At some point you draw a line through your app. This side free, that side costs money.
Draw it in the wrong place and one of two things happens. The free version is so limited nobody sticks around long enough to want more, or it's so complete that nobody ever needs to pay.
Gate depth, not function
The most useful principle here, and the one most often got wrong.
Free users should be able to do the core thing your app is for. Completely, and properly. What they hit is a ceiling on how much, how often, or how far.
A note app that lets you write notes but not save them has gated the function. Nobody converts, because nobody experienced anything worth paying for. A note app that gives you fifty notes free and charges beyond that has gated the depth. The people who hit fifty notes are exactly the people who've proven the thing is useful to them.
The paying customer should be someone who ran out of room while getting value, not someone who never got value at all.
Never gate the habit
A specific version of the same idea, and worth stating separately because it's easy to violate by accident.
Whatever brings people back daily, don't put it behind the paywall. That's the thing creating the attachment that eventually makes paying feel reasonable.
If someone uses your app every morning and one day the morning thing asks for money, you haven't converted them. You've removed the reason they were showing up and made them notice how easy it'd be to stop.
Charge for more of it, for extras around it, for the power-user version. The daily core stays open.
Freemium isn't the safe choice
Most first-timers pick freemium because it's what they see everywhere. It's worth knowing what it actually demands.
Free users cost you money. Servers, storage, support emails. If a large majority never pay, and they won't, the small paying group has to cover everyone else.
Freemium works properly when free users create value beyond themselves. They invite others, they make content, they're the reason the app is worth joining. If your free users are just costs with no upside, you've built something expensive to run with a small revenue base attached.
A straightforward trial is often the better shape for a first app. Everyone gets the full thing for a week or two, then pays. Simpler to build, simpler to explain, and the conversation about money happens while the value is fresh.
What people will actually pay for
Roughly speaking, people pay when something saves them meaningful time, makes or protects money, removes a real irritation, or is tied to who they want to be.
They rarely pay for slightly nicer. Cosmetic upgrades convert badly unless identity is genuinely part of the product.
Worth checking which of those your paid tier is offering, honestly. "More features" isn't one of them, and neither is "the version without the annoying limit we added specifically to annoy you," which people can always tell.
Keep it to two options
New builders often design four tiers with a matrix of features.
Two is nearly always better for a first app. Free, and paid. Maybe monthly and yearly as the only variation.
Every extra tier is a decision you're asking someone to make, and decisions cause hesitation, and hesitation loses sales. You can add complexity later once you know what people actually want. You'll be guessing if you do it now.
A quick story. My app's plan had a founding-member tier: a one-off lifetime price for the first five hundred people, meant to bring in some money early and create a reason to act.
I never built it. The waitlist stayed free the whole way through.
So the thing I learned was that people would give me an email address for something free, which tells you almost nothing. Whether anyone would pay, and how much, and for which version, stayed completely unanswered through months of work. The free line was easy to draw because it never had to survive contact with anybody's wallet.
If you want help drawing the line
That last check is worth doing sincerely. It's uncomfortable and it's better than finding out from a conversion rate six months later.
Next: Picking a price →
Go deeper: Chapter 25, Monetization Models