“We pay Google so that people typing our name find us. They would have come anyway, wouldn’t they?” Monthly report open between us, the “Brand” line highlighted in marker. Nearly every advertiser ends up asking this, and it is a fair question.
Short answer: most of the time, yes, they would have come. That does not mean you should switch it off.
A brand campaign is not an acquisition lever. It is insurance. And insurance is judged on the risk it covers, never on the return it displays.
1. Why the brand line always posts the best number
Open any Google Ads account and sort campaigns by return on ad spend. Brand comes first, often at two or three times the score of everything else. Tiny cost per click, high click-through rate, high conversion rate. On paper, it is the best campaign in the account.
On paper only. Someone typing your name has already decided to come to you. The ad did not create that intention, it placed itself on the path. The account then thanks the campaign for a purchase it never caused, because last click collects everything.
My position is blunt: I never look at the return on ad spend of a brand campaign. That number does not measure the campaign, it measures how well known the company is. Which is also why paid and organic have to be read together, something I took from the other end in combining SEO and SEA.
2. The only figure that settles it
The real question is not “how much does this campaign bring in”, it is “what would I have had without it”. The trade word is incrementality: the share of the result that would not exist if you switched it off.
On that point there is one serious, public measurement. In 2012 eBay switched off its brand ads on two search engines while keeping Google as a control. The result, published by three researchers, two of whom worked for eBay at the time: 99.5% of the forgone click traffic was immediately recaptured by the natural result. The organic link sat right under the ad, and people clicked it. The same authors measure that a 1% drop in paid visits produces 0.5% more organic visits and 0.23% more direct arrivals. One door closes, visitors take the next one.
One caution before turning that into a rule. eBay is eBay: its natural result is untouchable on its own name. A small company whose brand is an ordinary word, or which shares its name with another business, is nowhere near that situation. The figure is not a law, it is a boundary. It tells you how far substitution can go when the brand is solid.
3. What you look at, and what you think you are looking at
The reading error is almost always the same: people judge the brand campaign using the brand campaign’s own columns. That scope is precisely the one that lies.
| The figure | The naive reading | What it hides |
|---|---|---|
| Campaign return on ad spend | “Best campaign in the account.” | The intention existed before the ad. You are measuring fame, not advertising. |
| Paid clicks on the name | “We gained those visits.” | A good share came from the organic link sitting an inch below. |
| Attributed revenue | “The campaign brought in this much.” | Last click harvests a sale decided elsewhere, often long before. |
| Total brand sessions, paid plus organic plus direct | Nothing spectacular to announce. | It is the only figure that genuinely moves when you cut. |
My method when a client wants a decision: I never switch off the whole account at once. I cut brand across part of the territory and leave the rest running as a control. The two zones are then compared on the total, not on the Ads line, and it runs until the control zone has stacked up enough conversions for the gap to be readable. Same demand for a proper baseline as in everything GA4 no longer sees: without a comparison point, an isolated number proves nothing.
4. The four situations where I keep brand, no argument
Cutting on principle is as dumb as paying on principle. Four cases justify the spend, and I check them before any recommendation.
First case, a competitor is bidding on your name. That takes seconds to confirm in the auction insights report, where overlap rate tells you how often someone else’s ad showed at the same time as yours. A high rate on your brand keywords, and the question is closed: you stay.
Second case, your natural result is not first. A brand built from a common word, a namesake in another sector, a distributor ranking above you on your own name. The eBay reasoning no longer applies at all: there is no free escape door sitting just below.
Third case, you have a message organic will not carry. A dated promotion, a change of address, a hiring push, a product recall. On a website you have to freeze a message; on an ad you change it the same morning. The organic title takes weeks to move, the ad obeys immediately.
Fourth case, your brand results page is crowded. Directories, comparison sites, review platforms, an old reseller still hanging around. When the first result is not you, the ad takes back control of the top of the page, and that is the direct continuation of the work described in when a review platform outranks you.
5. The competitor buying your name
This is the case that frays nerves, and the one where I see the most money badly spent. The reflex is to raise bids until you take back first place. It is the most expensive and least effective move available.
Start with the complaint. Google states plainly that it will not restrict the use of trademarks as keywords, but that it will restrict the use of a trademark in an ad from a direct competitor, once the owner files a complaint. Both sentences sit on the same policy page and they change everything: you will not stop a competitor buying your name, you can have your name pulled out of their ad.
On the legal side, the Court of Justice of the European Union set the frame in its judgment of 23 March 2010: the origin function of a trade mark is adversely affected where the ad does not enable, or enables only with difficulty, a normally informed and reasonably attentive internet user to tell whether the goods come from the trade mark owner or from a third party. The exact reach of any given case is decided on the file, and by a lawyer, not by a paid search consultant.
The bet is easy to frame. At worst the complaint is refused and all you lost was the file you put together. At best the competitor loses your name from their headline, and an ad that no longer contains the searched word does not harvest much.
6. What automated campaigns do with your brand
The subject changed nature once Performance Max and automatic query expansion arrived. A campaign meant to win new customers absorbs brand traffic, converts it effortlessly and reports a flattering result. You think you are funding conquest, you are funding renewal.
Google provides the guardrail: brand exclusions, which keep a campaign from serving on queries containing your name, its variants and its misspellings. On Performance Max they apply to Search, Shopping and YouTube search inventory, with the option of letting Shopping ads through. I put them in place before reading a single figure, exactly as in Performance Max: taking back control.
Without that separation none of the earlier questions has an answer. You cannot decide whether to pay for your name while you still do not know how much you already pay for it without realising.
The most expensive mistake
Cutting brand, watching the Google Ads line collapse, panicking and switching it back on. You learned nothing, except that paid traffic is paid. The figure to watch is total revenue in the zone you cut, against the zone left running. At worst you confirm the campaign does something and you switch it back on knowing at last what you are paying for. At best you free up a whole budget line and put it back where it creates demand. Reading paid and organic against each other is the heart of a Google Ads engagement and of the analytics setup that goes with it.
7. What I actually answer
To a small company with a brand its customers know, a natural result in first position and no competitor in the auction report: cut it, and move the money onto the queries that do not know you yet. To a young brand, a namesake, one resold by others or attacked by a competitor: keep it, and steer the spend by a ceiling rather than by the reported performance.
In between there is no rule, there is a test. And this is the moment to be subtle rather than greedy: you do not cut a campaign to save money, you cut it to find out what it was really paying for.
In the crucible: separating what you buy from what you already had
An ad account permanently mixes two materials that look alike: demand you created and demand that existed before you. Both arrive through the same channel, wear the same costume, land in the same column. Only the test of removal tells them apart.
That is the whole job. Take out a piece, look at what still stands, and pay only for what collapses without it. A brand you buy back every month is not a brand you own. It is a brand you never weighed.