Your Google Ads get clicks.
So why doesn't the phone ring?

Every account we take over has the same chart. Cost per click holding steady, impressions healthy, click-through rate perfectly respectable, and a phone that rings maybe half as often as the math says it should. The owner assumes the campaign is broken somewhere inside Google's machinery. It almost never is. The leak is nearly always in one of three places between the click and the call, and all three are things you can check yourself in under an hour. Here is the order to check them in, the same order we use when a client calls us convinced Google Ads "just doesn't work" for their business.

The Line Is UnpluggedClick to Call

Start here: are the
clicks even real?

Before you blame tracking, your landing page, or your keywords, confirm that the clicks you are paying for are the clicks you think you are buying. Open the Search Terms report (Campaigns, then Insights and Reports, then Search Terms, or filter straight from the Keywords tab) and read what people actually typed, not just the keyword you bid on. This single report is the fastest ten minutes you will spend on this problem, and most owners have never opened it.

Broad match is the usual culprit. Bid on "brake repair" in broad match without a real negative keyword list, and Google will happily show your ad to someone searching "how to replace brake pads yourself" or "brake repair cost calculator." Both are clicks. Neither is a customer calling your shop. You paid the same three to eight dollars for the click either way, and the only way to see the difference is to read the actual search terms sitting underneath your keyword, not the keyword itself.

Check device and time of day too. A cluster of clicks from mobile at 2 a.m., or a spike from a single zip code far outside your service radius, is worth a second look before you conclude the ad itself is the problem. We are not talking about outright click fraud here, which is rarer than forum threads make it sound. We are talking about ordinary broad match doing exactly what it was built to do: find more clicks, not necessarily better ones.

Problem one: your tracking
is lying to you.

This is the layer almost nobody checks, and it is where we find the actual answer more often than any other. Google's Smart Bidding optimizes toward whatever conversion signal you hand it. If calls are invisible to that signal, you are not running a calls campaign. You are running a clicks campaign and hoping the rest sorts itself out.

The most common failure is the simplest one: the same static phone number sits on your website, your Google Business Profile, and every paid landing page. Without a dedicated tracking number tied specifically to the ad, Google Ads has no way to attribute an inbound call back to the click that produced it, so it never counts as a conversion no matter how many times the phone rings. Dynamic number insertion solves this: every paid visitor sees a unique tracking number that forwards to your real line, and that call gets logged against the exact click, keyword, and ad that generated it.

Second, open Goals, then Conversions inside Google Ads and confirm a "Calls from ads" conversion action actually exists and is marked primary. We regularly find accounts where this conversion action was either never created, created but never set as primary so Smart Bidding ignores it, or set with a minimum call length so low that every accidental dial and instant hang-up counts as a "lead," which quietly convinces the algorithm that junk clicks are working while real calls get buried in the noise. A minimum length of 60 to 90 seconds filters most of that static out.

An algorithm cannot optimize toward a conversion it cannot see. Every layer below this one only matters once the platform can actually count what you are asking it to find.

At Network Collision, our Gilbert, Arizona collision client, we built the account around call tracking from the first week live, because insurance-driven collision leads live or die by the phone. A missed or uncounted call to a body shop usually goes straight to the next name on the claims adjuster's list within the same conversation, so the cost of blind tracking there is not a vague inefficiency, it is a lost car sitting in someone else's bay. The same math applies at a smaller dollar amount to every service business running Search ads toward a phone number.

Run the math on what blind tracking actually costs. A shop spending $2,000 a month at a blended $6 cost per click buys roughly 330 clicks. If even a third of those clicks would have called with accurate tracking and a clean handoff, that is close to 100 calls Smart Bidding never got credit for, which means it spent the next month's budget as if those clicks had done nothing. The algorithm is not guessing badly. It is working perfectly against a number that was wrong from the start.

If your "Calls from ads" conversion count is at or near zero while your real phone log shows calls coming in, stop reading past this section until it is fixed. Everything below is invisible to the bidding algorithm until tracking tells it the truth.

Problem two: the landing page
kills the call before it happens.

Lost Before It FillsThe Handoff

A huge share of underperforming accounts send every ad, regardless of service, to the same generic homepage. Someone searches "transmission repair near me," clicks an ad promising exactly that, and lands on a homepage listing twelve different services with the phone number buried in a header nobody reads on a six inch screen. You already paid for the click. You lose the conversion entirely on the handoff between the promise and the page.

A landing experience built to produce calls needs a short list of things, and most pages are missing at least two of them. The phone number visible above the fold, large enough to actually read on mobile, with a tap to call button rather than plain text. A headline that matches the ad's specific promise, service plus location, not a generic tagline. One trust signal near the number, a review count and star rating, current hours, or an "open now" indicator, because a stranger deciding whether to dial wants a reason to believe the call will go somewhere. And a page that loads fast, since we typically see a meaningful share of mobile visitors abandon a page that takes longer than three seconds to render, and a caller who already hit back never sees the number at all.

If every service line in your account points to the same homepage, you are also destroying your own data. A dozen different search intents are getting flattened into one page's numbers, which means you cannot tell whether brake repair searches convert better than transmission searches, because they never had separate pages to convert on in the first place. Build or adapt one service specific page per major category before you spend another dollar sending traffic to a page that was never built to answer the question the ad just asked.

Problem three: the wrong
searches are winning.

Finding the Right OneKeyword Intent

If tracking is clean and the landing page matches the ad, and calls are still thin, the keywords themselves are usually winning the wrong searches. Not every search containing your service name is a buying search. "Brake repair," unqualified, catches a shop owner looking for a quote, a parent looking up how to teach their teenager to change pads, and someone researching a completely different problem that happens to share a word. Broad match without a disciplined negative keyword list treats all three the same, and Google's job under broad match is to find volume, not to guess which searcher is closest to picking up the phone.

The highest intent tier is specific and transactional: "[service] near me," "[service] cost," "[service] today," or a named part plus "repair" or "replacement." These are people who already know what is wrong and are deciding who to call. The lowest intent tier is informational: "how to," "DIY," "symptoms of," "is it safe to drive with." Those searchers are doing research, not booking an appointment, and clicks from that tier cost you the same three to eight dollars as a transactional click while converting at a fraction of the rate.

Broad match without a real negative keyword list is not targeting. It is an open invitation for Google to guess, and Google's guess optimizes for volume, not for your bay count.

Build the negative list from what you actually find in the Search Terms report, not from a generic template. Common candidates across almost every local service account include "jobs," "salary," "how to," "DIY," "reviews," "complaints," "free," and your own brand name if you are running a separate branded campaign that would otherwise cannibalize itself. Review that list weekly for the first month on any new or restructured campaign, then monthly once the account stabilizes.

Keyword intent tiers matter just as much as the negative list. We generally split campaigns into three tiers: emergency and near me searches at the top, since those convert fastest and can carry the highest bids; named service and part searches in the middle, since those are still transactional but less urgent; and informational or comparison searches at the bottom, kept on a tight budget or excluded entirely depending on how aggressive the account needs to be. Treating all three tiers the same budget and bid strategy is one of the fastest ways to burn spend on clicks that were never going to call in the first place.

The fix,
in the right order.

Once you know which layer is leaking, fix them in the same order you diagnosed them. Fixing the landing page before tracking is accurate just means you get a better page with no visibility into whether it actually worked.

1. Fix tracking first. Put a dedicated dynamic tracking number on every paid landing page, confirm a "Calls from ads" conversion action exists, is marked primary, and has a 60 to 90 second minimum call length so real conversations count and accidental dials do not.
2. Fix the landing page second. Build or select one page per major service category with the phone number and a tap to call button above the fold, a headline that matches the ad's promise, and one trust signal near the number.
3. Fix the keywords third. Pull the Search Terms report weekly for the first month, add negatives immediately, and shift spend toward exact and phrase match on the highest intent terms. Pause or restructure broad match terms that are not producing calls after a real sample, generally 50 to 100 clicks per term.
4. Only then touch budget. Once tracking, the landing page, and keyword intent are clean, more budget buys more of something that is actually working. Before that, raising budget just buys more of the same leak, faster.

This is also where the old playbook, print ads, radio spots, coupon mailers, direct mail, simply does not apply. None of it can be diagnosed this way at all, because none of it is trackable enough to tell you which layer is broken. Skip it entirely and keep every dollar somewhere you can actually measure and fix.

When it's not Google's
fault at all.

Sometimes every layer above is clean. Tracking is accurate, the landing page converts, the keywords are tight, and the phone still is not producing the outcome you want. At that point the problem has moved past Google Ads entirely, and it usually lands in one of two places.

The first is calls that genuinely ring but nobody answers, or they roll to voicemail during business hours because the front desk is already on another line. That is not a Google Ads problem, it is a missed call problem, and it needs its own fix: text back automation, call routing, or a simple answering system, not a keyword change. Our breakdown on missed call recovery for local service businesses covers that exact fix, and it picks up precisely where this article leaves off, after the phone has already started ringing.

The second is calls that get answered but never turn into a booked appointment. The conversation itself has its own failure points, a slow quote, no urgency, a caller put on hold too long, and those are conversation problems, not advertising problems. Our piece on diagnosing why leads aren't converting into booked appointments walks through that gap in detail, and it is a genuinely different diagnostic than the one in this article.

If you are running Meta Ads alongside Google and seeing a similar pattern there, clicks without action, the diagnostic order is nearly identical. The layers are the same shape, offer, tracking, creative, and budget, just with audience and creative standing in for keyword intent. Our guide on why your Facebook ads aren't actually working walks through that version of the same process. And if you want the deeper mechanics of structuring a Google Ads account that produces calls instead of just clicks, our Google Ads management page covers how we build campaigns, tracking, and landing pages together from the start.

Frequently asked
questions.

My cost per click looks fine, so why aren't I getting calls?

Cost per click measures traffic, not conversion. A healthy CPC with no calls means the leak is downstream of the click, almost always in tracking, the landing page, or the search terms actually triggering your ad, not in how much you're paying Google for the click itself.

How do I know if my call tracking is actually working?

Pull your real phone log for the last seven days, count inbound calls during business hours, then compare that number to the Calls from Ads conversion count in Google Ads for the same week. If the gap is wide, your tracking is blind to calls that are genuinely happening, and Smart Bidding is optimizing without the one signal that matters most.

Should I pause my Google Ads campaign if it isn't producing calls?

Not yet. Pausing resets the learning data the algorithm has already built and you lose it. Diagnose tracking, landing page, and keyword intent first. Most accounts we take over had a fixable leak, not a broken channel, and pausing would have thrown away weeks of usable data for nothing.

How fast will fixing this actually get me more calls?

Faster than people expect, because the clicks you need were already arriving. You're not waiting on Google to find a new audience. Once a tracking number is live and the landing page matches the ad, you'll typically see the real call volume show up within 24 to 48 hours, because you stopped losing the clicks you were already paying for.

Can broad match keywords ever work for a calls-focused campaign?

Sometimes, but only with a disciplined negative keyword list reviewed weekly and a Smart Bidding strategy set to optimize for calls specifically, not clicks or generic conversions. Without both of those in place, broad match tends to buy volume at the expense of intent, which is the opposite of what a calls campaign needs.

Mitchell Wolfert Founder, Provision Digital. Running paid media, SEO, and funnels for local and national brands from Queen Creek, Arizona.