How content experiments boosted rewarded ad views by 40% and IAPs by 20% in a six-year-old hyper-casual puzzle game

Hidden Objects is a hyper-casual hidden object game from the studio Diesel Puppet where players must find all quest items in a location within a time limit. The project is now in its seventh year, which, by hyper-casual standards, is close to retirement. 

A year ago, we released Hidden Objects 2: a game with the same core mechanics, but richer content and additional equipment. At that point, it seemed logical to gradually phase out the original Hidden Objects. However, we decided not to rush that decision and ran several additional iterations instead. One of those decisions ended up delivering an unexpected boost across nearly all key metrics, including in-app purchases.

Content update

In the release version of Hidden Objects, each level contained five quest items. Players completed levels in under a minute and regularly complained that everything happened too fast. Reviews often said things like: “The locations look great, but there’s barely time to actually see them.”

This also created a cost-effectiveness problem. Locations are expensive to produce, yet players spent less than a minute in each one. There was no real incentive to buy anything or even engage with rewards. The levels were too easy, meaning players didn’t need extra time, hints, or assistance — exactly the opposite of what we need when our long-term strategy is to systematically increase the share of IAP revenue.

Fast level completion also hurt our metrics. Over time, R1 steadily declined as more experienced puzzle players entered the game. For them, the difficulty felt trivial. The project had clearly stagnated, and something had to change. The most obvious solution was to increase the number of quest items per level.

Before touching content, we had to do a fair amount of technical prep work. Each location needed additional objects placed manually. We also had to rework our FireBase configuration so that eight quest items could be highlighted instead of five. After that, we prepared setups for testing 10 and even 12 items per level. Most old locations already contained up to 60 items in total, which gave us enough visual variety, but we’ve also increased the number of items on newer locations that weren’t supposed to have 60 initially. The next challenge was UI polish, making sure the bottom panel with quest item previews remained readable and visually clean as the number of items increased. All in all, this preparation phase took about a month.

We started by increasing the number of quest items from five to eight. The results were immediate: reward views increased by more than 40%, LTV grew by 20%, R1 increased from 28% to 30%. Encouraged by these results, we kept eight items as the new baseline.

We also saw a clear increase in session duration and overall playtime. Players now spent more time in each level, later-on retention improved, and the game finally started offering a meaningful challenge. As a result, most metrics showed sustained growth over time.

Next, we tested 10 quest items per level. We expected a smaller uplift, but instead LTV increased by another 15%, retention grew by 1-2%, and reward views jumped by another 40%.

We then experimented with 12 quest items per location, but ultimately abandoned that option. While the metrics did improve slightly, we quickly realized that content production wouldn’t keep up. At that volume, we simply wouldn’t be able to deliver new locations at a sustainable pace.

In the end, we settled on 10 quest items per level as the optimal balance instead of 5. This change increased total LTV by more than 30%, raised R1 by 2-4%, extended average playtime, and nearly doubled rewarded ad views.

We also rolled out the changes specifically to long-term users — players active for more than 60 days — and saw almost a 2x increase in LTV within that cohort.

It’s worth noting that these weren’t the only updates during that period. We also added some nice-looking animations, significantly improved search previews, used heatmap analytics to adjust locations, and introduced bonus levels after every 10 regular ones. These levels helped diversify the game’s mechanics, among other things.

Nuances of preparing hidden object levels

At first glance, it might seem that if increasing the number of quest items consistently improves metrics, we could just keep increasing them indefinitely. In reality, content pacing becomes the main constraint. Currently, each Hidden Objects location contains 60 items, 10 of which become quest items. That means a single location can be replayed six times, which results in roughly 1,000 playable levels set in a loop.

Even that isn’t a huge number. Some players complete two to three thousand levels. But if we keep increasing the number of quest items per level, the total amount of unique content players experience will shrink significantly.

Interestingly, players don’t react negatively to repeating locations. They’re focused on finding items and often don’t even notice the repetition. We specifically monitored churn after players ran out of unique content, and didn’t see a spike. That said, if repetition becomes too obvious, the situation could change quickly.

To address this, we’re actively using AI tools to speed up location production, an approach that already proved successful in Hidden Objects 2. We’re also gradually transferring stylistic solutions from the second project into the original one. Our long-term goal is to reach 12 quest items per level, since competitors typically run 15. But before that, we need a faster and more scalable content production pipeline.

We also experiment heavily with funnel structure — reshuffling levels and testing different entry points. This even spawned an internal meme: the curse of the first level. For two years, we tried to replace the first level with newer, more polished versions. Every single time, retention dropped noticeably. Somehow, the original first level, despite being visually weaker, consistently performed better.

Since then, we treat the first levels of the funnel with extreme caution across all projects. Changing them is risky, expensive, and difficult to roll back. Every funnel change requires a new release, and rolling back a live update is not always an option.

The project’s future 

The release of Hidden Objects 2 didn’t mean we abandoned the original game. Quite the opposite. We’re actively transferring best practices from the newer project: not only locations, but also monetization ideas. For example, Hidden Objects 2 performs very well with ad-skip ticket sales, while the original game doesn’t. One hypothesis is that Hidden Objects simply has fewer natural points where such tickets feel useful. Historically, HO1 had very few offers — mostly subscriptions, coins, and hints. 

We’re also refreshing older locations and moving away from overly dark color palettes. Player feedback clearly showed that early locations felt bright and pleasant, while later ones became too gloomy.

By experimenting with both content quantity and quality, we’ve unlocked strong growth even at this stage of the project’s lifecycle. This allows us to scale user acquisition, reinvest in ads, and grow the IAP segment through added challenge. In upcoming updates, we’ll introduce new offers: booster packs, hints, and ad-skip tickets.

Today, Hidden Objects shows year-over-year LTV growth of over 30% — an exceptional result for a six-year-old hyper-casual game. Most importantly, we’ve significantly increased the share of payments in total revenue, making the project far less dependent on advertising.