Bidding on people who already knew you

Your best-performing campaign is almost certainly the one that deserves the least credit. It is the one with your own product name in it, and on paper it looks magnificent: the lowest cost per signup in the whole account, a conversion rate none of the others can touch, steady week after week. Of course it does. Every single person who clicked it typed your name into a search box first. They had already decided. That ad was a turnstile, not a salesman.

There are only two jobs in paid search, and they are not equally hard. One is creating demand — putting you in front of somebody who has never heard of you and giving them a reason to care. The other is harvesting demand that already exists — being the first thing a decided person clicks. Both cost money, both show up in the same column of the same dashboard, and one of them is roughly ten times harder than the other.

Why the harvest always wins on paper

Because credit goes to the last click. Somebody reads something you wrote in March, thinks about it for three weeks, finally searches your name in April and clicks the ad sitting at the top. The ad gets the conversion. The thing that actually did the persuading gets nothing, because it happened too long ago and left no trace a platform can see.

Now run your account the way every dashboard invites you to and shift budget toward the cheapest cost per signup. You will steadily move money out of the work that creates demand and into the work that only collects it. Your reported numbers will get better the entire time. Your growth will quietly stop, because you have optimised your way into harvesting a field that nobody is planting any more.

the cheapest clicks are the ones you already paid for

So should you switch it off?

Not automatically, and the honest answer hangs on one thing you can check in about a minute: search your own product name and look at who sits above you. If a competitor is buying ads against your name, your brand campaign is defending a door somebody is actively trying to walk through, and that is usually money well spent. If nobody is up there and your own organic result holds the top spot anyway, you may be paying a toll on traffic that was already yours.

The way to settle it is a holdout, not an argument. Pause the brand campaign for two full weeks and watch the number in your own records — accounts created, purchases made — not the number in the ad account. Suppose that campaign had been reporting forty signups a month at three dollars each. You pause it, clicks on your organic result climb, and your real signups fall by four. Then the campaign was never buying forty signups for a hundred and twenty dollars. It was buying four, at thirty dollars each, and the other thirty-six were people who would have found you regardless. Same spend, ten times the true price, and no dashboard was ever going to tell you that on its own.

Two lines, never one blended number

Keep brand and non-brand apart in your reporting permanently. A blended cost per acquisition is one of the most flattering lies in marketing: it gets cheaper every time more people learn your name, which means it improves fastest precisely when your ability to reach strangers is getting worse. The two numbers answer different questions. Non-brand cost per signup is what it genuinely costs to acquire someone new, and that is the one that belongs in any plan about growth. Brand cost per signup is not an acquisition number at all — it is the price of a defensive position, and you should weigh it the way you weigh insurance.

There is a structural version of this mistake as well, and it is worth fixing today if you have it: brand keywords sharing a campaign and a budget with non-brand ones. The auction will do exactly what you asked and pour money into the conversions that come easiest, so the easy half eats the budget while the hard half — the strangers, the actual growth — starves, and your averages look terrific the whole way down. Split them. Separate campaigns, separate budgets, separate targets you set deliberately.

The number almost nobody watches

Here is the flip side, and it is the useful part. How many people search your name each month is itself a metric, and it is the cheapest honest read you will get on whether all the unmeasurable work is doing anything. The video, the guest appearance, the writing, the thing somebody said about you in a group chat — none of it produces a clean conversion you can point at, but all of it ends the same way: eventually someone types your name. Brand search volume rising month over month means more people are carrying you around in their heads. It moves slowly, it lags by weeks, and it is a monthly number rather than a daily one — check it every morning and you will see nothing but noise.

All three pieces live in tools you have probably already connected: brand queries in your search console data, the brand campaign in the ad account, real accounts in your own records. We pull those into one place so you can hold brand and non-brand up as two separate numbers instead of one blended average, and when we draft a campaign it arrives paused and waits for you — which matters more than usual here, because a brand-versus-non-brand split is exactly the kind of decision you want to look at before any money moves.

The one thing to keep: never judge the campaign that spends against people who already know your name using the same number you use for the ones who don't. They are not two versions of the same campaign. They are two different businesses sharing a login.