One saying has stuck with me whenever I consider using free apps, website registrations or services:
"If you're not the customer, you're the product."
If I'm not paying money to the company, I'm probably being sold to someone else. Or maybe both.
The source of the quote gives more context. They were talking about advertising. For example if they give a "free" paper out at the metro station, the product being sold is how many people read the paper, and the customers are the advertisers. It's the same with TV, eyeballing the tube is free, but the more eyeballs the higher the bidding for the advertising slots.
For a small app developer they'll be almost all advertising driven and not data-selling. That's because you would really need to pay someone to oversee that, and unless there's quite a bit of data, it's not worth the value of paying someone a wage to handle that part of things. Also there's the cost of maintaining a server, database, database management etc, professional relationships with who you'd be selling the data to, etc. Basically I can't see this being a cost effective way to make money until it's scaled up quite a bit. What's someone's data worth? A few cents maybe? And you need to hire a pro to run this for you. Say you pay someone low wages of $30000 a year to handle this, and just pretending you don't have any infrastructure costs at all, and you get 10 cents worth of data a year per user, you'd need 300,000 users to break even on that. And that's vastly underestimating the actual costs. Data collection would actually be a little bit more involved than just pushing a button and data flow out and money flows in.
Feeding ads is much easier. You only need a lightweight back-end then, since the ads are served directly from your affiliate sites. Also they don't have to collect data about you to target ads. if you're playing an RPG about dragons for example, then the developers can charge more to advertisers in general that are relevant to that stuff. Just what the app
is is relevant data for targeting ads, and being more specific than that may be diminishing returns. This would in fact be far easier: just make a ton of apps that are themselves niche, and feed each of them different ads targeted to the specific niche of that app.