Uncovering Anti-Retention Patterns
A Guide to Identify Hidden Roadblocks to Retention and Build Stronger Habit-Forming Products
If you are like most growth professionals out there looking to measure activation, you are likely thinking about analyzing which feature has a 'positive' correlation with retention. However, great growth professionals approach these problems differently than most do.
Focusing only on the positive correlation between aha and retention won't get you a complete picture. It’ll probably even hurt you. Why? Because there might be some features in your product that, if used too early, might break activation and even retention! At HyperGrowth Partners, we call these anti-retention patterns.
Take a real-life example from my time as SVP of Marketing at Auth0. To experience their aha moment, Auth0's customers needed to sign up and implement the SDK to get their users to log in. But they need more than that to build a habit with the product; that takes time.
We then realized that some people were using Multi-Factor Authentication (MFA) within their first week, which is a significantly more complicated feature to understand than others. Diving into the data, we learned many users got stuck trying to figure out how to use it.
And because of this, a staggering 50% of those who tried MFA in their first week didn't retain eight months later. Had they not tried it so early, these cohorts could have retained and even drove revenue. But that didn't happen because they used a feature "too soon" based on their product knowledge.
So how do you identify these anti-retention patterns? And once you do, how do you rethink your activation and retention to avoid their side effects? In this post, we'll reveal this and more.
What you should know about retention
Modern literature breaks activation and retention down into a linear UX of four steps:
Sign up: the steps for new users to create an account with your product.
Set up: the info signed up users must share to be 'set up' to experience the 'aha.'
Aha: the product experience that guides set-up users to experience core product value.
Habit: signs that show users return to the product to repeat the aha moment, exemplifying that they built a routine with it.
According to this literature, the most proven way of measuring activation is to:
Pinpoint 2-3 core events that exemplify your aha moment
Analyze which one has the highest positive correlation with retention
Design a UX with the path of least resistance to perform that aha feature
As simple as this process sounds, it's not. But it might also produce hidden side effects that will work against activation and retention.
Linear frameworks create blind spots. While literature strives to model reality by simplifying journeys in funnels and loops, users break these frameworks exploring features in parallel. That's why linear models create blind spots for identifying ancillary events that negatively affect retention, our not-so-dear anti-retention patterns. This is especially common in SaaS platforms with horizontal use cases.
Over-exploration can break activation. Even if most SaaS comes with an onboarding experience, users still have the freedom to explore. If you give them too many options, they might initially try something too hard to process, breaking their activation flow. It becomes then critical to guide product exploration intentionally to de-risk anti-retention.
Activation takes time to build, but it can break instantly. Activation can take 1-2 weeks to build, whereas anti-retention patterns can break it in seconds. This is precisely why it's vital to identify these negative patterns as much as the positive ones because they could be much more detrimental in achieving activation and long-term retention.
Retention starts with activation, but it doesn't stop there. While most teams focus on the signup and onboarding flows to work out activation, they often overlook what comes after the aha moment, ultimately driving long-term habit-building and retention. As we've seen with Auth0, anti-retention events can also arise in that part of the experience! Even once users are activated, they can still churn if they hit the wrong feature too soon. That's why it's essential to keep an eye out for these patterns across the entire user experience, factoring in ample lifecycle timeframes of 3-12 months post-activation.
Identifying anti-retention patterns
Most of today's B2B SaaS comes with tens of features, especially those working on bundling various use cases as platforms. Think Notion, Figma, and Hubspot. Despite the number of features, only 2 or 3 will often contribute to solid retention, depending on their personas and platform users.
Also, as you become more proficient with the product, the features that retain you change. For example, when you start using Notion, your aha feature might be how easy it is to build simple wiki pages to tell the rest of the company what your team does. But once you're more proficient with the product, you might be retained because of their database feature, which allows you to update and share information across multiple wiki pages from only one place.
Time matters. Product expertise matters.
And it's primarily for these reasons that all the other features can virtually become distractions to new users trying to activate. That's also why, if you're early-stage, you should go to market by nailing just one core use case — ensuring you're doing it 10x better than the competition — and then layering more as you scale.
Activation and retention are learning curves
Regardless of your company stage, treating activation and retention as a learning curve would be best. After signing up for a new product, every user becomes more proficient with your product as time passes. This journey is a give-and-take game of investments and returns:
Effort: Users invest their time and effort trying to understand how to use the product.
Value: In return for their efforts, they expect growing value from using the product.
When users sign up and get set up with the product, they need considerably more effort for small amounts of value in return. But as they progress through their journey, the value experienced during the aha moment should pay back all the effort invested until then. Successful retention means that as users become more proficient with the product, they also get comparable greater value for less effort.
As a growth professional, your task is to guide users through this learning curve by feeding them enough value that compensates for their effort investment, especially early on. But it would be best if you kept doing so every step of the way until they built a habit.
And to do so effectively, you need to identify the anti-retention features throughout the journey (e.g. red dots in the chart) so you can stir users away from them as they try to activate. On the other hand, you want to keep them focused on pursuing the activating features to ensure a positive balance of value over effort.
As time goes by and users become more proficient with the product, the anti-retention patterns shrink while the activating patterns expand. Depending on your product, you could keep your UX/UI more focused at the beginning when users are less expert and have experienced less value, and progressively release it as they gain more confidence and receive growing value from the product.
Using data to spot anti-retention patterns
To identify your anti-retention patterns correctly, let’s take the Auth0 example mentioned above. We changed the actual data here, but it reflects what happened.
The table compares two cohorts of users after one year from signup. The table on the left-hand side shows users with 50+ MAUs, which we consider retained. The table on the right-hand side shows the ones with 0 MAUs deemed churned.
For the two cohorts, each table analyzes when users tried different features for the first time — during their first hour, first day, week, and so on — and their % impact on retention.
As you know, percentages matter, but you should compare them with absolute cohorts to ensure that numbers are large enough to make an educated guess.
The analysis pinpoints in what timeframes retained users implemented certain features to understand which ones correlate positively — the red arrows on the left-hand side table.
Things get more interesting on the right-hand side table, highlighting the anti-retention patterns, which define at what point specific features negatively correlate with retention in the journey.
As mentioned in the intro, what comes across from the analysis is that:
38% of users who used the MFA feature on the first day didn’t retain after a year
15% of users who used the first non-try feature in the first hour didn’t retain after a year
15% of users who used a custom database connection in the first week didn’t retain after a year
You can then use these insights to:
Understand how many retained users you’re losing because they used a feature too soon
Devise when is too soon to have that feature in the learning curve
Plot incremental retention gains if you delayed that feature in the UX
Spot anti-retention patterns without data
If you're still early-stage or just starting to build your product, you should think about anti-retention patterns too, but you need to think creatively since you have no data. You can use a scoring framework and update it once you have available data.
Here's a simple step-by-step process inspired by Darius Contractor's psych framework:
List all the potential events users can take across signup, set up, aha, and habit UXs.
Map how many users hit these events in critical timeframes of your journey.
Score each event based on the following:
Effort: The higher the effort, the lower the score.
Value: The higher the value, the higher the score.
Use case frequency: higher frequency features correspond to a higher likelihood of users building a habit with them and retaining more; the higher the frequency, the higher the score.
Assign specific values to quantify the qualitative metrics defined above. For example, 1 for low, 3 for medium, and 9 for high. These qualifiers follow a specific geometric projection, making the analysis easier. Remember to verify your assumptions by checking in directly with your customers through user research or product recordings.
If the effort for that feature is high, consider pushing it down the line, as long as you have something with a low effort that you can have earlier. If you have no low-effort features, you must invest in content like FAQs or tutorials to help users navigate that specific learning curve.
If you're ready to gain deeper insights into your retention, give us a call at HyperGrowth Partners. We'll match you with an expert from our partners' roster and provide the guidance and support you need to succeed.
Overcoming anti-retention
So how can we build a better retaining product and set up-to-habit UX based on anti-retention patterns?
Keep the setup & aha UX in focus mode
Remember. New users expect value soon in their journey and with minimal effort to learn how to use your product. To deliver an outstanding experience, keep them focused where you know they'll get it.
Hide anti-retention features and related UI. Prevent them from going down multiple rabbit holes that you know will hurt their attention levels. Hide anti-retention features until they have formed strong habits with your core use case. Only then, make them available and discoverable.
Recommends the "next best action." Based on your anti-retention analysis, craft a sequence of events in a specific order that sets users up for success. Prioritize features and events that offer a higher balance of value and attention while aligning with your core value prop.
Introduce time and action requirements before 'releasing' the complete UX. While this might add some friction to the most explorative users, make them aware that you're simplifying their new user experience to set them up for success.
Use pulses, pointers, and product tours to keep users focused. These tactics could be less forceful in the UX and result in more native, depending on how technical your product and personas are.
BONUS: gamify the experience!
Consumer apps like Duolingo have led the way to gamify learning with streaks, quests, leaderboards, in-app point systems, and more. And no one said you couldn't do the same in B2B SaaS!
Armed with your insights on anti-retention patterns, you can gamify your set-up-to-habit UX to give them additional attention and delay their expectation of value later in their journey.
Some tactics to get you going:
Make the aha moment a must-hit event before 'unlocking' additional features. When you position anti-retention features as 'unlocks,' you indirectly educate users that those are more advanced ones that can be accessed only when you master the basics.
Reposition your learning tasks as games. When you label learning tasks as personal or social challenges or quests that unlock XPs (experience points) or other in-app rewards, you'll appeal to the user's intrinsic desire for self-mastery or recognition. This tactic creates focus, compounds their attention, and delays the expectation of product value.
Use scarcity, countdown, and social elements (e.g. comments, kudos, and leaderboards) to drive users away from anti-retention events and focus them on building habits on the core use cases you know are contributing more to retention.
There's no definitive formula for this. Use these tactics to brainstorm and test a few experiments with your unique product and audience.
Closing thoughts
In conclusion, understanding activation and retention is essential for growth and revenue in any product. While pinpointing core features with the highest positive correlation to retention is a standard practice, more is needed to identify the anti-retention patterns that can negatively affect activation and retention.
Anti-retention patterns are features that, when used too soon, can break or even reverse activation and retention. Growth teams must dig into their data and map users' learning curves to identify the related features. This knowledge will enable teams to optimize their user experience to avoid these side effects and build stronger habit-forming products.
Editor’s Note: This article was written in collaboration with
, who assisted in bringing to life @Gonto’s experience and insights through his writing.