Is your growth problem actually a retention problem?
Driving exponential growth is every business’ dream. Growth is so much at the center of focus that some overlook the impact of retention on driving exceptional growth. When someone mentions having a growth problem, we advise checking retention first:
“Are you sure you have a growth problem and not a retention problem?”
Without strong retention, you have a leaky bucket; you simply can’t scale. As each cohort churns, you lose momentum, stalling growth.
Fixing retention is your first milestone in building a scalable business.
How can you go about solving a retention challenge? There are two main approaches: design (before use) and analysis (after use).
We will focus on the design approach in which you build retention into your product using loops and other means. Great design will reduce the need for costly re-build and save you valuable time, especially critical for startups.
The different, equally beneficial, analysis approach is using tools to conduct user behavior analysis (e.g. cohort analysis) to determine when your customers “stick” and retain. However, this is harder to do in the beginning when you lack data.
The best approach to excellent retention is to start with design and then optimize using analytics.
By the end of this article, you will know how to build retention loops which are a lever and prerequisite to scaling your business. We’ll share examples throughout to get you thinking about how you can make retention loops for your business.
To provide an overview and some orientation, we have broken down retention along the user lifetime into 3 phases. Each phase has its challenges and problems as described in the figure. In this article, we focus on the middle part: habit creation.
The 3 phases of retention:
- Start: Onboarding and value (wow-moment) discovery.
- Middle: Habit creation and dormant user reactivation.
- End: Churn prediction and delay.
All phases are crucial for achieving strong retention: from activating that user to driving a habit to preventing churn.
After successful activation (first use) and onboarding (bringing them to the wow-moment) in phase one comes the core discipline of retention: getting users to use your solution frequently — ideally creating a habit of use for your users.
To be clear, this is a challenging task, and only a few do exceptionally well. Even less are successful in creating real habitual use. Unicorn companies are masters in creating habitual usage of their products.
Before we enter into how to do this, allow us to set your expectations by looking at some benchmarks. World-class retention rates start at 25% (media & finance apps) to 35% (e-commerce) eight weeks after activation, based on a 2019 Mixpanel benchmark report.
It is critical to understand that:
Habits are created by adding triggers to an existing behavior which then are rewarded in such a way (through dopamine spikes in anticipation) that they form a new additional habit.
Allow us to get a little technical for a moment and stress two crucial aspects:
- We create new habits by latching on to existing ones. Meaning we add rather than create isolated new behaviors.
- The reward turns a mere action into a habit by creating a dopamine spike in its anticipation. The highest dopamine spikes and most addictive habits are created using variable rewards, as perfected by the gaming and gambling industries.
“Dopamine is not about pleasure; it’s about the anticipation of pleasure. It’s about the pursuit of happiness.” — Robert Sapolsky, professor of biology and neurology, Stanford University
The loop of habit creation is also known as operant conditioning:
The elements we see to creating habit loops are:
- Cues, which we call triggers.
- The desired action we want the user to take.
- A reward that follows when the desired action has been taken.
Passing through this cycle of trigger -> action -> reward multiple times reinforces the habit creation.
To increase retention even further, some form of “lock-in” is added in addition to habit loops.
There are 3 main forms of lock-in which make it disadvantageous for the user to leave:
- Time and data investment (psychological).
- Creating switching costs by making it hard or uncomfortable (time and money).
- Straightforward platform lock-in (impossible due to closed systems).
Nir added the element of investment to habit loops in his Hooked model:
So let’s start building habit loops for your business.
The ultimate goal is to place an internal trigger on an emotion so that it automatically triggers product use. This works better with negative emotions.
For example, every time you feel bored, sad, or depressed, you open up your social network of choice, read some posts and engage with them. Just watch the people around you: a smartphone’s average daily pick-up rate is an astonishing 50 to 100 times for Gen Z and Millennials.
Here are some common negative feelings which could serve as an internal trigger:
Because it takes weeks or months until internal triggers are formed, we need to rely on external triggers first.
A trigger needs to be straightforward to capture the user’s attention and drive action. There are three key types of external triggers:
- Paid triggers (due to cost rather used for acquisition than retention) — All types of paid advertising or paid channel messages.
- Owned triggers (sustainable, controllable, affordable) — The red circle on an app icon, push messages, browser notifications, e-mail.
- Earned triggers (unpredictable and challenging to sustain meaningful levels) — Press, social media exposure, mentions in forums or platforms.
Through user research, you can start to understand what triggers occur and when. The end goal is to be able to fill in this template:
Every time users [feel internal emotion], they [first action of intended habit]. Until they feel the internal trigger, we trigger them using [external trigger] when [exact point in time in the form of existing action].
After you’ve triggered your user, you want them to take action.
When it comes to action influencing factors and creating habits, the Fogg Behavior Model is the one to turn to. B.J. Fogg found that:
Behavior = Motivation x Ability x Trigger
The obvious insight is that it takes a higher amount of motivation when something is hard to do and less motivation when something is easy to do. Also, you need a trigger; otherwise, no action will be taken, which we already discussed.
When it comes to motivation, it is derived from any of these three core motivators:
- Seeking pleasure and avoiding pain.
- Seeking hope and avoiding fear.
- Seeking social acceptance whilst avoiding social rejection.
Now we will let you in on a little secret. Ability and triggers are bigger obstacles to habit formation than motivation.
Ambition and motivation are not the obstacles to taking action and changing behavior. The real obstacles are lack of ability and missing triggers.
This confirms: the fewer actions necessary to arrive at the wow moment, the better. Or in other words:
Those who radically simplify products often experience radical growth.
Basically, any unicorn of the modern area is a champion in UX and UA design. But do not take it from us; take it from Evan Williams (cofounder of Blogger and Twitter) who realized that connecting a human desire with the shortest possible fulfillment is a receipt for success in business.
“Take a human desire, preferably one that has been around for a really long time … identify that desire and use modern technology to take out steps.” — Evan Williams, cofounder of Blogger and Twitter
Here is how to simplify your product.
There are six main areas that impact a task’s difficulty:
- Time: How long it takes to complete an action
- Money: The fiscal cost of taking an action
- Physical effort: The amount of labor involved in taking action
- Brain cycles: The level of mental effort and the focus required to take action
- Social deviance: How accepted the behavior is by others
- Non-routine: How much the action matches or disrupts existing routines
To reduce complexity, you need to map out every step and challenge it against the six factors that impact simplicity. Ideally, you try to cut out as many steps as possible between the first point of contact and wow-moment.
Next to making things simpler to increase ability, one can also utilize heuristics, i.e. shortcuts we rely on to operate efficiently, to increase ability.
For software, have a look at the 10 most common heuristics in User Interface design according to Nielsen Norman Group.
For physical products, take a look at the 40 design heuristics of 400 award-winning physical products.
In summary: set triggers, reduce complexity, ensure motivation is present, and use heuristics.
One note about solving retention for low frequency products
While it is easier for a high usage product like Facebook, how do you achieve this for lower usage products? How do you continue to get users to take action for them? We have one example for you.
Funda has identified this issue and attempted to solve it. Funda is a Dutch mortgage website that usually people would only use once every 5–10 years, basically when they are buying or renting a house. For products that do not have frequent use (minimum once a month), the utility needs to be very high:
So, what do they do? They offer a value checker where you automatically get notifications about the value of your house every month.
This allows them to remind you more frequently of their core product so that when you buy or sell again, you are more likely to use them.
For lower usage products it is about creating more moments to engage and to think of the brand beyond just a one-off usage, e.g. through providing relevant content or related products/services.
We talked about triggers and rewards in the beginning. A trigger is the start and the reward is the end of the loop. The reward is the differentiating factor between simple marketing and habit creation. Understanding and designing the right rewards is therefore critical.
Without a well-designed reward, there is no habit loop.
How do rewards work?
Rewards are fueled by the dopamine spike that occurs when we anticipate rewards, not when we receive them. The consequence is that a consistent reward becomes boring, but a variable reward keeps our anticipation and hence our interest high because the uncertainty increases the pre-reward expectation levels.
Variable rewards perform significantly better in habit creation than consistent rewards.
Type of rewards
Different rewards types can appeal to different emotions. The reward must match and align with the feeling connected to the trigger to work well.
Good rewards are designed to make us feel accepted, attractive, important, and included. — Nir Eyal, Hooked
Nir highlights three categories of rewards (with adaptations by us):
- Social: this comes in the form of likes, comments, shares on social networks, as a proxy for status and recognition as well as a feeling of being well-liked or connected to others. Examples include: expert or moderator status, premium or achievement badges, status or visible amount of likes, shares, points etc.
- Needs: the desire to hoard or obtain something as a sign of success or power like money, information, or physical goods drives this behavior.
- Self: this reward is different as it is more intrinsic. Based on a need to advance in skill, to master, or to complete. Take Fitbit that sends you a badge when you reach specific achievements, you end up wanting to beat your maximum steps for your own sense of achievement:
Before we move on a word of caution. Not all types of rewards work equally well. For example, offering monetary incentives is known to backfire because they are simply not meaningful enough to matter. Anything above some token appreciation and below the threshold of being substantial is perceived as an insult.
Money as a reward is simple: don’t use it,
unless it is a meaningful amount.
It is more sustainable and more effective to go for social, need or self-driven rewards.
There is a good reason you want to create some form of value-added investment for your users.
Gaining investments from your customers raises the barrier to leaving.
Like the IKEA effect, which shows that we place a higher value on furniture assembled by ourselves, we also value the product experience higher, if we invest time (or data) into it.
The more investment we put into a solution (or product) in the form of customization, data provided, data accumulated such as historical use, time spent with the solution, integration into other solutions, etc… the more likely we will retain and invest further effort to use and improve the solution.
When to trigger an investment
When is the best time to trigger an investment?
The ideal time to ask for an investment is right after having provided a reward — leveraging our tendency to reciprocate favors.
Some brands even use this investment up front, e.g. getting you to complete a quiz to identify which products best suit you. Because you have spent time investing into a product by answering those questions, it makes you more curious about the outcome and therefore more likely to complete an action.
Another example is creating your own profile, like on LinkedIn. This serves as ability increasing orientation with a clear next step and as a reward by giving you blue progress feedback bars with a status (here: Intermediate). Can you feel the need to complete your profile and achieve a higher level?
Nir often talks about this one core habit loop that drives retention. In reality, we see more than one loop. One combination we see is a smaller engagement loop before the core habit loop.
Let us take LinkedIn as an example. The following is one of the main core habit loops of LinkedIn: to meet, search and contact someone:
In this core loop, trust plays an important role. The less deep the connection, the more is the need to come across as trustworthy. And this is achieved by having a complete and well executed profile. Because this task can be daunting and quite tedious, LinkedIn puts a trust enhancing profile completion loop before or on top of the core habit loop:
Once LinkedIn sees that you have completed a meaningful amount of your profile, they focus on triggering other loops such as networking loops, vacancy fulfilling loops, sales activity loops (all paid services) and content creation loops (growth loop).
This is a common theme we see across brands as the competition for attention increases. Sometimes it is through an easier, smaller first loop, at other times it is through creating engagement through content first. Once they are in the habit of consuming content, it is easier for them to invest in more difficult parts of the product/service.
Meaning small asks before bigger ones achieve higher rates of success (which we all know thanks to Robert Cialdini).
Let us have a look how these loops connect and reinforce each other:
In our last and introductory article Funnels vs. Loops we suggested: it is the combination of funnels and loops that drives exceptional growth. We continued our series by deep diving into retention in this article.
We decided to start our deep dive with retention loops because without having this solved, it is difficult to build sustainable linear traffic sources and acquisition loops.
In order to get retention right, you need to get in your customer’s head through qualitative research and connect variable reward to emotional triggers. Having a valuable product helps a lot, but it takes more to reach habitual use.
There is a continuous fight for the attention of our customers. What brings them in may not be what keeps them coming back, as we have seen with the LinkedIn example. There are many and different types of loops at play.
Until next time.
Our next article will deep dive into acquisition (growth loops), followed by one focused on linear traffic sources (funnels).