Thinking · Newsletter
Mobile game ads in 2026: formats, hooks and what is running
Mobile game ads in late 2026, by format and hook: what 12,311 public game ads read by Flixr Studio show, and what to make next.
TL;DR. Mobile game ads in late 2026 are mostly video, and most are not pure gameplay: in Flixr Studio's read of 12,311 public Meta ads from 23 game brands, gameplay-led ads were 34.9%, with animated or CGI stories (20.0%), motion graphics or statics (17.9%) and skits that cut to gameplay (16.2%) making up most of the rest. The most common opening is a character or an emotion (29.6% of ads), ahead of a plain gameplay opener (23.5%). About half close on an end card, and 30.8% show some gameplay that is not in the game. Our read: build each idea across several formats, open on a person or a feeling, and keep every version honest to the game.
Two meanings of "mobile game ads"
The phrase covers two different things. Developers who earn from advertising mean the ads a game shows its players: Google AdMob lists banner, interstitial, rewarded and native formats, where "Interstitial ads are full-screen overlays" and a rewarded ad is one where "users opt in to engage with an ad in exchange for a 'reward' within the app".
This page is about the other meaning: the ads a studio buys to sell its game, on Meta, TikTok, AppLovin and the in-app networks. That is mobile game advertising for user acquisition, and the creative is most of the job.
The data behind this page
Flixr Studio reads public Meta Ad Library ads from game brands with a model that watches each video or image and labels its format, hook, footage, characters and whether the gameplay is real. The set used here is 12,339 ads from 23 game brands, launched between October 2024 and November 2026 and nearly all (11,893) since July 2026, watched by Studio since October 2026. 12,311 have a reading, and every share is out of those.
Three limits. It is Meta only, so it has no playables. Eight of the 23 brands are labelled social casino, with 28.7% of read ads between them, so casino conventions weigh on the totals. And the readings are a model's: checked against a second model's blind labels, agreement is about 70% on format and about 95% on fake versus real gameplay. Competitors have no spend in Studio, so this shows what is running, not what is working.
Mobile game ad formats in 2026, by how they are made
84.0% of read ads are video and 15.6% static images, and 2,349 of all 12,339 ads (19.0%) ran as dynamic creative (DCO). The more useful cut is how each ad is made.
| How the ad is made | Share of read ads | Brands using it | What it is |
|---|---|---|---|
| Gameplay-led | 34.9% | 23 of 23 | Real or fake gameplay carries the ad, with captions, voice-over or an end card |
| CGI or animated story | 20.0% | 23 | A 3D or animated scene carries the ad; gameplay may follow |
| Motion graphics or static | 17.9% | 23 | Animated text, icons, slideshows or a still |
| Skit, then gameplay | 16.2% | 20 | A staged skit with actors (real or AI), usually cutting to gameplay |
| Creator over gameplay | 6.7% | 18 | A creator's face or voice over play: reacting, commentating |
| Live action | 2.2% | 11 | Filmed people or places that are not a skit or a creator |
| Creator talking to camera | 0.7% | 9 | Selfie-style UGC about the game |
| Character to camera | 0.7% | 8 | A game character or mascot speaks to the viewer |
Three things stand out. Pure gameplay is about a third of the market, not the default. Every one of the 23 brands runs CGI or animated stories. And people are everywhere even where the format is not "UGC": 31.5% of read ads (3,873) carry a creator, real-person or skit concept, led by scripted skits (14.0%), celebrity or pro endorsements (8.5%, concentrated in six brands), streamer reactions (3.8%) and reviews to camera (2.6%, but used by 17 brands).
Below the format, the craft details are consistent. 52.2% of read ads (6,427) close on an end card segment, 50.9% (6,261) carry a voice-over, and 10.1% (1,248) include footage the model judged visibly AI-generated or AI-enhanced.
Mobile game ad hooks: what the first three seconds do
| Hook (first three seconds) | Share of read ads |
|---|---|
| A character or an emotion | 29.6% |
| Plain gameplay opener | 23.5% |
| A reward or offer | 10.6% |
| Text question or claim | 9.2% |
| Spectacle or scale | 7.3% |
| Story cliffhanger | 5.4% |
| A choice or challenge to the viewer | 5.2% |
| A satisfying action | 3.4% |
| A fail or mistake | 2.4% |
The deliberate-fail opening that defined puzzle ads is now one of the rarest hooks in the set, at 299 ads; our puzzle teardown argued it was getting old. Offers as hooks are common (10.6%) but sit in 14 of the 23 brands, which fits the casino weighting.
Inside the ad, the messaging devices most often read were a big win celebration (22.6% of read ads), social proof such as player counts or rankings (16.3%), a how-to or tutorial (10.9%), a rivalry or versus frame (9.3%) and a dare to the viewer (7.8%). Come-back letters aimed at lapsed players ("we miss you") were 6.7%, from nine brands. Our taxonomy of hooks covers what each family is doing.
Real and fake gameplay in game ads
Of the 12,311 read ads, 44.5% showed real core gameplay, 30.8% some fake gameplay and 24.6% no gameplay at all. Fake gameplay is concentrated: the median brand's share was 11.9%, and four brands were at 50% or more. On 4 November 2026 the UK's Advertising Standards Authority published four rulings on mobile game ads, each upholding at least one complaint, and in King's banned Farm Heroes Saga ad it found the "Not actual gameplay" style disclaimer "likely to contradict, rather than clarify". Our stance is that the ad has to promise the game that exists; a hook is a promise explains why.
The full read on why studios run fake gameplay and where the UK regulator drew the line is in fake mobile game ads, and every number here is on the monthly State of mobile game ads page.
How to plan mobile game ads for 2026
- Read your category before you brief. Pull your rivals' live ads and tag them by format, hook and footage, as above. The gap is usually a format nobody in your genre is making.
- Build concept systems, not single ads. Take one idea and make it as a gameplay-led cut, an animated story, a skit cutting to gameplay and a motion cut. Those four formats together are 89.0% of read ads.
- Test a person-led opening against a gameplay opening. Character or emotion hooks lead the set at 29.6%; whether they beat plain gameplay for your game is a test, not a given. Read hook rate as a diagnostic and installs as the result.
- Cast UGC to a role. Skits and endorsements dominate the people-led ads. Write the role first, then cast; our UGC ads guide has the method.
- Make the playable and the end card from the same brief. Studio's Meta set cannot show playables, so brief them alongside the video; and 52.2% of the ads in this set include an end card. See the playable ads guide and the end cards guide.
- Keep the gameplay real, then stage it. Curated wins and a tidy board are still the game; an invented puzzle is not.
- Plan the refresh before launch. Decide what replaces each concept when it tires. Our note on how often to refresh mobile game creative gives the cadence logic.
Flixr makes the video, UGC, motion, playables and end cards from one brief, and reads the category with Flixr Studio. Examples of each format are on the showcase, in our mobile game ad examples gallery and in our game UGC and playable ads pages.
Questions people ask
What are the most common mobile game ad formats? In Flixr Studio's read of 12,311 public Meta game ads, gameplay-led video was most common at 34.9%, then CGI or animated stories (20.0%), motion graphics or statics (17.9%) and skits that cut to gameplay (16.2%). Playables are not in that set because it is Meta only.
Why are mobile game ads so weird? Partly because many are not gameplay: a fifth are animated stories and a sixth are skits. And 30.8% of read ads showed some fake gameplay, which the UK's advertising regulator has ruled misleading in specific cases.
Do mobile game ads use real gameplay? Most that show gameplay do. Of read ads, 44.5% showed real core gameplay, 30.8% some fake gameplay and 24.6% none.
How long should a mobile game ad be? It depends on the network and placement, so check each network's own specs. On AppLovin, see our AppLovin creative guide for the published numbers.
What makes a good mobile game ad? In our view: an opening that earns three seconds, gameplay the player will actually meet, and an end card that closes. Judge it on installs and early retention, not on clicks.
Sources: Flixr Studio public game ad readings (Meta Ad Library, 23 game brands), queried 9 Nov 2026; Google AdMob, monetize your mobile game with ads; ASA ruling on King.com Ltd, 4 Nov 2026; MobileGamer.biz on the ASA rulings, 4 Nov 2026.