The most important growth metric for early startups

About the author

Jeff Chang (@JeffChang30) is a growth technical leader at Pinterest and angel investor. If your startup is looking for an angel investor who can help with all things growth, please send over an email!

As a startup, there are a lot of different metrics you could optimize for: how many signups you have, how much revenue you’re generating, your WoW growth rate, the list goes on. For most early startups, I think the most important metric to understand and attempt to grow is cohort retention rate (will shorten as just retention rate). Retention rate is a strong determining factor of whether or not your startup has a long future. Some people even define product-market fit as retention. In this post, I’ll talk about how to measure retention correctly and why it’s important.

Sometimes I see startups report retention as something along the lines of “each month, 90% of our users will still be users the next month.” Looking at your entire userbase as a whole while calculating retention is not a good way to calculate retention because it depends on your mix of old and new users. It doesn’t make sense to compare retention rates of a newly joined user and a user who has been around for a long time, so the correct way is to measure it by cohort. A cohort is a group of users who joined around the same time. So, you should graph a cohort retention graph by taking a single cohort of users (week or month is a typical time period), and plotting how many of them come back or do a key action over time:

image_preview (1).png

If your cohort retention chart looks like the above, you’re in good shape! 

However, if your retention chart looks like the below, despite having a higher retention rate at week 5, you should iterate on your product to look more like the first chart.

image_preview (3).png

Cohort retention graph tips:

  • To get significant value from your product, does your product need to be used daily (messaging products), weekly, or yearly (travel products)? In the above graphs I used “Weeks in”, but you would change this to your optimal product usage frequency

  • For the y axis, % retained should be a key action for your product. For example, it could just be “visited app” for content sites, “message sent” for messaging apps, “trip booked” for travel/transportation apps, or “still subscribed” for subscription businesses.

  • If your product has both a website signup and app signup, you should make multiple cohort graphs for signups by web and signups by app, since they will look significantly different

There’s a specific point on the cohort retention graph that’s very important to know - founders and growth leaders should basically have this memorized:

image_preview (2).png

At around 5 weeks, this retention graph starts to flattens at 35%. This data point is important for two purposes. First, you can expect around 35% of your signups to be around long term, which can be used for back of the envelope calculations for things like paid acquisition. The 35% number is your long term cohort retention rate, which is the most important growth metric for early startups.

Next, if you launch any big product features or run any experiments, you know that you will need to wait around 5 weeks to get accurate data on how many additional retained users your change has acquired. The takeaway here is not that you should run experiments for 5 weeks, since that is too slow, but to take retention data previous to 5 weeks with a grain of salt. An experiment that acquires a lot of additional low intent users may seem promising in the beginning, but fall back down to control levels as it approaches this 5 week mark.

Why is long term retention rate for startups so important? In general, acquiring users is much easier than getting them to stay. There are a lot of acquisition channels that you can use to acquire users at scale, such as SEO, referrals, and paid, but the growth strategies used to retain such as emails and notifications can only change your retention rate by a certain amount. Over time, acquisition rates of good startups change a lot - they can potentially be 1000x or more, while retention rate doesn’t usually change as much, maybe double at best.

Great retention rates make everything easier. They make your LTV much higher, which allows you to spend more to acquire users. They make your acquisition efforts higher leveraged. They help your company survive even when acquiring users gets harder. That’s why a lot of investors are looking for great retention rates for startups that have been around long enough to be able to measure it. Your WoW user numbers, revenue, signup rate will change a lot over time, but your retention rate will likely stay around the same through the life of your startup.


  • As an early startup, retention rate is the most important metric

  • Build a cohort graph to measure your retention rate

  • If your cohort graph bottoms out near 0%, improving that is your #1 growth priority.

I only post every few weeks, so follow me on Twitter to receive daily growth advice!

If your startup is looking for an angel investor who can help with all things growth, please send over an email!

Jeff Chang