It seems like mobile ads are extremely incestuous. Game A advertises games B to M, which in turn advertise all the others.

In many ways, yes they are. Especially if you like inside individual genres. But mobile games also have so so many players and a rotating player base. Even old games can still attract new players etc. But yes, they are pretty incestuous.

But that's the market. It's unlikely to see massive growth like it has in the past. Mobile games have become so common that they've pretty much saturated the market and rotate players around. The same idea could kind of be said about things like movies or theaters, but the business still works.

The games themselves probably all work on a freemium model, but even given the whale dynamics there, it seems unlikely that the games produce enough revenue to offset literally billions of ads.

Whale dynamics are a huge part of this, and the spenders on these games absolutely do produce enough to pay for the ads. If they didn't, the companies wouldn't be running them.

Let me put it this way - I've seen companies run games all the way through the process from "fake ads" to a fully released game... And then shut it down because the players "only" end up spending 2-3x what it cost to acquire them through advertising. 3x their investment is seen as a failure because of the cost to build them. That's how important it is to them that they run these fake campaigns so they can bail on the failures early. And their targets for successful games land in 3-8x the advertising budge to be successful. Though exact ranges depend on genres and the "longevity" of a player and lots of other things.

I'll also add, as expensive as you might think running ads is, actual development is significantly higher. Ads will likely be run for a long time on a successful game, but the advertising for 6 months is way cheaper than spending 2 years with engineers, artists, designers, QA, and management all on the project. If they can spend 200k on advertising in 6 months to gauge interest, that's only costing the salary of like 2 engineers, so it's highly worthwhile. Most mobile game "success" rates are well below ten percent.

Also, how exactly does your analysis square with the fact that I've seen the exact same game ads for years? It doesn't really make sense to advertise 5 years for such throwaway products.

To be honest, I can't answer this one with confidence. I've seen multiple companies using the strategy I outlined, so I know it's pretty common. I also know that those companies were copying the strategy from other companies in the space. So I know it's prevalent. But that's not going to be every single ad you ever see.

I'll point out a couple things:

How exact is exact? Are you sure it's the exact same video down to a T? They may be floating multiple ideas at a time, and games can live in this "fake ad" state for multiple years while they iterate on it. Everything from different sound effects but the same video, different visual themes, running cuts of players doing well vs poorly, changing individual words in the messaging, etc. They then test these against each other to see which do better. I've seen some run for a while, but I've never felt confident it's actually exactly the same.

And if that's the case... Is it possible someone saw that and ad was fake but thought it was a good idea, and now a different company just literally copied and posted the same video?

Second, this may just be a "market analysis" learning vehicle. They may never intend on building the game. For example, if a company is thinking about game A, they may run ads, see it doesn't work, kill the project, and start considering game B. Now they already have data on how game As ads ran, and they can use their original ads as a "control" and try different variants to see what does better, and then use that data to determine how to best advertise game B. Or they may test game B against game A. Then they might see that it's doing worse than A, and try something else.

Third, some of this may be chasing measuring "seasonality". Game genres trend back and forth over time. They may use an old ad they put together to test the waters now to test the water again.

Fourth, I'm not totally convinced these are always studios running the ads. These might be publishers that never intend on building the game, but are trying to find info on what types of games are trending and what genres they should be invested in. Or they might be the advertising networks just running bullshit ads to gauge how much they should bid for ads in a particular genre. Or maybe it's some giant joint venture like Tencent who owns tons of studios and is gaging what they should be recommending their studios work on.

Data is extremely valuable. In many forms. And many people will pay for that data. And this type of data is such an accurate gauge of actual user behavior because it is literally actually current user data.

There are about 3 companies that dominate 90% of the mobile games market and these ads are extremely cheap to produce

Also there is tons of money in mobile gaming, just look at Blizzard, their highest revenue brand is Candy Crush