• Matthew Lerner

How to find your "aha moment" (plus fresh examples)

First of all, why does this matter? Because “Monthly Active Users” (MAUs) hide an important truth. (Investors might want to read this too)

In the startup world, you’d naturally want to increase the number of (daily / weekly / monthly) active users, right? Sort of.

Erwin Schrödinger and the half-dead startup

Generally, you want more MAUs. But, just as Schrödinger’s fabled cat was neither certainly alive nor dead… some of your MAUs aren’t as “alive" as others.

Imagine you could read your MAUs’ minds, you’d see a mix of 3 mindsets:

  1. Happy loyal customers 🏆

  2. Uncertain new customers who are just trying out your product 🧐

  3. Unhappy customers who might churn 😡

For your investor updates, you might be able to sneak by with “MAUs.” But for your own sanity, you really need understand which MAUs are “new” vs. “loyal” and the conversion rate between them.

First, new users will want to try your product before deciding if it’s for them. In your data this might just look like an “activation” but it’s not that simple. Since you can’t actually read their minds, how do you know when they switch from “trying” to “loyal”? You’ll need to look for a behaviour pattern - your "Aha moment."

How to find your “aha moment"

Your data will show some kind of behaviour pattern that tends to predict loyalty. For Facebook, it was 7 friends in 10 days. For Twitter new users needed to following at-least 30 users with 10 following you back. And, new Netflix customers needed to add at-least 3 titles to their queue in the first 90 seconds. (Notice it's time-bound).

If you can do fancy analytics, you can infer this from past data by establishing what early-life behaviour pattern predicts month-3 retention. Or, you can just spend time with new customers, see how they initially use the product, and use your common sense. Here are some fresh examples:

  • For payments, that first $1 transaction is a test, but 5 transactions of different amounts from different senders in 48 hours = signal.

  • For a design tool, making an image is “trying” but exporting it to social is signal.

  • For expense software, tracking expenses is “trying” but submitting an expense report over $100 is signal.

  • For B2B SAAS, inviting 2 co-workers is “trying” but inviting your whole department is signal.

If you can’t do the fancy analytics, or don’t yet have enough historical data, that’s fine. Don’t over-think this, just pick a behaviour pattern that sounds reasonable, and run with it for now. (Delays from over-thinking this are more dangerous than starting with the wrong metric).

Simple next step

Establish your “aha moment” and start tracking “loyal” MAUs, “new” MAUs, and the conversion rate between them (your habituation rate.)

How do you improve your habituation rate?

This thinking (and even the Schrödinger reference) come from a reclusive genius named Diego de Jodar. He’s kindly put together this video to show you how to improve your habituation rate. (warning: contains strong language). If you love his stuff as much as I do, give him a “like” and “subscribe."

What about “likely to churn" customers?

That’s beyond the scope of this “two minute” email. But check out my epic Tweetstorm “How we cracked a $100M Churn problem at PayPal with a spreadsheet, some SQL, and a physicist named Ben." (Not Erwin)

110 views0 comments

Recent Posts

See All