How to evaluate the effectiveness of a mobile app

All companies investing in mobile apps sooner or later face the question of how to objectively measure their effectiveness. The correct assessment answers not only how many users have opened the app but also how well the product works for business purposes.

In this article, we will consider what user activity means, why it should be tracked, the correct calculation of the key metrics dau, wau, and mau, what mistakes to avoid in the calculation, and what practical steps can help improve app performance.

What does «user activity» mean?

User activity is defined as those actions that indicate the involvement and value of a product for a particular person. Activity can be different: from a simple session when the app is opened to a targeted action such as registration, payment, or publishing content. It’s important to separate the number of visits from the number of unique visitors: the former measures the performance of the interface, and the latter measures the real base of returning users.

Let’s look at an example to better understand. After simplifying the first acquaintance with the application, the user gets to know the interface faster, gets to the main action without unnecessary steps, and manages to complete it on the first day of use. This change increases early engagement and sends a signal that onboarding is working. If the user comes back for another week, we get a more stable retention rate.

Why track activity in mobile apps

Regular analysis of user activity allows you to see how the app is actually used, not how it was intended at the development stage. Such data forms a practical business context: it shows whether the user returns to the app, how often they interact with key functions, and at what stages their interest in the app decreases. It is activity that becomes the basis for informed marketing and product decisions, not intuitive hypotheses.

Activity metrics help to:

• predict LTV and evaluate the profitability of engagement channels, as they allow you to understand which users do not just install the application, but stay with it and create long-term value for the business;
• Identify problem areas in the user journey – from the first encounter with the application to performing a key action and regular returns, which allows you to improve the interface and use cases;
• make informed decisions about investments in retention, such as push notifications, email communications, or app updates, based on real behavior, not just the number of installs;
• plan releases and resources, taking into account peak periods of activity, team workload, and expected infrastructure load.

How to calculate DAU, WAU, MAU metrics correctly

The most common activity indicators are DAU (daily active users), WAU (weekly active users), and MAU (monthly active users).

They provide a different time perspective of app usage and serve as a basis for making product and marketing decisions. The main principle, as we have already mentioned, is to count unique users, and the calculation itself should be transparent and reproducible by all team members working with data.

DAU metric

The metric shows the number of unique users who were active during a specific calendar day or within a selected 24-hour interval. Activity can be counted as follows: opening an application, successful authorization, performing a key action by a user, for example, sending a message, making a purchase, publishing content.

When calculating, it is important to

• ensure that each user has a single identifier (user_id); this allows you to count a person only once, even if they opened the app several times a day;
• decide whether to count guest sessions or only authorized users, and apply this rule permanently;
• take into account time zones: the «for today» report should be clearly defined by UTC or local market time.

In addition, it is important to document which events are considered activity, as this ensures correct calculation and allows other team members to understand where the numbers come from. After the first calculation, compare the metrics with backend logs to quickly identify discrepancies.

WAU metric

This metric measures the number of unique users per week. You can calculate the metric for a fixed calendar week from Monday to Sunday or for , a rolling period that covers the last seven days in a row, regardless of the date on the calendar. However, the choice must be made immediately and applied consistently so that the metrics remain comparable. WAU smooths out day-to-day spikes and gives you a better idea of the regular interaction with your app.

When calculating, you should:

• choose one approach to the interval and document it;
• remember that WAU better shows the behavior of users with a cycle of use several times a week (media applications, delivery services, etc.).

WAU also helps to identify if an application has cyclical behavior, such as weekend spikes, and whether communications need to be tailored to fit these cycles.

MAU metric

The metric means the number of unique users per month. Again, you need to choose between a calendar month and a rolling period within the last 30 days. It is useful for assessing the scale of the audience, long-term dynamics, and for calculating ratios, for example, DAU/MAU as an indicator of engagement.

When interpreting, pay attention to:

• seasonality and marketing campaigns that may increase the indicator in the short term;
• the ratio of DAU/MAU and WAU/MAU: a low value may indicate a passive audience, while a high value may indicate regular use;
• the way users interact with the app: whether they are one-time guests or regular customers.

MAU gives a general picture, but always combine it with other metrics and high-quality feedback from real users to make decisions.

Let’s look at practical examples of how to count and how to interpret the results.

To begin with, let’s imagine a small social application with a mixed audience: some users log in frequently, while others log in sporadically.

On a particular day, the system recorded 1,200 unique user_ids, and the total number of sessions was 3,400.

• This gives DAU = 1,200 for that day.
• Over the past 7 days, the system counted 3,800 unique user_ids → WAU = 3,800.
• Over the past 30 days – 9,500 unique → MAU = 9,500.

The DAU/MAU ratio of ≈ 0.126 (12.6%) shows what share of the monthly audience the app attracts on a daily basis. For a social app, this may mean that a significant portion of users visit less than daily, which is a signal to look at the engagement mechanics, i.e., feed, notifications, social triggers. At the same time, it’s worth comparing WAU/MAU to understand how many users come back every week: if WAU/MAU is significantly higher, it means that the app has a weekly usage cycle.

Consider another product – a subscription-based SaaS application with paid subscriptions. Over the past 30 days, MAU = 2,000 paid users, and the average daily rate is DAU = 400.

• Then DAU/MAU = 400 / 2 000 = 0.2 (20%).

This value means that on average, 20% of the paid audience uses the app every day. For services with daily use, such as work tools, this may not be enough, but for B2B products with work cycles 2-3 times a week, it’s normal. It is important to analyze the indicator together with business metrics: average revenue per user, churn, and planned usage scenarios. At the same time, check whether the DAU/MAU calculation correctly takes into account users who are temporarily inactive due to technical reasons or changes in the application version.

Stickiness metric

Special attention should be paid to the so-called stickiness metric, an indicator that helps to understand how regularly users return to the app. It doesn’t measure engagement directly, but it is a good indicator of the quality of daily or weekly interaction with the product over time.

The stickiness metric is usually calculated as the ratio of DAU to MAU or DAU to WAU. In the first case, it shows what share of the monthly audience uses the app on a daily basis, and in the second case, it shows how actively users interact with the app during the week. The higher this indicator is, the more likely it is that the app fills a regular need rather than being used sporadically.

In practice, the stickiness metric helps to:

• assess whether the app has become part of the user’s daily scenarios;
• compare the dynamics of activity after updates or changes in functionality;
• identify the risks of decreased engagement even before the MAU drops.

It is important to consider the context of the business model. For financial or work services, a high level of stickiness is expected, while for travel or e-commerce apps, lower values may be the norm. Therefore, this indicator should not be analyzed in isolation, but together with other activity metrics and taking into account the logic of using the product in the future.

Mistakes to avoid when calculating

Let’s take a look at common mistakes that can occur when preparing reports:

1. The criteria for «user activity» are not defined. If you don’t define what action a user must take to be considered active, the metrics lose their practical value. For each application, you need to document which events form activity and apply these rules consistently across all reports.
2. Trying to measure everything at once or blindly following industry standards. When a user is evaluated through dozens of indicators without prioritization, analytics becomes complicated and data interpretation becomes superficial. It’s better to focus on those metrics that show the real value that the user creates for the business.
3. Lack of user segmentation. Aggregate metrics hide the differences between groups: new, regular, and paid users behave differently. Without segmentation, decisions are made on the basis of an average user who does not actually exist.
4. Ignoring the user life cycle. Visitor behavior changes depending on the stage: acquaintance, activation, retention, or re-engagement. If you do not take these stages into account, the criteria for user activity will be incorrect at the calculation stage.
5. Using sessions as a substitute for a unique user. Counting by sessions leads to overestimated results.
6. Counting device_id instead of user_id. When a user changes the device, the system can count it several times, which distorts the real number of active users.
7. Unfiltered bot traffic and test accounts. A test user or automated traffic creates artificial activity and misleads the analysis.
8. Mixing different time periods without explanation. If a user is counted for different periods without a single standard, the data comparison becomes incorrect.
9. Incorrect interpretation of reactivation. Reinstallation or a short-term return does not always mean that the user has become active again in the product sense.

How to improve DAU, WAU, and MAU metrics

Improving engagement is a comprehensive work with the product, which always focuses on the user: their motivation, expectations, and actual behavior. For metrics to grow steadily, it is important to break down the work into stages of interaction that the visitor goes through. At each stage, the user has different goals, and the business has different tools of influence, so the logic of improvements should be consistent and systematic.

Initial acquaintance and first activation

At the start, the key task is to get the user from installing the app to the first targeted action as quickly as possible. To do this, you should minimize forms, gradually open functions, and show the value of the app in the first 60-90 seconds through short tips or demonstrations. It’s important to measure how long it takes for a user to perform the first action and how many people reach this step. These are the metrics that best show whether simplifying the interaction really helps to drive faster activation. Improving this stage directly affects the first-day DAU and forms the basis for further behavior in the WAU.

Push notifications and triggered communications

For a user to return to the app, communications should be tied to a specific action or need. Personalized triggers are a reminder of an incomplete action, a recommendation that matches a previous activity, or a notification of an event that may be useful to a particular person. The frequency, time, and channel of contact should be chosen so that the potential client perceives the message as help, not pressure. If the user sees real benefits, it has a positive impact on the return and the dynamics of WAU and MAU.

Product loops for repeated interaction

For an app to become a part of a regular scenario, the user must have a reason to come back again and again. Product loops work when they support the core value of the product for the user: regular content updates, rewards for repeated activity, social signals, or short internal challenges. It is important that the user feels the sense of repeated interaction, not artificial retention. The effectiveness of such solutions is evaluated through behavioral changes: how often a person logs into the app, how the duration of sessions changes, and whether the share of users who return regularly increases.

Experiments and rapid iterations

Sustainable growth of metrics is impossible without systematic work with hypotheses. Each change should answer the question of how it will affect the user and his behavior. The practice of «one hypothesis – one factor – one test» allows you to clearly see how the user reacts to a particular change. To do this, you choose clear metrics, form a test group, and analyze the result. If the visitor starts coming back more often or performing targeted actions faster, the change is scaled up, and if not, the team returns to a new hypothesis. This approach allows you to gradually improve the user experience while maintaining control over data quality and the real impact on DAU, WAU, and MAU.

Iryna Voitovych

Copywriter

Author's page