Keeping an app fresh, relevant, and useful can be a major challenge after you have launched. Nothing worse than watching your usage diminish after working so hard to get people to download and try the app in the first place.

Apps like Uber, Venmo and Instagram have proven to have staying power – but sometimes it is easier to learn the lessons of those that failed rather than those that succeeded. Consider Pokemon Go. When the app was released, it was a huge success. It was innovative, social, and – for fans of the 90’s cartoon – nostalgic. It wasn’t hard to find users walking around your city with their heads down trying to catch Charmanders and Jigglypuffs. In July 2016, Pokemon Go was the top selling mobile game on the market and quickly became the most active game ever, reaching over 40 million users. It was a big, fast success. It was the first to combine augmented reality, geocaching, and the ability to interact with other users in realtime (by “battling” at gyms) throughout a user’s city. The app even motivated other tech developers to spin off into other ventures: third parties created apps to help players track Pokemon and connect to each other in the game.

It is not to say that the app is still not a raging success – but the number of users dropped from nearly 40 million down to 15 million by October. And revenues fell at an even faster rate. They are still making money, but it is nowhere near the peak it had in those early days. Many of the game’s issues were related to the never seen before growth of the app, but there are definite lessons organizations can learn.

So, why did Pokemon Go lose their users and what can you learn from them?

Pokemon Go released many of its features at once and then, for several months, made few changes beyond bug fixes and some cosmetic updates. People got bored. Once they had found their favorite Pokemon and spent a few afternoons scouring the city’s monuments with friends, many users deleted Pokemon Go from their phones.

When creating your app, have a long-term plan to keep engagement up. You want to initially release a minimum viable product (MVP): get your core features right, but wait to release more. Take input from users, look at usage data and see what features get the most use and then start releasing items that enhance your MVP. Write down the great ideas that come up in the process and set a list of which one you will release next.

2. Listen to your users and (quickly!) implement changes based on what they are saying.

Pokemon Go failed to be meaningfully active in its community. Users discussed the app’s shortcomings on Facebook, Reddit, Twitter, and other forums, but Niantic made few changes in response to their complaints. Where changes were made, they were made too slowly, and momentum had already stalled. For example, the app has recently made some changes with regard to tracking Pokemon based on feedback, but the changes took months to implement. By the time this feature was updated, many users had already deleted the app.

To avoid this pitfall, listen to your users. Have a plan in place for how to prioritize the most important app updates and make those happen quickly. Some organizations don’t have the bandwidth or resources to spend hours and hours fixing issues with the app. So, as leaders in organizations, foresee updates for your app and consider how you will account for them.

By rolling features out on a schedule (see #1, above), you are more likely to have a manageable workload. You can roll out each big feature and consider any problems or feedback before moving on to the next one. If you are outsourcing app development, create your budget with this in mind. You will want to have some funds set aside for making changes to the app once it’s been unleashed. We typically recommend 15-25% of your total budget be earmarked for making updates and fixes after the app is live.

3. Be careful with user’s data: don’t lose it or compromise it.

Another place that Pokemon Go went wrong was with an early bug that lost user data and caused millions of users to have to restart their progress on the game. By losing user data, Niantic frustrated part of its fan base and lost user confidence in the game.

If you don’t want to undermine your own app’s success, be mindful of how you use data. First, during updates, don’t lose it. If there is a chance user data might get lost, reach out to users in advance, let them know the update is happening, and advise them to save their data by creating an account if they haven’t already. Being proactive about any potential issues can go a long way in building goodwill.