How to predict user behavior in the application

CEO App in the Air, Bayram Annakov, dismantled at Epic Growth Conference practices that help to multiple increase the retention rate of the user in the application.


Read the transcript of the report below.

The user always wants to tell us something.

Someone on the slide below sees poor planning. I see a pedestrian message to architects here. An analogy can be made with the product. How do users really go from point "A" to point "B".
')


We design the interface as we think the user should use it. The task of the product manager is to understand what the users want, based on their trajectories on the application and screenshots.

Why study users' messages?

1. The user is not a product manager.

There is a proverb: "Man proposes, but God disposes." You can apply it to the work of web services: "Product Manager assumes, and the user has". We can not always predict exactly how people will use the application. In order to analyze the situation in time and work on the errors, it is important not to “hammer” on user messages, but to study them.

2. The funnel is not all

Funnel is the most common way to study messages from users. However, the problem with funnels is that they are becoming insufficient at a certain moment. There are several reasons for this:

Funnel greatly simplifies complex, rich user behavior.
The slide on the left shows how users behave on the screens of your product. Right - how do you reduce this behavior in a set of stages, assuming that the stages of the user's movement are sequential.



The user can go on onboarding and then switch to another application, because the SMS came with the message “come back”.

The user does not go clearly to the goal that you set for him. He is guided by complex behavior, and the funnel greatly reduces his behavior. It does not allow you to see all the complexity and richness of the user's message.

User wasting time

The funnel does not take into account the time dimension. There are accurate screens where users spend more time and less. In our application, for example, we know that if a user begins to read a privacy policy, then most likely he will leave the application and will not return.

At some point, you feel that the funnel no longer provides answers to your questions. Then you have to learn the trajectories of users.

3. User schedule is key

What is a trajectory? Imagine: you have standard events (Google Analytics, Firebase, Amplitude). Events have a time sequence. You represent user behavior as a sequence of actions with transitions from one to another event.

Nodes are events (as a rule, these are screens). Transitions are jumps between screens. When we draw a screen layout, we use about the same tool.

It would be cool to analyze all the trajectories of all users, to find patterns in behavior and what they are trying to tell us. But when the number of users exceeds 100 million per month, there is not enough time for manual analysis. You have to use an automated tool.

4. Frequency analysis = benefit

We have developed a toolkit for tracking the trajectories of users who buy and do not buy our product. We use the matrix of the frequency of use of the product.



Along the edges of the slide are different cohorts of users. Two charts show the proportion of users who buy our subscriptions from each cohort. On the X axis - we see an indicator of the frequency of using features, on the Y axis - users.

When you build a similar matrix, you begin to see the fundamental differences of one cohort from another. Knowing that as a result there are differences between what percentage of users subscribe and what not, you can understand which screens, events and actions lead to the user's understanding.

5. Through the group graph you can see insights.

We are interested to see the sequence in which users use features, and build what we call a “group graph” - a graph characterizing a certain group. For example, the key functions they use.

Further, depending on your application or tasks, you make people move along the trajectory that gives you maximum results.

If you clearly understand that your product is suitable for different categories of users, then build onboarding. You can also sharpen the whole part of the product for this use case.

6. Cycles lead to user outflow.

When you get a tool that automatically analyzes the graphs, and built a transition graph on one of the cohorts, you start to see the loss in this graph.

For example, we lost about 5% of users after one of the onboarding screens on which the user could connect a calendar.

This happens because of looping: the user goes around a set of screens, repeats the same actions, and then closes the application. Cycles are very easy to find if you build a mathematical graph - the more cycles a user makes, the lower his retention rate.

7. Dynamic Counting

We found out which cycles of the sequence of user actions collected in the trajectory make the greatest contribution to the fact that a person leaves. We started flashing these cycles.



Using trajectories, you define patterns of user behavior. To do this, you can impose ready-made mathematical tools, for example, search for cycles - they will quickly show which cycles lead to people leaving.

You immerse yourself in these cycles, cross-check for a couple of users, view the entire cycle, understand what the problem is, and flush the cycle. This instantly gives a profit in the user retention rate.

A good example: Imagine that your user arrives at Dubai Airport and is lost. This is one of the most - in terms of navigation - incomprehensible airports. At some point, an airport clerk notices him and points in the direction of the exit. For your service, you can dynamically change the UI to maximize retention.

We thought: “It's great to do this within the company. But it is even more cool to provide all these tools and give product managers the opportunity to use them. ”

Work with Google Analytics or any analytical tool. A set of tools will help you automatically build graphs and make a person’s prediction on the latest X events.

How does analytics evolve in many companies?

Let's imagine that we have two axes. One end - "I know", the second - "I do not know." The second axis operates on the same principle. Observation of many companies and the evolution of analytics showed that we all move within the framework of this quadrant.

What place in the quadrant is occupied by the retenshening and the tools described?

1. "We only know what we know."

Usually this is the main analytics dashboard. We know how many downloads we have, users, how much is our income. At this level, “factology” occurs. This can not be called "analytics", just statistical information. Many companies still remain at this level.

2. "We know that we do not understand something"

They know, for example, which coefficient of retention or which LTV. They begin to measure this in many ways in order to predict the future.

Why measure retention? To predict the future number of active users. Why measure LTV? To understand how much we spend and how much we ultimately earn from the user. How to relate these data to each other? When we are at the stage, “we know what we don’t know,” we gradually consider them and try to look into the future.

3. "We do not know what we know"

This is the place of retensing and many machine learning approaches. We already know how to measure user trajectories. We know that users are trying to tell us something. But we do not analyze this information. Tools help us pull messages out of users and get insights to improve the product or, conversely, turn it off.

4. "We do not know what we do not know."

When you understand retenseniringom, you need to move in this direction. This stage can be described as going astral in analytics. You are constantly looking for ideas, trying to apply them in your product, to check and analyze the results.

More product marketing reports available at the @epicgrowth Telegram Channel.