Refocus has 2,407 App Store ad records tracked in the last year, across 24 EU storefronts, which makes it the second heaviest advertiser among app blockers. It did that on four creatives, and between those four there are only two headlines.
One of the two has been running for 249 days.
Everything below comes from Apple's public Ad Repository, the transparency dataset the EU Digital Services Act requires. It shows every App Store ad an advertiser ran, where, in what language, and for how long. The repository lists the publisher as Labi, LLC. Data window ends 24 July 2026.
One English line, 24 storefronts, 249 days
The line is "Blocker, Limit Screen Time". It is in English, it runs in all 24 storefronts Refocus buys, and it was still recorded on 23 July 2026, the last full day in our window.
It carries the volume too. The version with a screenshot asset accounts for 859 records on its own, and the icon-only version another 324. Apple groups the same headline twice, once as an icon ad and once as an icon plus asset ad, which is why four creatives comes down to two pieces of writing.
For most of the year, that one sentence was the whole account. Every country, every impression, one line of copy in one language.
The one experiment it ran, and killed
The second headline was "Block Websites, Control Time". It ran for 25 days and stopped on 18 November 2025, in Bulgaria, Estonia and Slovakia.
Three of the smallest storefronts on the board, three weeks and a bit, then off. That reads like a small-market test: put the alternative angle somewhere cheap, see if it beats the incumbent, drop it when it does not. The winning line went untouched, and nothing has been tested against it since.
Worth sitting with how little that is. Refocus spent a year at the second highest record volume in its category and wrote two headlines, one of which it abandoned inside a month. AppBlock, the one advertiser above it, ran 20.
No local-language copy anywhere
Refocus sells in Bulgaria, Estonia, Slovakia and twenty-one other EU storefronts in English. There is no German line, no French line, no Spanish line, nothing in any local language at any point in the window.
The obvious reading is that this is a gap. The data does not support that as cleanly as it should.
AppBlock does the opposite thing and also survives. It has 2,973 records tracked, 20 creatives, and copy in seven languages: French running 197 days, German 193, Italian 182, Portuguese 182, Dutch 166, Swedish 127, Spanish 87, alongside an English line, "Limit Screen Time | Block Web", at 233 days across 23 storefronts. Almost all of it was still live on 23 July.
Localizing cost AppBlock something, though. Its Dutch line "Stop met uitstelgedrag" lasted 29 days and stopped on 9 February 2026. Its Spanish "Distraerse menos" lasted 86 days, stopped on 29 April 2026, and was replaced by "Limitar tiempo de pantalla", which is at 87 days and still live. So AppBlock's seven languages include two lines it had to bury, and Refocus's zero languages include an English line older than AppBlock's oldest.
One more fact against the easy conclusion: the oldest still-running screen-time ad in the EU belongs to neither of them. It is detoxi's German line "Reduziere digitalen Stress", which Apple dates past its one-year window and which was still recorded on 23 July in 20 storefronts, most of which do not read German. Localization, no localization, and German copy in Ireland and Portugal all have a survivor. Anyone who tells you the data settles this is reading their own preference back to you.
Search results only
Placement is one-sided and there is nothing to interpret. Of the 1,200 Refocus records we analyzed, every single one is an App Store search results ad. No product page, no search tab.
That is the category norm rather than a Refocus decision. AppBlock's 1,173 analyzed records are all search results as well, and so are Opal's 518. Refocus buys the person who already typed "app blocker" or "screen time", and it does not buy attention on a competitor's product page.
What we cannot see
Records are repository records, a proxy for volume. They are not euros, and nothing in Apple's dataset tells us what Refocus spends, what it pays per install, or whether any of this converts. Apple also caps each country query at roughly 50 records, so of the 2,407 records tracked for Refocus we analyzed 1,200. The headline counts, dates and storefront lists hold. A spend ranking would not.
What a founder can take from this
- You are allowed to stop writing. Refocus found one line, put it in 24 countries, and left it alone for 249 days. If a creative is still running after eight months, the useful question is what else you could be doing with the hours you would spend replacing it.
- Test in the cheap markets and let the test die there. The "Block Websites, Control Time" experiment ran in Bulgaria, Estonia and Slovakia and never reached the big storefronts. Whether it lost or just did not win by enough, the shape is right: challenge the incumbent line somewhere small, and only promote a challenger that earns it.
- Localization is a choice, not a checklist item. The two heaviest advertisers here do opposite things and both keep a line alive past 230 days. Pick based on whether you can maintain seven languages, because the losses AppBlock booked in Dutch and Spanish are the real cost of that decision, not the translation fee.
You can pull any advertiser at recao.app, free, no account, in about a minute.
Method note: figures are computed from Apple's public Ad Repository (DSA Article 39), swept across all 25 EU storefronts with zero failed country queries, data window ending 24 July 2026. "Days running" is the span from a creative's first to last recorded impression across those markets. "Still live" means the creative was recorded on 23 July 2026, the last full day in the window. Apple reports creatives older than its one-year window as over a year rather than a date, which is Apple's own marker, not ours. Record counts are repository records across those storefronts, a proxy for volume, not euros and not spend, and Apple caps each country query at about 50 records, so 1,200 of the 2,407 records tracked for Refocus were analyzed. Longevity is a proxy for performance, not a measured conversion rate.