The Auctions You Don’t Want to Win: What Branded App Store Search Is Really Buying
Anyone who runs a sizeable app store search budget will recognise the pitch. A vendor has “looked at your Apple Ads” and noticed that your brand terms have surprisingly low paid share of voice. Wouldn’t you like to fix that?
Daniel Mori, Performance Marketing Lead at Life360, gets some version of that message almost every week. He understands why it keeps coming back: it works because everyone in the room has a reason to say yes.
“If you’re in marketing, spending more on branded search promises a lower CAC, at least if

you’re looking at paid channels only,” he says. Finance sees a channel that looks extremely efficient on paper. Leadership hears “dollars left on the table” and doesn’t want to be the one who missed it.
None of this is dishonest or irrational, Mori argues. It’s a “dashboard growth mindset”, where more branded spend produces more installs at a falling CAC, exactly the chart everyone wants to show their boss. The problem is that “the system is probably rewarding the wrong answer before anyone has asked the right question.”
That question is usually framed as incrementality, and Mori is a supporter of testing it. But he points to something more fundamental: when you buy branded search in an app store, what are you actually buying?
The $1,500 anomaly
Search marketers tend to assume they are buying intent. A user types a query, we infer what they want, and we bid accordingly. Mori’s view changed after a routine audit when he moved into user acquisition at Life360.
He opens every ad account himself rather than delegating to agencies or consultants. Google, Meta and TikTok threw up nothing major. Apple was mostly fine too, apart from one detail in a broad match ad group.
The long-tail keyword “find my friends” had spent around $1,500 over six months on clicks for the search term “Wizz dating app”, producing 54 installs. Small enough to fly under the radar for a business of Life360’s size, but it made no sense. What kind of search intent leads someone looking for a dating app to install a family location-sharing app?
The term was excluded immediately (“you don’t need incrementality testing to know” those users churned within minutes). But the question stuck.
Mori, who has a background in online dating, checked whether other dating apps triggered Life360 ads. Plenty of Fish, Zoosk, eHarmony, Match.com, Tinder, Hinge and Bumble: nothing, bar a single Hinge impression. Then he found Wiz’s old logo. It was the same purple as Life360’s.
To test the theory, he looked for another purple app in the dating and social space. Peeps had triggered around $900 of Life360 ads over the same period, with 66 installs.
The green wall
If visual similarity was driving taps, it should show up elsewhere. Mori turned to banks, which rarely defend their brand terms in app stores because their customers need the app regardless.
A search for TD Bank surfaced Cash App as the top ad. Another returned Acorns. Not every result followed the pattern (Experian sometimes won), but the “green wall effect”, as he calls it, was hard to miss. Chase returned OnePay. PNC Mobile returned a fintech app with the same blue background and orange mark.
And it wasn’t only colour. Rocket Money appeared above Ally with a near-identical logo shape in different colours. A prescription savings app won an auction for Embrace Pet Insurance. iTrust, a homophone, regularly appeared for Truist.
The pattern held on Google Play too. Because Google app campaigns don’t expose search terms, the team took queries from Apple Search Ads and ran them manually on Android devices, deliberately without automation so they could see as much as possible.
The explanation, Mori says, is in how the auction works: “The auction isn’t actually selecting for relevance, it’s just selecting for tap rate. And relevance is an important part of tap rate, but it’s not the only one. People react to what they see.”
What this means for brand defence

For Life360, the finding shaped branded search strategy before any incrementality results came in. The team defends its brand “enough”, but doesn’t bid on every auction or chase impression share at all costs.
If a meaningful share of taps on brand terms comes from visual confusion rather than genuine intent, then paying aggressively to own every one of those auctions isn’t the opportunity the vendor email suggests. “This is not a missed opportunity,” Mori says.
Life360 does run brand defence on both mobile web, where attacks are more aggressive, and in app stores, with separate tactics for each. He declined to share specifics: “I really love not spending a lot on brand search.”
An open question on conquesting
The finding raises a less comfortable question for challengers. Mori has never been a fan of conquesting. The brand owner’s tap rate on its own navigational query is so much higher that a challenger needs a far higher CPC just to compete.
But if visual adjacency lifts tap rate, could a shrewd advertiser use it to make conquesting cheaper? Cash App bidding on Chase probably gains nothing, he suggests.
Cash App bidding on TD Bank might gain just enough tap-rate lift to bring CPCs down. A challenger to Rakuten, such as RetailMeNot, might consider custom product pages that echo the competitor’s look. Life360 has no plans to test this itself. It has, however, put defences in place in case others do.
Behind the broad match
The anomaly surfaced because of broad match, which Mori defends as a deliberate “tax we pay to ultimately grow”. It accounts for under 15% of Life360’s Apple spend and has uncovered several search terms now running as exact match.
What makes broad match workable is the signal behind it. Life360 has built a synthetic in-app event: after registration, users go through a short observation period and are scored on early behaviours predictive of staying for at least a year.
Those above the threshold become a “quality new registration”, which feeds the machine learning that adjusts bids. For a product used as a group, much of that scoring centres on whether a new user brings others in early.
It’s effectively an in-house version of what Meta does, Mori says, and for Apple Search Ads it’s necessary. In his view, Meta remains more advanced than Apple for campaign optimisation.
The takeaway
The vendor pitch about branded share of voice will keep landing in inboxes. Mori’s point isn’t that branded search is worthless, or that incrementality testing doesn’t matter. It’s that before you pay to win every brand auction, it’s worth understanding why users are tapping in the first place, and accepting that some of the auctions you’re losing are ones you never wanted to win.




Comments