Google said the ads lost money for years. The tracking was never set up.
A founder-led performance-apparel brand had a Google Ads ROAS below 1.0 for more than five years, a number that says every dollar spent came back as less than a dollar. The owners assumed the ads were failing. The real story was simpler and stranger: the conversion tracking had never been set up, so most of the purchases were never counted.
The account was being graded on a fraction of the truth.
Lifetime reported return sat at 0.81 while real sales kept landing. Event accuracy was near 40 percent, and a large share of revenue ran through marketplaces the platform never sees.
We rebuilt the signal before anyone judged the spend.
A server-side tracking stack fired from the brand's own infrastructure, so every purchase was captured as first-party data and reconciled against the store as the single source of truth.
"Fix the ads, get the tracking right, then we scale."
A founder-led performance-apparel brand with genuine engineering advantages and loyal repeat buyers came to us with a tidy, tactical ask: make the ads work, get the tracking right, then put more money behind it. Customers who bought tended to stay. Getting them in the door felt like the bottleneck. It was a reasonable ask, and the product and the repeat loyalty were genuinely strong.
But the feeling underneath it was the one every operator knows: the business looked profitable, the sales were real, and the dashboard still kept saying the marketing was losing money. Working hard, with a good product and steady demand, and still stuck because the numbers and the bank account would not agree.
It wears on you in a specific way. You stop being the founder who builds with conviction and start being the one who second-guesses every decision, because the scoreboard you are supposed to steer by keeps calling your best bets losses. The question stops being how to grow and becomes whether you can trust your own dashboard at all.
That belief, that a low return means weak ads, feels safe because nearly everyone shares it. The moves it leads to are the ones every team reaches for: more spend, new creative, swap the vendor. The catch is that all of them assume the number is true, so they spend against a reading that was never honest, and the gap never closes. Safe is not the same as cheap.
What we found when we looked closer
So before touching a single campaign, we pulled the lifetime numbers and held the reported return up against what the store had actually recorded. Google showed a lifetime ROAS of 0.81, built across roughly five years and a spend in the mid six figures. Read literally, that says the brand sent money out, got less back, and gave product away at a loss.
Then the conversion tracking told the real story. It was not set up incorrectly. It had never been set up at all, buried under legacy tags and orphaned plugin code, with a large share of revenue flowing through marketplaces where the ad platform sees zero conversions. Event accuracy sat near 40 percent. The platform could not count most of the purchases, so it could not learn what was working, and the return it reported was measuring almost nothing.
You already know this in your own work: a scale only weighs what is set on it. If half the order never makes it onto the scale, the reading is light, and trusting it ships the wrong amount. The brand had been weighing its entire paid program on a scale that was catching less than half of what was real, then making spend decisions off the weight. A new creative or a new vendor would have changed what sat on a broken scale without fixing the scale.
The intervention
We fixed the scale before we read another number off it. The tracking stack was rebuilt from zero: a new tag container, server-side events fired from the brand's own infrastructure, hashed customer data, and every tool tied together so conversions were captured as first-party data the brand owned rather than third-party signals the browser kept dropping.
Once the events fired reliably, event accuracy moved from roughly 40 percent to 100 percent. For the first time the platform could see the purchases and reconcile them against the store as the single source of truth, with the gap between reported return and recorded revenue finally closing. Nothing about the product changed. We changed what the numbers were counting, and what the brand could finally trust.
Rebuild the signal before you judge the spend.
Separate the return the ad platform reports from the revenue your store actually recorded, then rebuild server-side tracking so the platform can count every purchase and reconcile to the store as the single source of truth. Until the signal is honest, the return you are reading is fiction, and the gap between the two numbers is where the truth has been hiding.
The result, in context
With the signal clean, the account stopped lying. The real paid return settled in the 2.5 to 2.8x range, blended return ran roughly 4 to 6x through peak season, and since the rebuild the brand recorded revenue up about 107 percent and orders up about 87 percent. The 0.81 was never the ads losing money. It was the cost of never having set the tracking up.
What this case does not claim is a rescue. Performance beat the brand's own best-ever baseline, but the binding constraints sat outside marketing, in unit economics and seasonality, and we said so plainly. Honest measurement is what let everyone see that clearly, which is the point. The number is the evidence, not the trophy.
Figures are real and client-attested (agency-measured), banded to protect confidentiality, and not third-party audited.
For years I thought the ads were the problem. The ads were never the problem. We just were not counting what they did.
Who this is for
If your Google Ads ROAS reads below 1.0 while real sales keep happening, this is for you.
Not because the ads are weak. A reported return far below 1.0 rarely means the ads are losing money. More often it means the purchases are not being counted, because the conversion tracking was never set up correctly, or because privacy changes and marketplace sales are quietly hiding the conversions from the platform.
It is also for you if a large share of your revenue runs through channels the ad platform cannot see, and you cannot tell how much of your real return the dashboard is actually capturing. You probably already suspect the two numbers do not agree.
Picture opening your account and your store side by side and, instead of arguing about whether the ads work, seeing exactly which purchases the platform counted and which it missed. That is the shift. Not a better number, a number you can finally trust enough to spend against.
Once you have seen which purchases the platform counted and which it missed, the question you have been asking changes on its own. It stops being whether the ads work, and becomes how much real return has been sitting there uncounted the whole time. That gap is yours to find.
We open your ad account and your store next to each other and show you the gap, on your own numbers, before anyone touches a campaign. It is not a sales call. It is a Second Opinion for brands that suspect their data is lying and want to know for certain.
Clarity is only the start. When you can see how your business really works, the hard calls get easier and the thing begins to run on its system instead of on you. That is what we are actually building toward.
Before you scale another dollar
Why is my Google Ads ROAS below 1.0?
A Google Ads ROAS below 1.0 often means purchases are not being counted, not that the ads lost money. If conversion tracking was never set up correctly, the platform records spend with little or no revenue against it, so the number reads like a loss even when real sales are happening. Reconcile the reported return against the revenue your store actually recorded before you judge the ads.
Why is my reported ROAS lower than the sales my store shows?
When the reported return is far below the sales you can see, the platform is usually counting only a fraction of your purchases. Events may never have been set up, may fire on the wrong action, or may be blocked by privacy changes, and any revenue running through marketplaces sits outside the store pixel entirely. Treat the store as the single source of truth and rebuild the tracking until the two numbers reconcile.
What is server-side tracking and do I need it?
Server-side tracking sends conversion events from your own server instead of relying only on the browser, which makes them harder to lose to ad blockers, privacy settings, and cookie limits. It turns third-party signals the browser keeps dropping into first-party data you own. If your event accuracy is low, a large share of revenue runs through marketplaces, or your reported ROAS does not match your store, server-side tracking is usually the foundation that has to be rebuilt first.
Wondering whether your ROAS is even real?
The first step is seeing what your tracking is actually counting. The diagnosis is independent and yours to keep. There is no obligation to have us rebuild it, and no half-answers that end in a referral list.
Start with a Second OpinionA representative engagement from the Business JetPack Diagnosis Library. Identifying details have been removed or changed to protect client confidentiality. Figures are real and client-attested, not third-party audited, and reflect a specific engagement; results vary.