To reduce your product’s churn rate, first find out why users stay

Most of us have learned to think about products in terms of user churn at specific steps in the funnel. We keep asking ourselves “Why do users leave?” and then we try to find and fix the reasons for this. We assume that solving those issues and removing friction will improve the key product metrics.

Funnel optimization surely is a good approach to improve key product metrics. However, it doesn’t work in all situations. And when it does work, it usually only brings incremental improvements, not fundamental changes.

Today, we’re going to talk about a different approach when examining your product. This approach can boost your product, and sometimes it can take your product to a completely new direction.

I suggest posing the question “Why do users stay?” before “Why do users leave?”

If you want to learn how data can help you build and grow products, take a look at GoPractice! Simulator.

To reduce your app’s churn rate, first find out why users stay

Why do users stay?

In this article, we will use the following product model:

  • The core is the key value that a product delivers to its users.
  • Everything else (positioning, distribution channels, creatives, landing pages, onboarding, email chains, scalable infrastructure) are supporting constructions that help convey the benefits of the product core to the users.

When you’re trying to find out why users leave your product, you focus on individual pieces of these supporting structures. This approach often makes you lose sight of the product’s core value that these constructions are supposed to serve.

When you shift to “Why do users stay?” you start focusing on the product’s key value and how thoroughly it is delivered to the users at all product levels.

Building the product backwards

Each product starts with an idea around which you create your first version. Then, when you launch this first version of the product, reality kicks in and brings you back to earth, and you find out how much value the original idea had. And then you start searching for the right product and the right audience.

Often, this investigation proves that the original idea bears no potential. Sometimes this process may lead you to a working solution for the original problem you were trying to solve. And occasionally it may take you to completely new directions. In some cases, a small piece of the initial product turns out to be the very essence of the product’s true value.

When you find the true value of your product for a specific audience, you’ve reached the point we call the “product/market fit.”

The way you work on the product before it reaches its product/market fit differs greatly from what you do it once the real value is there.

“Before” finding the product/market fit, you wade forward through the unknown looking for the value. Once you find it, the “After” part is all about retracing your steps and trying to show others where this value stands. Now your goal is to bring this new value to the market.

On your way back, you start building a product around what you’ve found. You crystallize the core, get rid of the clutter, and discover how to convey the created value to the target audience. Here is how Robert McKee describes this process in the context of working on a story:

“Once the Climax is in hand, stories are in a significant way rewritten backward, not forward. The flow of life moves from cause to effect, but the flow of creativity often flows from effect to cause. An idea for the Climax pops unsupported into the imagination. Now we must work backward to support it in the fictional reality, supplying the hows and whys. We work back from the ending to make certain that by Idea and Counter-Idea every image, beat, action, or line of dialogue somehow relates to or sets up this grand payoff. All scenes must be thematically or structurally justified in the light of the Climax. If they can be cut without disturbing the impact of the ending, they must be cut.”

Before you reach product/market fit

Multiplying any number by 0 gives 0. Likewise, when you increase the funnel conversion rate of a non-working product core, nothing changes. That’s why there is no point in trying to solve user churn before you find the key value of your product.

At this stage, the supporting constructions are temporary. You need them to check the next version of the core value. These constructions can be implemented in the product itself or replaced by, say, a person. For instance, you can manually set up the product for your new clients and teach them how to use it in person. This option is very good when you are in the stages of testing and validating your hypotheses about the product core. It speeds up the process of filtering non-working solutions and brings you closer to the working ones.

At this point, the main task is to find those who will use this half-baked version of your product and are happy with it. Also of interest are users who might be disappointed with your product but continue to use it for some reason. Their willingness to continue using the product and the feedback they will provide will be the base material for building your new and improved product. These are the people who will give you the answer to the question “Why do users stay?” and the equally important question “What kind of users stay?”

Answering these questions will show you what kind of value your product delivers and who are its real customers. If the answer sounds just right, you can start building a real product around it. Note that this is not always true; sometimes, the potential market for the product you discovered may be very small and not something you’d invest your efforts into.

Once you reached product/market fit and built the first version of the product

It still makes sense to ask yourself “Why do users stay?” even after you’ve found the key value and built the first product’s version around it. Answering this question thoroughly will help to take the product to the next level.

Surely at this stage, it is also time to ask “Why do users leave?” Solving the problems at different steps of the funnel will increase the share of users who experience the product’s core value. These users are likely to become your customers in the future.

Nevertheless, it is crucial to understand that the potential growth of the product metrics through the improvement of its supporting structures is limited. At best, you will be able to convey the core value to 100% of potential users. This is where it stops.

The main growth potential lies within increasing the product’s core value. For example, one of the ways to do this is to develop a completely new feature. Another way is to find the hidden value of your product and bring it to the surface. That’s what we’re going to talk about.

Bringing the product’s hidden value to the surface

The way people are using the product usually shows what we should build next. In most cases, the current version of the product already contains hints of what will be the next successful feature. Users often hack the product in different ways to achieve their goals, and that’s exactly where the hidden value lies. By exploring how people hack your product, you get to know what to do next.

Do you remember the first versions of social networks back in 2005-2008? They were simple catalogs with people’s profiles. The main use case was searching for pages of the people you knew. On their profiles, you could find answers to questions that people ask each other when they meet, like what is their current mood, what they’ve been up to, their relationship status, their political views, etc.

It’s natural to expect people to visit a friend’s profile once and never come back to it, or do so very rarely. But users kept coming back to their friends’ profiles many times during the day. Why? Because they wanted to know whether their friends had anything new to share. This behavior formed the basis for the Facebook Newsfeed, one of the most successful products in the entire history of the Internet. This was a product that people used before it was actually created.

I highly recommend reading the story behind the News Feed. You can find a short version here, and a more comprehensive version here.

When you answer the question “Why do users stay?” you are looking for the hidden patterns in the product usage. You can then get cues from these patterns to develop specific features. You’re learning how certain users (probably very few so far) benefit from your product. Once you find these ways of getting the value, your task is to teach other users to do the same.

This is exactly what you did before. There is only one little difference compared to the stage when you haven’t found the value yet. Here you need to think about how you are going to deliver this value not only to your new users, but also to the old ones.

Then, you keep working on two things in parallel. First, you rebuild and optimize the paths to the value that you have found. Then, you continue searching for ways to increase the product’s key value by asking “Why do users stay?”

How data analytics can help to answer the question “Why do users stay?”

Once you learned the theory, it’s time to get down to practice. You can do it here.

To answer the question “Why do users stay?” I suggest using the following approach: You need to take users who have been using the product long enough (e.g., they first launched the product 30 days ago) and are using it regularly (e.g., every day during the last week).

After examining their usage and behavior patterns in the product, you will learn who your target users are, and how they get value from your product.

These users have been there for quite a while and studied the product well. They made up their mind about what really benefits them. Their usage was naturally reduced to the key product value. And this is what you are looking for.

To study such users, I recommend using the following methods:

  • Conduct personal interviews with several such users. Find out for what tasks they use the product? Why did they choose your product over others? What other products have they tried? Who are these users? Do they have any common characteristics? How did they find out about your product? Does it provide clues about how and where you can attract more users like them?
  • Rate the popularity of different features among these users. This cohort will probably use the product in a different way than the overall audience. The overall feature popularity might be skewed due to the presence of new users who are just exploring the product.
  • Take a look at the sessions of these users. How exactly do they use the features? In what sequence? Do they all use the same things or are there different clusters of users with different behavior?
  • Take a look at their first sessions in the product. The way the successful users have studied the product will show you how to onboard others.
    How are successful users different from others? Is it about the traffic sources, device specifications, geography, or any other relevant characteristics?

Summing it up

Most products are very difficult to understand. But we usually tend to create simplified models when we’re in the initial development stages. The purpose of this article was to give you one more model to think about your product and make it better.

Why do your users stay? Feel free to let us know in the comments.

 

P.S. If you want to learn how data can help you build and grow products, take a look at GoPractice! Simulator.