Average cost per click in Google Ads rose 12.88% last year. Average cost per lead went down. Both numbers come from the same dataset, and the gap between them is where most accounts are quietly losing money.
If your Google Ads cost per lead keeps climbing, the auction is rarely the whole story. Clicks did get more expensive. The accounts that held their number steady absorbed that with better conversion rates, cleaner conversion actions, tighter negative lists and bid targets somebody actually revisited. The rest are paying this year’s click prices at last year’s conversion rate. Two changes Google shipped in the last month push it further: a bid strategy change that finished rolling out globally on 27 August 2026, and the AI Max migration running through September.
The auction got more expensive. Leads got cheaper anyway.
WordStream’s 2026 benchmark study, published on 19 May 2026, covered 13,474 US search campaigns running between 1 April 2025 and 31 March 2026. Average CPC came in at $5.42, up from $5.26, a rise of 12.88%. Average cost per lead came in at $66.69, down from $70.11. That is the first drop since 2020. Average conversion rate was 8.18%, and 87% of industries recorded a conversion rate increase.
Read those in the right order. Clicks got roughly 13% more expensive and leads still got cheaper, because conversion rates rose faster than click prices did. So here is a position I will defend: if your cost per lead is up more than about 13% year on year, competition isn’t your explanation. Everything above that line came from inside your account.
Plenty of buyers will push back on that, and they have a case. The average hides a wide vertical spread. Attorneys and legal services sat at $131.63 CPL while automotive repair and parts sat at $29.96, and real estate CPCs jumped 27.27% on their own. If you sit in one of those, 13% is the wrong yardstick. Before I accept “it’s the market” I want to see your auction insights showing a new entrant holding sustained impression share.
That dataset is US search campaigns, and most of the accounts I work on sit in Australia and the UK, where the auctions price differently. I use it for direction and pull each account’s own year-on-year CPC before concluding anything. The full 2026 benchmarks by industry will have your vertical.
Your Google Ads cost per lead is probably not the number on the dashboard
The most common reason a cost per lead “climbs” is that it was never real, and something finally exposed it.
An Australian roofing account I audited reported 318 conversions at roughly $10 each. Once we traced the conversion action, 96% turned out to be taps on the phone-number button on mobile, counted every time, whether or not a call ever connected. Real traceable leads over sixteen months, form fills plus call-extension calls that actually connected, came to twelve. True cost per acquisition was closer to $263.
Twelve.
The reporting error was the smaller half of the problem. Smart Bidding had been trained on that signal for over a year, so it learned that cheap Display taps looked like conversions and bought more of them. The full breakdown sits in the 318-lead mirage. A conversion action that fires on intent instead of outcome does two jobs at once: it misreports your cost per lead, and it teaches the bidding algorithm to buy whatever traffic produces the fake signal.
Check yours. Go to Goals > Conversions > Summary, open each action, and look at two settings. “Include in ‘Conversions'” should be Yes only for actions that represent a real outcome. Under Settings > Count, phone and form actions belong on One rather than Every. (The Count setting is two clicks deep, and on handover accounts I have never once found it set correctly.)
When you fix this, your reported cost per lead goes up. That is the true number arriving, and it’s the one time a climbing figure is good news. Warn the client in advance, because that conversation gets much harder afterwards.
Two things Google changed in the last four weeks
Both are live now, and both push reported cost per lead in the same direction.
Target-based bid strategies stopped overdelivering
Google changed how budget-constrained campaigns optimise toward their targets. The rollout started on 17 August 2026 and completed globally on 27 August 2026. It covers Target CPA and Target ROAS, plus Target CPC in Demand Gen, across Search, Shopping, Performance Max, Demand Gen and Travel campaigns. Google’s FAQ on the target-based bid strategy change has the detail. If the account runs Performance Max alongside Search, check which campaign really earned the lead before you trust the CPA comparison between them.
Before this, a campaign limited by budget could quietly overperform its target. You set Target CPA at $120, the budget cap kept it in the cheaper slice of the auction, and it delivered at $70. Those campaigns now optimise consistently toward the number you set, so the $70 drifts up toward $120 and nothing in your account has broken.
(Yes, this means the number you typed in eighteen months ago and forgot about is now the number you get. Arguably that is how targets were always meant to behave. It still lands on your report as a cost per lead increase.)
Google will not adjust your targets for you.
The Target Adjustment Tool opens from the “Review your campaign targets” banner, or from the Campaigns page via Settings icon > Bidding > “Review campaigns”. Set each target against what the campaign actually delivered over the last 30 days, then wait one to two conversion cycles before judging.
AI Max is being switched on for you during September
Campaign-level broad match settings and standalone automatically created assets are being auto-migrated to AI Max between 1 and 30 September 2026. Dynamic Search Ads follow in February 2027. Google blocked creation of the legacy structures everywhere on 3 August 2026, interface and API included, so you’re being migrated whether you planned for it or not.
What this does to your numbers depends on how tightly your account was scoped. AI Max search term matching uses broad match plus keywordless technology to find queries you never targeted, and it’s on by default at ad group level. On an account with a maintained negative list, the extra reach is usually fine. On an account held together by exact match and hope, it will find queries you’d never have bid on, and your cost per lead moves within days.
Two controls matter before September ends. Search term matching toggles at ad group level. Brand exclusions exist only at campaign level while inclusions work at both, which catches people out. The search terms report now carries an “AI Max” match type and a Source column showing whether a match came from broad match expansion or keywordless matching, and Google documents how the matching works. This one costs people real money and almost nobody checks it. Read it weekly this month.
The drift nobody watches
None of the above explains the account where your Google Ads cost per lead climbs 4% a month for a year with no single change behind it. That one is almost always conversion action creep and lead quality decay running together. Somebody adds a newsletter signup to Conversions because it looked good in a quarterly report. Six months later a “view contact page” action goes in alongside it. Reported CPL drops, everyone relaxes, and Smart Bidding starts optimising toward the cheapest of those actions, which is never the phone call.
Lead quality decay is harder to spot because it doesn’t appear in Google Ads at all. Your CPL can sit flat at $58 while your cost per qualified lead doubles underneath it. The only reliable fix is pushing a qualification status back from your CRM, or failing that, asking whoever answers the phone. I have written more about where wasted spend actually hides in a Google Ads account, and the same reports catch both problems.
How to find your real cost per lead in about forty minutes
- Pull the last 90 days segmented by conversion action. Goals > Conversions > Summary, then note which actions have “Include in ‘Conversions'” set to Yes.
- Open each phone and form action and check Settings > Count. Anything on Every that is not an ecommerce purchase is inflating your numbers.
- Count the actual enquiries over the same 90 days from your CRM and your call log. Divide total Google Ads spend by that count. That is your real cost per lead.
- Compare the two figures. A gap under about 1.3x is normal attribution slack. A gap over 2x means you stop optimising and fix tracking first.
- Check when each bid strategy target was last edited. Campaigns page > Settings icon > Bidding. Anything untouched since before August 2026 needs a fresh look after the rollout above.
- Open the search terms report, add the Source column, and filter for AI Max matches. Anything irrelevant goes straight into negatives.
Steps one to four are where the money usually sits. If you’d rather not do it yourself, this is the first hour of any audit a Google Ads consultant should run before touching a single bid.
Where this diagnosis falls apart
The forty-minute version doesn’t scale down cleanly. There are four situations where I would ignore most of it.
- Low volume. Under roughly 15 conversions a month, 90 days gives you about 45 data points and your CPL swings on noise. You’ll find a “trend” that was three bad weeks.
- Seasonality. A trades account in Australia in September does not compare to the same account in March. Look at the same month last year.
- Timing coincidence. If your CPL shifted on 4 August, the bid strategy rollout that started on 17 August did not cause it. Line the shift up against the rollout window.
- The reporting shock. Fixing tracking makes CPL look worse immediately. Show a client that chart without warning and you’ll spend a month defending a fix that helped them.
Check the date before you blame the change.
What I still do not have a clean answer for
Accounts under about $2,000 a month. At that spend, 90 days might produce twenty conversions, and the difference between a genuine cost per lead problem and a run of bad weeks is not visible in the data. I have tried stretching the window to 180 days, which smooths the noise and buries the seasonality at the same time. I don’t know which trade-off is right yet, and I would rather say so than pretend the audit above works at every budget. If you run small accounts and you’ve found a rule that holds up, I’d like to hear it.
Before you touch a bid, find out whether the Google Ads cost per lead on your dashboard is the one your business actually pays. Our Google Ads management is senior-led, month to month, no lock-in and no reporting theatre, and you can book a free 30-minute audit.
Since September, Google has been auto-upgrading Dynamic Search Ads to AI Max, which changes how queries, ad copy and landing pages are chosen. If your cost per lead moved this month, start with what the AI Max upgrade changed and what to check.
Researched and drafted with AI assistance, then reviewed, fact-checked and edited by Rajat Verma before publishing. All account examples are from real campaigns managed by Turquoise Digital and are anonymised.