Diagnosis
Why are my Google Ads not converting?
Traffic arriving and nothing happening is the most common complaint in paid search, and it is almost never one mysterious problem. It is a short list of specific faults, and they have to be checked in a particular order.
Why are my Google Ads getting clicks but no conversions?
Ads that get clicks but no enquiries almost always fail at one of five points: conversion tracking is missing or misconfigured so the conversions that do happen are never recorded, the search terms triggering the ads are not the searches you wanted, the landing page does not continue the promise the ad made, the offer asks for too much too early, or enquiries arrive and nobody answers them quickly enough.
Check them in that order, because a tracking fault makes every other diagnosis unreliable. Start in the search terms report, which shows what people actually typed rather than the keywords you added.
First decide which kind of failure this is
There are only two possibilities and they lead to completely different work. Either conversions are happening and you cannot see them, or they genuinely are not happening.
Those look identical in a report and nothing else you do is reliable until you know which one you have. A practice that took eleven phone calls from ads last month and recorded none of them does not have a conversion problem. It has a measurement problem, and every optimisation made on top of that blind spot will make performance worse, because automated bidding is being trained on an incomplete picture.
Read the full breakdown: First decide which kind of failure this isHide the full breakdown: First decide which kind of failure this is
The quickest separation test is to compare the ad account against reality. Ask whoever answers the phone whether call volume changed when the campaign started. Check whether form submissions land somewhere a human reads. If the business felt something and the account shows nothing, stop and fix the plumbing.
If the business genuinely felt nothing either, then the failure is real and the rest of this page is the order to work through it.
One more piece of framing before the list: judge this over a meaningful window. A campaign with a handful of clicks a day cannot produce a stable conversion rate inside a fortnight, and reacting to a quiet week is how accounts get churned into a worse state than they started.
Work through it in this order
The order matters more than the individual checks. Each step assumes the one before it came back clean, and doing them out of sequence produces confident wrong answers.
Prove that a real enquiry would be recorded
Submit your own form and place your own call from a phone that has clicked an ad, then look for both in the conversions column. Google Ads can record website actions, calls from the ad, calls from a number on the site, app activity and conversions imported from a CRM. Each has to be configured separately, and a business with one form tracked and no call tracking is typically blind to most of its enquiries.
Check what is marked primary and what is marked secondary
Conversion actions configured as primary are what bidding optimises toward. Secondary actions are recorded for observation only. If page views or newsletter signups are sitting in the primary column, the system is faithfully buying you more of those instead of booked appointments, and the account will look busy and unproductive.
Read the search terms report line by line
This is the highest yield hour in paid search. The report shows the searches that actually triggered your ads, which is never the same list as the keywords you added. Expect to find jobs you do not take, towns you do not serve, salary and jobs queries, DIY how to queries and competitor names. Each one becomes a negative keyword. Google withholds very low volume terms for privacy, so the report is thorough rather than complete.
- Filter to terms with spend and no conversions over a long enough window to be meaningful.
- Add negatives at the right level: campaign wide for things you never do, ad group level for things that belong in a different ad group.
- Re-read it weekly at first. Broad match keeps finding new ways to spend money that nobody predicted.
See the remaining steps: Work through it in this orderHide the remaining steps: Work through it in this order
Tighten match types and check location settings
Broad match draws on recent search activity, your landing page and the other keywords in the ad group, which makes it useful once conversion data exists and expensive before. Move the money to phrase and exact while you are diagnosing. At the same time check the location options, because serving to people merely showing interest in your area is very different from serving to people actually in it.
Open the landing page on a phone and try to convert
Do it as a stranger would, on mobile data rather than office wifi. Time how long it takes to find the phone number. Count the form fields. If the ad promised same day service and the page leads with a history of the practice, you have found the problem without needing any data at all.
If all five come back clean and the account still produces nothing, the honest conclusion may be that the searches you are buying are not commercially valuable, which is a channel question rather than a settings question.
Why tracking has to come first
Every recommendation that follows depends on conversion data being real. When it is not, the account does not fail loudly, it quietly optimises toward the wrong thing.
Automated bidding learns from recorded conversions. If half of them are invisible, the system concludes that the campaigns producing those invisible conversions are weak, and moves budget away from exactly the work that was succeeding. That is worse than no bidding strategy at all, because it is confidently wrong.
For most local businesses the missing piece is calls. A person taps the number in the ad or on the page, the phone rings at reception, and nothing in the system connects that call back to the search that started it. Until that link exists, every report is a partial account of what the money bought.
Duplicate counting causes the opposite error. If a thank you page fires a conversion on every refresh and the counting setting is every rather than one, a single enquiry can be recorded several times and the account looks far healthier than it is. Check the count setting on each conversion action, not just that the action exists.
The search terms report is where the money is
Keywords are what you told Google you want. Search terms are what people typed. Nobody outside advertising knows there is a difference, and it is the difference that empties budgets.
The most expensive category is not obviously irrelevant terms, it is plausible ones. A cosmetic dentist bidding on veneers will collect searches for veneer cost, veneer before and after, veneers gone wrong, veneers abroad and dental veneer technician jobs. Every one of those is topically correct and commercially useless, and a broad match keyword will happily buy all of them.
Read the full breakdown: The search terms report is where the money isHide the full breakdown: The search terms report is where the money is
There is a second pattern worth naming: research stage searches. Somebody typing how much do implants cost is real demand at the wrong moment. If they convert at all, they convert on a page that answers the question rather than on a booking form, which is a content decision rather than a bidding one.
Negative keyword lists are the cumulative asset here. A list built over a year across a category is one of the genuinely valuable things inside an ad account, and one of the main reasons the account should belong to the advertiser rather than to whoever happens to be managing it this year.
Set a recurring time for this. Weekly at the start, then monthly once the account has settled. It is unglamorous and it consistently outperforms bid adjustments.
Where the drop actually happens
Impressions, clicks, landings and enquiries each have their own drop off. Locating which one is abnormal tells you what to fix, and stops you rewriting ads when the page is at fault.
If the click-through rate is healthy and the conversion rate is poor, the ad is doing its job and the page or the offer is not. Rewriting headlines will not help, and may make it worse by attracting even more of the same traffic.
If the click-through rate is poor, the ad is not matching the search, and the fix is copy and ad group structure rather than anything on the site.
If both look reasonable and the phone still does not ring in a way anybody notices, look past the website entirely at what happens to an enquiry once it exists.
This is also why one number can never diagnose an account. Cost per lead is an outcome. The stage rates are the diagnosis.
What a normal shape looks like
LocalIQ's Search Advertising Benchmarks 2026 aggregate campaigns run by their own customers across Google Ads and Microsoft Ads. Use these to spot which stage is abnormal, not as targets, and not as a market average.
SourceLocalIQ, Search Advertising Benchmarks 2026
Compare shapes rather than values. A category with expensive clicks and a strong conversion rate is a healthy account. A category with cheap clicks and almost no conversions is usually buying the wrong searches cheaply, which is the most common way an account looks affordable and produces nothing.
Signs this is not an ads problem at all
Some accounts are working correctly and the failure is happening after the click. These are the patterns worth ruling in before anybody touches a bid.
- Enquiries come in outside office hours and get a reply the next working day, by which point the person has called somebody else.
- The form asks for date of birth, insurance details or a long message before anybody has spoken to the person.
- The phone is answered by voicemail during advertised hours, or by somebody who cannot book an appointment.
- Nobody has agreed what a qualified lead is, so sales reports the leads as junk and marketing reports the leads as delivered.
- The ad promises something the business cannot actually do this week, such as same day appointments when the diary is full for a fortnight.
- Enquiries are counted in one system and sales in another, and nobody has ever joined the two, so the return on the channel is unknown rather than poor.
- The offer is right but the price point requires a conversation, and the page offers only a booking button and no way to ask a question.
Speed of response is the one most often underrated. For enquiries generated by an urgent search, the business that answers first frequently wins regardless of who was cheaper or better.
When the honest answer is that the channel is wrong
Sometimes everything above checks out and search ads still do not work, and the reason is that the demand being harvested is not the demand you needed. This is worth saying because very few pages on this subject will say it.
It happens most often with genuinely new or elective services that nobody searches for by name, with very small service areas where the monthly search volume cannot support a campaign, and with categories where the visible searches are dominated by price shoppers while your business competes on something else entirely.
Read the full breakdown: When the honest answer is that the channel is wrongHide the full breakdown: When the honest answer is that the channel is wrong
In those cases the money usually does better creating demand rather than harvesting it, or building organic visibility for the questions people ask before they are ready to buy. Neither is faster. Both can be more durable.
The test is simple enough to run before spending more: does the search volume for the specific services you sell, in the specific area you serve, support the number of enquiries you need at roughly the cost per lead your category runs at? If the arithmetic does not close, no amount of account management will make it close.
Where to look next
Each of these covers one of the stages above in more depth than a diagnostic page can.
Tracking where leads come from
Getting calls, forms and chats into one place, attributed back to the channel that produced them.
Tracking where leads come fromWhat counts as a good conversion rate
Why the number depends entirely on traffic source and what you are asking people to do.
What counts as a good conversion rateWhat a landing page is for
Why paid traffic converts better on a dedicated page than on a home page.
What a landing page is forConversion rate optimisation explained
How to improve the page systematically rather than by opinion.
Conversion rate optimisation explainedThe full Google Ads guide
Structure, match types, bidding and reporting in one place, including the self diagnosis for whether the channel fits at all.
The full Google Ads guideWant somebody to run this diagnostic on your account?
We will check the tracking, read the search terms and open the landing pages, then tell you which of the five it actually is.
Straight answers.
How do I know if my conversion tracking is set up correctly?
Submit a test form and make a test call from a device that has clicked one of your ads, then look for both in the conversions column within a day or so.
Also check the counting setting and whether each action is primary or secondary. A conversion action that exists but is set to secondary is recorded for observation and is not what bidding chases.
Where do I find the searches that actually triggered my ads?
The search terms report inside Google Ads. It shows the terms people typed that led to an impression, which is different from the keyword list you built.
Google omits terms with very low query activity for privacy reasons, so treat it as thorough rather than exhaustive.
Should I switch everything to exact match?
It is a reasonable move while you are diagnosing, because it removes the variable of unexpected search terms. It is usually too restrictive as a permanent state, since it limits how much volume the account can reach.
A common settled structure is phrase and exact carrying the core spend, with broad match used deliberately once there is enough conversion data for bidding to steer it.
My cost per lead looks fine but sales say the leads are junk. What now?
Agree a definition of a qualified lead first, then get that definition back into the ad account. Importing qualified leads from a CRM as conversions lets bidding chase the leads you actually want rather than every form submission.
Until that loop exists, the two teams are measuring different things and both are correct.
How long should I wait before deciding the ads do not work?
Long enough to accumulate a meaningful number of conversions rather than a set number of days. In a low volume category that can be a couple of months.
Conversion tracking should be in place from day one, though. Waiting to measure is what turns a two month test into a two month opinion.
Can bad landing page speed stop ads converting?
Yes, and it does double damage. A slow page loses impatient visitors, and landing page experience is one of the three components Google reports in Quality Score, so the click can cost more as well.
Test on a phone on mobile data rather than office wifi, which is where the difference shows up.
Where this comes from.
Primary documentation and published research behind the guidance on this page.
- Google Ads Help: About conversion measurement (opens in a new tab)
- Google Ads Help: About the search terms report (opens in a new tab)
- Google Ads Help: Negative keyword definition (opens in a new tab)
- Google Ads Help: About keyword matching options (opens in a new tab)
- Google Ads Help: 5 ways to use Quality Score to improve performance (opens in a new tab)
Talk to the team
A short call, a look at how the business currently shows up, and a straight answer on what we would do first.