After working with dozens of mobile products, one pattern shows up again and again:

Monetization rarely fails because of the model. It fails because the model doesn’t match how users actually behave. 

Paid Apps:

We’ve seen paid apps work best when the user arrives already convinced.

One example: a highly specialized professional tool built for a narrow audience. Users didn’t discover it through ads or app store browsing, they arrived via recommendations. 

Because the value was obvious and immediate, charging upfront filtered out the wrong users and reduced support noise.

In this case, monetization wasn’t about maximizing downloads. It was about attracting the right users.

Paid models succeed when trust exists before install.

In-App Purchases: 

Across multiple consumer apps we’ve worked on, users rarely consider paying when they first download the product.

The core experience is free and usable. Payment becomes relevant only at specific moments, when users want to move faster, get another attempt, personalize the experience, or bypass friction that suddenly matters.

It’s important to present these offers exactly when the user is motivated. Compared to subscriptions, in-app purchases require no long-term commitment, which makes the decision easier for users.
 

In-App Advertising: 

Advertising often looks simple, but results vary widely depending on how users interact with the app.

We’ve seen apps with large download numbers generate limited ad revenue because sessions were short or infrequent, and ad placement was poorly balanced: either too aggressive, leading to user churn, or too subtle to have any meaningful impact.

With this model, app usage frequency and the amount of time users spend in the app are crucial.

Ad format selection also plays a significant role.

• Banner ads are easy to implement but typically produce low revenue
• Interstitial ads generate higher returns but require careful timing
• Rewarded ads perform well when users receive clear, optional value
• Native ads integrate better into the interface but require design effort

Because the impact of ads is highly context-dependent, teams that succeed rely on experimentation. A/B testing different formats, placements, and frequencies helps identify what increases revenue without negatively affecting engagement.

Advertising performs best when it aligns with existing user behavior rather than interrupting it.

Freemium: 

Its primary purpose is to let users experience value early, then guide the right users toward subscriptions or in-app purchases. This approach works especially well when the goal is to drive high download volume.

We’ve seen freemium fail most often when the free tier delivers so much value that users never encounter a reason to pay. Engagement remains high, but monetization stalls.

Freemium isn’t about restricting access. It’s about creating a clear progression toward paid value. When done well, freemium qualifies users: casual users stay free, while high-intent users naturally move toward paid options.

Subscription: When the App Becomes Part of the Routine
 

Subscription success almost always comes down to timing.

We’ve seen subscription apps struggle when users are asked to commit before value is clearly established.

Paywall placement should be chosen carefully and based on user data, not assumptions.

Paywall design also matters. In our experience, the best outcomes come from A/B testing:  testing layout, wording, and pricing structure, including single-offer versus multi-offer paywalls.

We strongly recommend integrating tools to track and measure results, as the most effective monetization decisions are data-driven.

What We’ve Learned Across All Models

The strongest apps  start with monetization:

• Understanding user intent
• Observing usage patterns
• Designing value around real behavior

A revenue model only must follow real user behavior.

The most common mistake isn’t choosing the wrong model, it’s locking into one too early, without testing or data.

Monetization isn’t a fixed decision.  It evolves through experimentation, tracking, and iteration.