Lesson: Hooked with Nir Eyal

Step #7 Framework: Your most engaged users pave the road for more

So there are really five fundamental questions that, if you're looking to build a habit-forming product, you can ask yourself, does your idea, before you've built anything, map to this design pattern? The first question to ask is the internal trigger. What's the itch the user is coming to scratch? Then the external trigger, what's the thing that gets the user to use the product to take the intended action? The action is what's the simplest behavior done in anticipation of reward? Reward is does the user get what they want and does the product leave them wanting more, this variable reward component? Then finally, the investment—what's the bit or work the user does to increase the likelihood of the next pass?

Whether you have an existing product and you're seeing, “You know what? It's just not very engaging. It's just not forming a habit.” You can look at the hook canvas; you can look at these five fundamental questions and see if your product might be deficient in one of these four categories. And if you're building a product from scratch, before you build anything, and invest all that time, money, and effort into building something, you can do the same. You can look at the hook canvas and say, “Do we have these four fundamental components? And, if not, how do we patch that hole or make the existing part of that hook better?”

And then it's just a matter of, we still have to iterate, we still have to build the product. We give it our best shot with the hook canvas in mind and then it's about building, measuring, and learning. Then we're testing, we're figuring out how to optimize based on what we just built.

The best we have today is cohort analysis. You change some aspect of your product. You've got this cohort that now has this new feature. Hopefully, it's a subset of your existing user base and then you're testing to see is that user base more engaged? So you're coming up with some kind of metric based on how often you think your product should be used.

Some products are not used intradaily. That's okay. But then if it's a product that should be used once a day, that's what you're tracking. If it's a product that should be used once a week, that's what you're tracking. Whatever it is that you think is the frequency by which the user should use the product, that's going to be your number that you're tracking based on how you're changing it. And tracking how each cohort changes and to see if they're more engaged. Is that number tracking the percentage of users who are using the product weekly, daily, intradaily increasing based on your change in the product?

Another place to look is to do what I call habit testing. Habit testing is this process that I didn't invent. I kind of gleaned it from people in the industry, including Josh Elman and a few other folks, who basically use this process of trying to figure out who are your most avid users? Trying to figure out who is the user who is using your product already? This will only work, of course, if you have an existing product. Who uses your product a lot? What are they doing on that site that makes them use the product a lot? And most importantly, what's the path they took to become so active?

So in the story that Josh Elman tells when he was a product manager at Twitter, was that they noticed that people who followed others within the first few sessions of using Twitter, those were the people who became habituated. Those were the people who used the product every single day. So now, when you use Twitter, when you onboard to Twitter, guess what everybody on Twitter's going to do? Twitter's going to shape your path so that you also follow people right away. They've learned that based on what these first few users who were rabid users did, now everybody takes that same habit path as I call it. Everybody now has to follow a few more folks.