AI apps are experiencing high growth and high churn, but these are closely related. The abundance of new apps creates excitement, but eventually, the party will end. To succeed in the long term, founders need to focus on retention and low churn. Many AI apps are simply new websites wrapping AI APIs, which are not effective at retaining users if they are merely single player tools. Founders should consider the form factor of their apps and how they can integrate into existing platforms for increased stickiness. Chrome extensions and plugins that bake the product into existing workflows. Or build replacements for existing apps to take over muscle memory Network effects are crucial for AI apps to succeed in the long term. Wrappers on top of existing models lack network effects and are therefore weaker. You need other users that notify you, as social apps and collab tools do As the market progresses, AI apps will face slower growth and lower churn as novelty effects go away. We saw it on mobile apps and web3, that maturity means less novelty. Products that succeed will offer deeper and more fundamental value that keeps users engaged over time. We’ve seen similar waves of innovation before, like web 2 and mobile apps. Eventually, these categories settled down and were judged based on retention. Today, we care about high growth, but tomorrow we’ll care about high retention. Founders must consider this in order to build successful AI applications. Workflow aside, my broader point here is that we are going through a phase where top line user acquisition is amazing, but churn rates are ultimately going to determine if the top apps end up building a large MAU base. And I theorize that ultimately the best apps will need to have network effects. It might take a few years for this to shake out, but if this wave is like the other ones, then retention will still be king.

Do you have a newsletter?