What Does Winning UA with Playable Exploration Look Like
I write a lot about Chinese game companies. I do it from a place of admiration. Very few Western studios take this approach to UA creatives, and it's working really damn well.
In this piece I want to walk through how Microfun is doing it again, this time with a game called Flambe that caught some headlines recently for its growth and its creatives.
I wanted to dig deeper into its creatives and think about what goes into one of the most efficient game marketing and user acquisition engines in our market.
The game and its growth
Flambe started scaling about a year ago and now pulls roughly $10M in monthly IAP revenues according to Sensor Tower. It's a similar game to the other restaurant-themed merge and decoration games by Microfun, and their approach to creatives seems extremely methodical. At least evaluating by this chart.
What's happening with its UA creatives?
When the game launched, only ~200 creatives a month. Completely different concepts tested. No playables.
Fast forward to today: 720 new creatives. Out of them, 57 new playables uploaded in the past 30 days. About 25 completely new concepts and ideas (depending on how strict you are in defining "new concept").
Beyond new concepts, a lot of the rest are deep variations that target different styles of play.
A look at the scale of exploration
That's not the mind-boggling part.
What is, is how sophisticated these new concepts and ideas are.
I call this exploration of adjacent audiences. Audiences that would have a strong affinity to a target game but are used to other genres.
The most efficient way to reach them today is through creatives. As SDK networks like Applovin became extremely efficient at matching specific audiences to playables to full games, the way to "fish" for these audiences is by creating playables that balance two sides of the rope: adjacent audience appeal and the full game experience.
There are many voices in the game industry. But one thing I can't wrap my head around is how some folks expect to win by simply and mindlessly copying trending ad creatives.
A game doesn't grow because an ad creative just magically worked. It grew because that ad attracted an audience that enjoyed both the ad and the full game. An audience that converted to high-LTV users.
So if you have a car racing game and you see a trending ad for a puzzle game, copying that ad won't work. Yes, you might get some top-of-funnel interest for a limited time. But it'll probably do very little for revenues and ROAS.
To really run a data-driven UA creative strategy, you can't rely on blindly copying some trending ad because it worked for a different game.
The objective of a UA team today, among other things, is to take the game they're marketing, slice it into all possible aspects of play style and reasons to play, package them, and use them to reach pretty much every possible player audience on earth that might enjoy that game.
To fish for these audiences by throwing great playable baits.
The UA team as bait maker, if you will.
Let's see how the great team at Microfun makes baits.
Deep dive into the playables and reverse-thinking the rationale
Let's go through these playables and see what they're trying to achieve. Feel free to scroll through. It's super interesting.
Concept 1
A core concept that simplifies the merge puzzle into a more basic form: find similar items and connect them into a customer order. There's a layer of slot optimization with the bottom slots acting as temporary holding space until you need items.
Customers have limited time to be served. This appeals to a time-management-loving audience.
Concept 2
A more vanilla, pure merge gameplay stripped of customer order serving. This appeals to an audience looking for a more stress-free experience.
Concept 3
A concept that leverages the aesthetic of a hidden object puzzle together with merge. Note the thinking here. They're not blindly creating a hidden object playable. They're still connecting it to merge mechanics, fishing for an audience that comes to merge games from that angle.
Concept 4
The full game has a prominent meta loop of decorating and renovating a restaurant. Here the merge mechanic still appears, but as a mechanism for a kitchen decoration meta.
Concept 5
Another bait, this time for restaurant and cooking simulator lovers. An ice cream station mechanic. And yet, it doesn't give up on the core mechanic of merging items together into a final dish, a core component of the actual game.
Concept 6
Super simplified merge gameplay with only one item to merge, making it extremely easy and allowing for rapid fun gameplay. Targeting the pure merge audience once again.
Concept 7
An interesting one. The same concept of merging food to serve an order, BBQ station this time. But there's another mechanic: the conveyor belt. If that seems familiar, it should. Games like Pixel Glow and Sand Loop made the conveyor belt mechanism super recognizable and proved there's an audience that likes managing and optimizing resources. A conveyor belt isn't just a trend. It's also a great mechanism to attract the resource optimization audience.
Concept 8
This one targets an audience already familiar with the sliding/curling mechanism from Tasty Travel ads, merging drinks this time to get players making their way to bigger and bigger drinks.
A different style of play. A different reason to play. Adding an aiming mechanism to attract yet another type of audience.
Concept 9
Also familiar from Gossip Harbor. Adds the narrative aspect (save the girl) and combines it with sort mechanics to attract the sort and narrative audience.
Concept 10
Another one to attract the pure sort puzzle audience, but keeping the restaurant and serving customers theme intact.
Concept 11
Another interesting one. The classic save-the-girl and merge mechanic layered on top of home renovation. But note the extreme difficulty curve, aimed at bringing in an audience attracted to challenge and time management.
Concept 12
Once again, the Microfun exploration system is aware of the top trends in the UA creative market. This time using the color block jam puzzle as a hook to attract a new audience, but still keeping the customer-serving aspect.
Note the extreme difficulty curve here too, going from very simple to very hard straight away. The exploration of concepts is testing hypotheses about whether very hard playables can bring valuable audiences to the full game.
Concept 13
The merge mechanic makes room for a triple-match sort of mechanic. Still revolving around the core merge puzzle aspect (identifying matching items), but this time with a tap mechanic and still serving customers.
A time-pressure element is introduced as the board slowly fills up, and the player needs to act before items reach the counter.
Concept 14
nd lastly, a concept bringing in idle arcade gameplay married with 2D art instead of the usual 3D isometric art style.
Note the testing of various hooks like the tap-to-break piggy mechanic here, aimed at optimizing for players engaging with the ad first, then switching them into idle arcade gameplay once they interact.
Another playable without this hook also appears, so there's clearly a specific hypothesis being tested: does a better hook lead to more players engaging with the idle arcade gameplay?
The playbook for winning with UA: a system to hunt audiences and scale
Beyond getting ideas for how deep exploration could go, I think there are a few key takeaways:
Exploration is not throwing spaghetti at the wall.
Each playable is always tied to a specific aspect of the full game. It can look very different from the full game, but it still carries some core gameplay aspect that connects to the same reasons to play, bringing in users who will enjoy the full experience.
Everything is being tested.
Difficulty curves, art styles, various hooks, even switching quickly between a seemingly easy level and a very hard one.
Concepts target wide and adjacent audiences.
Pure sort puzzle players. Triple match. Hidden objects. Restaurant/cooking simulators. The net is cast wide, but always with purpose.
I actually like the bait metaphor. These winning UA teams are fishing for audiences with playables as bait that is always being tested, always connected to the fishing rod (the full game), always in many different flavors.
This playbook is what the teams scaling games today are mostly employing, and what we hope here at Sett to enable you to do at scale with agentic AI.
About the Author
Fishi is the Head of Marketing at Sett. His brain is a chaotic jukebox of ideas with more cultural references than any feed can handle. He collects sneakers and plays chess while you’re still counting sheep.
