a time you are normally busy. Mid-morning on a weekday is good. Then do it again at 7pm. Listen to what a paying customer hears. If it is six rings and a voicemail greeting that lists your hours, that is what your ad budget is buying.

Match your ad report against your call log for one month. You want three numbers: calls the ads sent, calls someone actually answered, and calls that turned into a booked job. That last ratio, calls answered to jobs booked, is the one I think every contractor should know. I explained why in the one number every contractor should know about their phone.

Look at the answered calls that went nowhere. If you record calls, listen to a handful. (Tell callers at the start if you record. About a dozen states require everyone on the call to agree, and the Digital Media Law Project keeps a list. Check your own state’s rule.) You are listening for the moments the call stalled. Did anyone take the address? Did anyone say what happens next? Did anyone follow up when the customer said “let me think about it”?

Check your texts. Scroll back a week on whatever number your listings show. Count the messages that got no reply, or a reply the next day.

Then decide. If the leak is small, great, go spend more on ads. If it is large, raising the budget now is like turning up the water pressure on a pipe you already know is cracked.

06Plugging the leak without hiring someone

The traditional fix is a person whose job is the phone. That works, and for some shops it is the right call. It is also not cheap. The Bureau of Labor Statistics put the median receptionist wage at $18.27 an hour in May 2025. Twenty hours a week at that rate is about $1,580 a month in wages, before payroll taxes, and that only covers twenty hours. Nights and weekends, when a lot of ad calls come in, cost more to cover.

The other way is a system that answers when your team can’t. Owners who buy ads have usually been oversold before, so here is what a good one does and what it does not.

It answers the calls your team misses and replies to texts. It takes the caller’s name, address, number and what the problem is, and helps them book the next step. If a customer does not book, it sends approved follow-ups about the next step and stops the moment they reply. That covers the “let me think about it” caller who would otherwise drift off. Anything that needs your judgment goes to the person you named, and the answer goes back to the customer.

What it will not do matters just as much. It does not quote a price. It does not diagnose a problem over the phone. It does not promise scope, coverage or a time slot. Appointments wait for you to confirm them, which I argued for in why an AI should never book a job without confirming it first. And it does not make cold calls. It only talks to people who already called or texted your business.

Your number does not change either. The ad keeps pointing at the same number it always did. If you are wondering how that works, it is covered in what happens to your phone number when you add an answering system.

Frequently asked questions

How do I know if my ads are bad or my phone is the problem?

Compare three numbers for one month: calls the ads sent, calls answered, and jobs booked. If a large share of calls went unanswered, the phone is the first suspect. If almost every call was answered and few turned into jobs, look at the lead source or how the calls are handled once someone picks up.

Is it worth paying for call tracking?

It can help if you run several ad sources and need to know which one sent which call. But you can get most of the answer for free by comparing your phone provider’s call log against your ad reports. Start there before you buy anything.

Should I turn my ads off after hours?

You can, and if nobody can answer at night it will save you money. The tradeoff is that some of the most motivated callers search at night. If you can get those calls answered, they are often worth keeping. If you cannot, pausing ads after hours is a reasonable stopgap.

Will an AI answering my ad calls hurt my reviews?

Not if it is honest and useful. It should say it is an AI if asked, take the details properly and set a true expectation about when someone will call back. What hurts reviews is a customer who called three times and never heard back, and that is the situation it is meant to replace.

Do I need to change the number on my ads?

No. Your business number stays the same. Calls your team cannot get to are forwarded to the system, so your listings, truck wraps and signs do not change.

The bottom line

Most ad budgets are not wasted at the ad. They are wasted in the twenty seconds after the phone rings and in the week after a customer says they will think about it. Before you change agencies or raise the budget, find out how many of the calls you already paid for actually reached a person and got a follow-up. That number tells you where the next dollar should go.

We book appointments for roofing companies: we find homeowners looking for roof repair or replacement, answer their replies and book them into your system. If that would help your team, see how it works at rushdial.com/roofing.