Is Brand Search a Waste of Money? Jones Road Tested It Four Times, Here's Our Take

Most of the money you're spending to bid on your own brand name is buying you customers you already had.

That's an uncomfortable thing to say, because brand search feels like the safest line in the whole account. It's cheap. The ROAS looks incredible. Your agency reports it as a win every month, and your Google rep keeps nudging you to spend more on it. So nobody questions it. But "cheap and high ROAS" and "incremental" are two completely different claims, and the gap between them is where a lot of budget quietly leaks.

I'll back the claim up, because you shouldn't take a contrarian line on faith any more than you should take the default one.

The cleanest evidence I've seen comes from people who tested it properly

There's a well-known beauty brand that ran a branded search holdout not once but four separate times over a couple of years, as the business changed shape, because the answer genuinely matters at scale. Every time, the read came back roughly the same. When they switched brand search off, about 66% of the clicks they'd been paying for simply moved straight to their organic listing instead. Same customer, same purchase, now free.

By the fourth test the cost to acquire a genuinely new customer through branded search worked out to something like $700, and even then the incremental lift was barely there. The honest summary from the operator running it was basically "do not let me run this again." Four tests. Same answer four times.

And this isn't one weird case. The most famous version of this story is a large home retailer whose paid search got switched off by accident. Days went by before anyone noticed, and when they checked, sales hadn't dropped. The founder spotted it before any of the marketers did, at a business doing over a billion in revenue. Think about what that means: a whole team was reporting branded search as a performing channel, and the revenue it was "driving" was sitting there the entire time, free for the taking.

Why the ROAS looks so good and means so little

Here's the thing about someone typing your exact brand name into a search bar. They've already decided. The intent happened upstream, on Meta, on YouTube, from a friend, from a podcast, from the box that arrived last week. By the time they're searching "[your brand]", the buying decision is mostly made.

When you bid on that term, you're not creating the sale. You're putting a tollbooth in front of a customer who was walking through the door regardless, and paying for the privilege. The platform is delighted to let you do this. Some of the data floating around suggests roughly 98% of people who search a brand end up buying that brand one way or another. So the ad gets the credit, the dashboard lights up green, and the underlying truth, that most of those people would have arrived for free, never shows up in a standard report.

That's the trap. Branded search isn't expensive in absolute terms, so it never trips the alarm. It just sits there looking like your best-performing campaign while doing a fraction of the work it claims.

So my default position is: assume your brand spend is mostly non-incremental until a holdout proves otherwise. Make it earn its place rather than inherit it.

But here's where I break from the purists

If I stopped there, I'd be giving you half an answer, and the half that's easy to tweet but wrong in practice. Because there are real situations where I'll keep bidding on brand even knowing the recapture rate, and I think the people declaring "brand search is always a waste" are missing them.

The question was never "is branded search incremental in a vacuum." It's "what's sitting above your organic listing when someone searches your name." If the answer is "just your own free link," switch it off. If the answer is anything else, you might be defending revenue rather than wasting money. Three cases in particular.

1. When Amazon or a competitor is squatting on your term

This is the big one right now, and it's changed the maths over the last couple of years. Amazon has been bidding hard on brand terms, and they don't show up shyly. When they bid, they tend to grab the top two or three links and a top-of-page rate up in the 90s. I've watched a single brand's Amazon impression share on their own name climb from around 20% a few years ago to roughly 60% now. That's not a typo, that's Amazon advertising your customers to themselves.

And it gets worse, because the marketing you're doing on TV and top-of-funnel Meta drives more people to search your name, and Amazon is sitting there waiting to intercept the extra demand you paid to create. So you can end up in a situation where you now have to pay the Google tax purely to avoid the Amazon tax. It's a slightly absurd place to be, but it's real.

In that world, switching off brand search doesn't send those clicks to your free organic listing. It sends them to a third party who'll happily take a thinner margin, sometimes undercut your own pricing, and pocket a customer you spent real money to generate. Here, brand bidding isn't incremental in the textbook sense, but it's defensive, and defence has value when someone's standing on your front lawn.

2. When your Shopping results are a mess of resellers

Same logic, different shelf. If you sell through any channel where third parties can list, and you search your own name in Shopping, you sometimes find a scrum of resellers, grey-market sellers, and the occasional outright scammer ranking on your product. The same beauty brand from earlier kept brand search live partly for exactly this reason, because the organic Shopping surface around their name looked rough.

If a customer's first impression of your brand is a cluttered, off-brand, slightly dodgy-looking set of results, the cost of bidding to sit cleanly at the top isn't really an acquisition cost. It's brand control. You're not paying for the click so much as paying to make sure the click that was always going to happen lands on you, at your price, with your presentation, rather than on someone reselling your stuff at a markup or worse.

3. When your organic listing is weak or barely there

The whole "66% goes to organic" argument quietly assumes you have a strong organic presence to catch it. New brands, or brands that ignored SEO for years while they poured everything into paid, often don't. If your organic listing for your own name is thin, buried under retailers, or just not compelling, then turning off brand search doesn't hand the customer a clean free path to you. It hands them a gap.

Plenty of brands spent a decade treating organic as an afterthought and are only now rebuilding it. Until that foundation is actually there, a small amount of brand bidding can be the thing holding the top of your own search results together. The fix is to build the organic muscle so you can stop paying for the gap, but you bid in the meantime with your eyes open, not as a permanent habit.

How to actually find out, cheaply, this month

The good news is you don't have to argue about this. Branded search is one of the easiest things in all of marketing to test, because the spend is small and the risk is contained. Worst case, you lose a little money for a couple of weeks and learn something genuinely valuable.

Here's the version I'd run if I were sitting in your seat:

• Pick a geo split, not an all-or-nothing flip. Hold out brand search in one set of regions and keep it live in a matched set. That way you're comparing like for like instead of guessing off a week-over-week wobble.
• Watch total brand revenue, not the campaign's own numbers. The whole point is to see what happens to your blended brand-driven sales when the paid layer goes away. If total brand revenue holds while you've stopped paying, you just found your answer.
• Run it long enough to clear the noise. A few days isn't a read. Give it two to four weeks so the confidence interval tightens and you're not making a permanent call off a random spike.
• Then check the three exceptions by hand. Search your own brand name in Google and in Shopping. Is Amazon or a competitor above you? Are resellers cluttering the results? Is your organic listing weak? The test tells you the incremental number. Those three checks tell you whether to keep a defensive sliver anyway.

That's it. A geo holdout and four manual searches will tell you more about your brand spend than any dashboard has all year.

So, waste of money or not?

My honest take: for most brands, most brand search spend is recapturing demand you'd have got for free, and you should be sceptical of it by default. But "default off" is not the same as "always off", and the three cases above are where a blunt rule costs you money instead of saving it.

Run the holdout. Do the four searches. Then make the call with your own numbers instead of your Google rep's. And if you find yourself defending a chunk of brand spend, make sure it's because something's standing on your lawn, not just because the ROAS looked nice.

What's sitting above your organic listing when you search your own name right now? Go and look. That single check decides most of this for you.