Your Google Ads Account Isn't Underperforming. It's Lying.
Before you change a single bid, raise your budget, or switch campaign types, check whether your conversion tracking is telling the truth. A "conversion" is just the action you want a visitor to take, a phone call, a filled-out form, a booked appointment, and Google counts those actions to decide where to spend your money. When that counting is wrong, your account is not underperforming. It is lying to you, and every optimization you make on top of bad data just teaches Google to spend faster in the wrong direction.
There are two ways the lie shows up, and both are cheap to fix yourself. The first is phantom conversions: counting "Every" action instead of "One," thank-you-page reloads, and tags that fire twice, all of which make a bad campaign look like a winner. The second is invisible conversions: phone calls and closed deals that never make it back into Google, which make your best campaigns look like losers and get them shut off. Fix those two things first. Everything else you do to the account depends on them being right.
Bad tracking quietly burns your budget at the industry's going rate
While the data lies, your money keeps moving. According to WordStream by LocaliQ's 2025 Google Ads Benchmarks report, compiled by Susie Marino, the average Google Ads conversion rate in 2025 was 7.52% and the average cost per lead was $70.11. So at typical performance, roughly one in thirteen clicks turns into a lead, and each of those leads costs you about seventy dollars. If your tracking is inventing conversions that never happened, Google thinks leads are cheaper than seventy dollars, so it pours more money into the campaign producing those fake wins, and your real cost per lead climbs while the dashboard looks great.
The damage scales with your industry. The same WordStream report found cost per lead ranging from roughly $28.50 in auto repair to roughly $131.63 for attorneys and legal services. For a law firm, every lead Google misjudges is a hundred-and-thirty-dollar decision, so a tracking error that hides which campaigns actually produce signed clients is not a rounding mistake, it is real money walking out the door every day. A roofer paying closer to the low end still feels it, because the cheaper the lead, the higher the volume, and small per-lead errors multiply across hundreds of clicks a month.
Think about a roofing company running a single search campaign. The thank-you page that visitors land on after submitting a quote request gets bookmarked, refreshed, and reopened. Every reload fires the conversion tag again. By the end of the month Google reports 80 conversions on 40 real form fills. The reported cost per lead looks like half of reality, so the owner raises the budget, and now twice the money chases a number that was never true. The account did not improve. The lie just got more expensive.
Owners trust the dashboard because the dashboard never says "I don't know"
The reason this goes unnoticed is psychological, not technical. A reported number feels like a fact. When Google Ads shows 80 conversions, it does not show error bars, it does not flag a thank-you page that reloaded, and it does not whisper that half those calls were existing customers checking their appointment time. The interface presents confidence, and a busy owner reading on a phone between jobs takes that confidence at face value.
The default settings work against you too. For website conversion actions, Google sets the counting method to "Every" by default, meaning it counts every time the tag fires, even if the same person triggers it five times. That default is correct for a store selling five separate products to one buyer. It is wrong for a lead-generation business, where one person filling out one form is one lead no matter how many times the confirmation page loads. Most owners never open that setting, so they inherit a default built for a different kind of business.
The second blind spot is the part of your business Google cannot see. A "tag" is a small snippet of code on your website that tells Google an action happened, and tags only fire inside the browser. When a customer calls and books over the phone, or when a lead you captured last month finally signs a contract this month, none of that happens in the browser. If you do not deliberately send those events back to Google, they are invisible. Your phone might be your best salesperson and your CRM might be full of closed deals, and Google's automated bidding, which is only as good as the data you feed it, will never know any of it happened. That is exactly the budget risk that comes with handing your bidding over to Google's automation before the inputs are clean.
Google's own guidance and the practitioners agree on the foundation
This is not a fringe opinion. Google Ads Help, in its documentation on conversion counting options, recommends using "One" counting for lead-generation actions and "Every" for sales, on the logic that for a lead you care whether it happened at all, not how many times the page fired. The same documentation confirms "Every" is the default for website conversion actions. So the company that built the platform is telling you the default is wrong for your business model, and almost nobody changes it.
Practitioners say it more bluntly. Miles McNair of PPC Mastery put it plainly in January 2025:
"Without proper conversion tracking, you will never get your desired results. It's the absolute foundation for Google Ads success."
Miles McNair, PPC Mastery, January 23, 2025
He goes further on what fixing it is worth, and this next line is his claim rather than a measured statistic, so treat it as a practitioner's confidence rather than a number you can audit:
"If you implement the 3 tracking techniques we have for you in this section, you will already be ahead of 90% of your competitors."
Miles McNair, PPC Mastery, January 23, 2025
On the specific failure of inflated numbers, the team at Teach Traffic describes the mechanism directly: double-counting happens when users reload the thank-you page or a tag fires more than once, which inflates your conversions and your return on ad spend, and the fixes are to switch lead actions to "One" counting and audit how the tags are set up. That is the same problem the roofing example showed, named by people who clean it up for a living.
There is one piece of recent news worth understanding so it does not scare you. As of June 15, 2026, Google blocked new offline-conversion imports through its older, legacy Google Ads API and pushed users toward the newer Data Manager API, a change reported on the Google Ads Developer Blog in May 2026 and corroborated by PPC Land and GroAS in June 2026. The measurement world is tightening, and that is the larger trend. But this mainly affects developers, tools, and agencies that upload data through the API. If you are a DIY owner working inside the Google Ads interface, this is not an emergency for you. It is context. Separately, in April 2026, Google unified its Enhanced Conversions, the feature that improves match accuracy by securely sending hashed customer details like an email address, into a single setting that accepts data from website tags, the Data Manager, and the API all at once, which makes the in-platform setup simpler than it used to be.
The fix is four steps, and you can start three of them today
None of this requires a developer. Open your Google Ads account, click the tools icon, and go to the conversions section under measurement. Then work through these in order.
Step one: switch every lead action to count "One." Open each conversion action that represents a lead, a form submission, a contact request, a quote, and find the counting setting. If it says "Every," change it to "One." This single change stops thank-you-page reloads and repeat visits from inflating your numbers. Do it for lead actions only. If you genuinely sell multiple separate orders to the same customer online, leave those sales actions on "Every," which is what Google recommends.
Step two: confirm phone calls and form fills are tracked as conversions, not just clicks. A "call" that only counts when someone taps your phone number is a click, not a conversion, because it does not tell you whether anyone actually connected or talked. In your conversion actions list, look for call tracking and form tracking that record the completed action. If your leads come by phone and the only thing you are counting is website form submissions, you are optimizing toward a fraction of your real business.
Step three: turn on Enhanced Conversions if it is available to you. Since Google unified this into one setting in April 2026, enabling it is far simpler than it once was. It improves how accurately Google matches the conversions you are already getting, which means cleaner data feeding every decision. This is a toggle, not a project.
Step four: get your offline wins back into Google. An "offline conversion" is a sale that closes outside the browser, the phone deal, the contract signed two weeks after the form fill. This is the single biggest accuracy upgrade available to most lead-generation businesses, because it tells Google which clicks became actual paying customers, not just which clicks became inquiries. This step often needs your tracking person or agency, especially given the June 2026 API change. So if you do not do it yourself, ask them this exact question: "Are we importing offline conversions, meaning our closed deals, back into Google Ads, and are we set up on the Data Manager API now that the legacy import was blocked on June 15?" If the answer is no or a blank stare, you have found the gap that is starving your account of its most valuable data.
Once tracking is honest, the rest of your optimization work finally has solid ground to stand on, which is the right moment to run a proper audit of where your spend is being wasted.
Watch the right numbers, and distrust the comforting ones
After you make these changes, give the account two to three weeks before you judge anything, because conversion data needs time to accumulate and the early days after a tracking change will look noisy. Then compare your reported conversions against something you can count in the real world: the leads in your inbox, the calls in your phone log, the deals in your CRM. If Google says 40 leads and your inbox says 40 leads, your tracking is telling the truth. If Google says 80 and your inbox says 40, you just caught the lie in the act.
The number to trust most is cost per lead measured against your own benchmark, not the industry average. The $70.11 figure from WordStream is a reference point, not a target, because an auto repair shop at $28.50 and a law firm at $131.63 live in different worlds. What matters is whether your cost per lead is stable and believable once the phantom conversions are gone. Expect it to look worse at first. That is not your account getting worse, that is the inflation draining out and the real number surfacing.
Be suspicious of the comforting metrics. Click-through rate, impressions, and raw conversion count are the vanity metrics that feel like progress while telling you almost nothing about money. A campaign with a gorgeous click-through rate and a thank-you page that double-fires will show a beautiful conversion count and quietly lose you money. The metric that survives scrutiny is cost per real, verified lead, and then, once offline conversions are flowing, cost per closed deal. Everything else is decoration. The trap is that the decoration is exactly what the dashboard shows you first.
One more habit worth keeping: re-check the "One" versus "Every" setting after anyone touches your website or your tags, because a site redesign or a new plugin can silently reset tracking, and the dashboard will keep reporting confident numbers the whole time. Tracking is not a thing you fix once. It is a thing you verify, the same way you would reconcile a bank statement.
Frequently Asked Questions
What is the difference between "One" and "Every" conversion counting?
"Every" counts a conversion every single time the tag fires, so one person who reloads the thank-you page three times gets counted as three conversions. "One" counts only the first conversion per ad click, so that same person counts once. Google Ads Help recommends "One" for lead-generation actions, because you care whether a lead happened, not how many times the page loaded, and "Every" for sales where each purchase is genuinely separate. "Every" is the default for website conversion actions, which is why so many lead-gen accounts are silently inflated.
Why do my Google Ads conversions not match the leads in my inbox?
Almost always it is one of two problems. Either you are counting "Every" instead of "One," so reloads and repeat visits inflate the number above your real lead count, or you are missing conversions entirely because phone calls and offline deals never get sent back to Google. Compare the two numbers directly: if Google is higher, you have phantom conversions, and if your inbox and phone log are higher, you have invisible ones. Both are fixable inside your account, and both distort every spending decision until you fix them.
Does the June 2026 Google API change mean I have to fix my tracking right now?
If you manage your tracking inside the Google Ads interface yourself, no, this is context rather than an emergency. As of June 15, 2026, Google blocked new offline-conversion imports through its legacy Google Ads API and moved users to the newer Data Manager API, which mainly affects developers, tools, and agencies uploading data through the API. It matters to you only if someone is importing your closed deals into Google on your behalf. If so, ask whether they have moved to the Data Manager API, because that pipeline is how your most valuable data gets back into the account.
Most owners spend their time arguing about budgets and bidding strategies because those feel like the levers of control. They are the last thing you should touch. A bid is a bet, and you cannot place a smart bet on numbers you have not verified, which is the same reason chasing Performance Max ROAS without checking real incrementality leads people astray. Fix the tracking, watch the real number surface, and brace yourself, because the campaign you have been proudest of all year may turn out to be the one quietly costing you the most.
