Written by Nir Eyal.
Apple. Facebook. Twitter. Google. Pinterest. These companies all have one thing in common — they create habits among their users. People use these products habitually on a daily basis and they’re so compelling that many of us struggle to imagine life before they existed.
But creating habits is easier said than done. Even though I’ve written extensively aboutbehavior engineering and the importance of habits, there are few resources to give designers the tools they need to design and measure user habits. That’s why I wrote “Hooked: How to Build Habit-Form Products,” which explains the four-step Hook Model that can help form habits among your users.
In this post, we’ll take a look at how you can use Habit Testing to build products that users not only love, but are hooked on.
The Hook Model
Before we take a look at Habit Testing, it’s important that we familiarize ourselves with the Hook Model.
As described in my book, the Hook Model is a four-step process companies use to hook users and form habits among them. The four parts of the model include: trigger, action, investment and variable reward.
Image Source: Nir Eyal
Let’s break that down:
- Trigger: This is the spark for a behavior that gets someone into a system. There are two types of triggers: external and internal. The external trigger alerts users with something like an email, link or icon. Internal triggers happen within the system and are formed as a user cycles through successive hooks while using the product.
- Action: This is the behavior taken when a reward is anticipated. For instance, the action of clicking on an image on your Facebook feed. When you click that image, you anticipate that you’ll be taken somewhere interesting, such as the latest listicle on Buzzfeed. This is an important part of the model because it draws upon usability design to drive users to take action.
- Variable Reward: This is the part of the model that allows you to create craving in users. Rather than using a conventional feedback loop, you can serve up a multitude of potential rewards to hold a user’s interest. For instance, Pinterest does this by showing you images that are relevant to your interests along with other things that might catch your eye.
- Investment: This is the part where the user now has to do some work. Think of it as giving back into the product, and that can take the form of time, data, effort, social capital or money. But this isn’t solely about swiping their credit cards. Investment is an action that will improve the product as well — such as inviting new people into the system, giving feedback on features, etc.
The Hook Model is a great way to start consider how you bring users into a system, hook them and keep them there through variable rewards and internal triggers. In the end, you’ll be able to build habits among your users that will have them coming back for more.
An Introduction to Habit Testing
Habit Testing fits perfectly into the Hook Model.
With Habit Testing, you’re better able to answer three vital questions:
- Who are your devotees?
- What part of your product forms habits?
- Why do those parts of your product form habits?
Image Source: Nir Eyal
A prerequisite to Habit Testing is having some kind of product up and running. Of course, it’s a good idea to complete your business model hypotheses and explore how your product willcreate user desire before launching even a minimal viable product.
Once you’ve released your site or app into the wild, you must start reviewing data. Don’t track everything: focus on when users interact with your site along with the path of their visit.
Let’s go through the three steps of Habit Testing.
Step One — Identify
This is where we look at the first question of Habit Testing: “Who are the habitual users?”
Here’s some tips to do that:
- Define the characteristics of a devoted user. Ask yourself how frequently someone should use the site, assuming that most bugs will be resolved and the product is nice and polished.
- Be real with yourself. If you’re working on a social network app like Instagram, you’d expect habitual users to be on the app many times a day. But if you’re building a movie recommendation site, you might not expect people to visit more than three times per week
- Crunch the numbers. Once you know how often someone should use your site, check the data to see how they stack up against expectations. Create a cohort analysis as a baseline for measuring upcoming iterations.
When defining the frequency of use, don’t be overly optimistic by thinking only of super users. Aim for a reasonable guess. For example, you could average out the product use between yourself and your coworkers.
The more frequently someone uses a product, it’s more than likely they’ll form a habit.
A good way to measure whether someone will actually use your product is to see whether your own team is using it. Take for example Twitter. The social media giant was born within Odeo — the original company Biz Stone and Jack Dorsey founded. They knew they were on to something with Twitter because the engineers couldn’t stop playing with it.
In any case, you’ll hopefully have at least a few users who interact frequently enough with your product for you to call them devotees.
Editor’s note: To learn more about how to influence behavior with design, check out the free e-book Interaction Design Best Practices.
Step Two — Codify
The next step is to codify the steps devotees took using your product so you can understand what hooked them.
First, how do you know you have enough devotees? A safe guideline is roughly 5%.
Keep in mind, however, that your rate of active users will need to be much higher to sustain your business. If at least 5% don’t find your product useful, you have a problem. If you have that 5%, then it’s time to find the Habit Path, a series of similar behaviors shared among your most loyal users.
Not every user will interact with your product in the same way. Each one will have a unique data fingerprint, which will reveal patterns in usage and that in turn will help you discover the Habit Path.
Let’s go back to Twitter. The social media app discovered that once new users followed enough other members to odds of more people using their site increased.
Here’s how you determine which of the steps in the Habit Path were critical for creating devoted users:
- Create a hypothesis. Where was the turn that turned passers-by to devotees? This could feel like assuming causation from correlations — but in the murkiness of launching a new product, it’s the best you’ll have.
- Talk to users: Learn how they use the products and why they use it. For instance, you can walk them through the actual steps that got them hooked.
Step 3 — Modify
Now that we have a hypothesis, it’s time return to the build. We need to measure, learn and take new users down the Habit Path we’ve discovered.
Image Source: Twitter
For example, leveraging their Habit Path, Twitter’s onboarding process suggests news users who to follow. You’ll want to do something similar once you’ve identified the path that turns onlookers into devoted users of your product.
Habit Testing is an ongoing process that applies to every new release. Continually comparing new user activity to devotee usages can guide you on how to evolve and further foster habits.
Products fail when they don’t adequately create user habits. And if a product doesn’t create a habit, then it’ll get left behind in the ever increasing distractions we all face.