In this post, we’ll dive deep into what retention is, why it is important, how to keep your startup alive by focusing on retention and what tips and tricks you can use to improve it.
Focus on retention to keep your company alive
Simply put, retention is the measure of how many users return to your product over time.
Even for the best products, it’s inevitable to lose a portion of users in a few days.
Based on Quettra’s data, the average app loses 77% of its DAUs (daily active users) within the first 3 days. In 30 days, it’s 90%. After 90 days, it’s over 95%. So the retention rate for the average app after 90 days is around 5%. That’s very low.
So if you want to grow your company, you need to focus on retaining users, by improving retention. Retention is really the foundation of all growth as it directly impacts the virality, the lifetime value (LTV) or even the payback period for any product.
In most cases, companies and software products look at what we call N-day retention. This is probably the most spread retention metric you’ll find reported anywhere.
N-day retention shows what percentage / portion of users come back on the ‘Nth’ day after their first use of the product. In practice it means that if 100 people signed up yesterday, and only 10 came back, my 1st day retention (N=1) is 10% (10/100). By looking at multiple days in a row, you can plot these retention values over time and draw a retention curve highlighting these values.
This is what we call an app retention curve.
A curve like the blue one shows a weighted average of all Nth day retention values for all user cohorts in a certain time period.
Let’s assume that today is 31 December. So 365 days have passed from the year. I have 364 1-day retention numbers. 363 2-day retention numbers. 362 3-day retention points. And so on. By weighting the averages of all these data points, we can come up with a curve like the above one. So the app retention curve shows the average percentage of active users every day within a specific timeframe.
In real life many companies and startups look at specific Nth day retention dates and compare those to each other and industry benchmarks. The most common ones are the 1-day, 7-day and 30-day retention numbers, also referenced as D1, D7, D30 retention. This is a good metric if the usage of your product is daily. It is not a good metric though if you are not operating in a day-to-day usage type of industry (like an accounting or tax software).
To get a better picture for the usage, you can focus on weekly, bi-weekly or monthly retention numbers too. In this case you can look at a bigger timeframe, not just specific days. So if you have a 40% week 1 (W1) retention, it means that 40% of users come back at least once in 7 days (1 week) after signing up. It could have happened on day 1 (D1) or even day 4.
A company can successfully and sustainably scale its user base if it can retain users. When there is a certain amount of people using the app on a regular basis, we say that the company has reached product-market fit.
The definition of product-market fit is very vague. There are multiple ways to look at it.
According to Mark Andreesen: Product-market fit means being in a good market with a product that can satisfy that market.
As he puts it: “You can always feel when product-market fit isn't happening. The customers aren't quite getting value out of the product, word of mouth isn't spreading, usage isn't growing that fast, press reviews are kind of "blah", the sales cycle takes too long, and lots of deals never close.”
This concept is great, because it’s very general, so you can apply it to multiple situations. But it lacks some clarity and definition.
Brian Balfour describes the phenomena more qualitatively and quantitatively: “If your retention curve flattens out, you have product-market fit (Product A).” It could be only for 5% of your user base, but for those, you have product-market fit. If the app retention curve will drop to zero, you don’t have good fit (Product B).
Sean Ellis test
Sean Ellis, the person who coined the term growth hacker, put this in a more quantitative way a few years back: “I asked users a simple question.
How would you feel if you could no longer use [the product]?
- Very disappointed
- Somewhat disappointed
- Not disappointed (it isn’t really that useful)
- N/A - I no longer use [product]
This is called the Sean Ellis test. If over 40% of the users respond to the survey saying they'd be "Very disappointed", there's a good chance you have found product-market fit. Sean Ellis compared data from over 100 startups back in the days and found that those companies who had a 40%+ answer rate to the first option had strong traction. While those who were scoring lower had no or small traction. It is advised to ask at least 50 users to get statistical significance, but the more you can ask, the better informed you are.
Sean generally recommends to survey the following users:
- People that have experienced the core of your product offering.
- People that have used your product at least twice.
- People that have used your product in the last two weeks.
By looking at any of the three definitions, it’s hopefully clearer how you can look at and define retention.
There are certain tips and tactics to improve user retention. The two most common methods are:
- Shift the retention curve up.
- Flatten the retention curve.
Shifting the curve up focuses mostly on the first-time user experience, the onboarding flow and tries to pass along the core value of the product to new users.
Flattening the curve increases the baseline level of users delivering a better product experience over time, so it results in long-term user retention.
Of course you can not only look at the portion of active, retaining users (like W1 = 23%), but you can track the net number of users too in a certain timeframe. The most common metrics for this category are DAU (daily active users) or MAU (monthly active users).
To define any retention metric, you first have to settle on what the core action is that you measure against. In other words, what qualifies a user as active? Is it opening the app? Is it visiting the website? Is it creating a new project? Is it adding a new contact?
The answer depends on the nature of your product. If you are a social app, the number of friends you add, the conversations you have, the time you spend consuming content are good indicators of a core action to measure against. If you are a project management tool, then you want to track how many active projects are in the user’s profile, how often they create and complete projects, how well they collaborate. If you have a PM tool, but have no projects in it, no matter how many times you open it, you’ll unlikely to be a successful / active user. Defining the right and correct retention metric has a great impact on your future success.