MAU is a bad metric, and Twitter’s recent stock fluctuations prove it

Twitter grew its number of monthly active users (MAUs) 18% this quarter, meeting predictions and breaking the 300 million mark. The company’s stock is 20% lower since the announcement.

Last quarter, Twitter had 288 million MAUs, missing its target by 4 million. 48 hours later, the stock was up 15%.

“But, there’s more to the story,” I hear you grumbling at your phone (or whatever you’re reading this on). Well, yes of course: revenue growth is what the shareholders really seem to care about.

In which case, what’s the point of reporting MAUs? The metric correlates neither to revenue, nor the stock price. In most consumer businesses, the more users you have, the more money you can make. Apparently that’s not true of Twitter’s MAUs, which seems to make the metric redundant.

Why have a user metric at all?

A good user-based metric can be really valuable to a consumer product. It (1) guides product and marketing decisions (it should be actionable), and (2) is often a leading indicator of monetization and revenue.

But MAUs — especially in Twitter’s case — are neither actionable, nor a reliable leading indicator of revenue.

Consider two users, Francine and Jemima: Francine has been encouraged to sign up for Twitter by a friend. She does so by downloading the app and registering for a new account. She doesn’t really see the value and never returns. Jemima has been using Twitter for years. She checks her feed multiple times, every day. She tweets and re-tweets often.

Both Jemima and Francine count as an MAU. But which of these users is truly active? Which of these should Twitter want more of? Which is likely to create value and revenue opportunities for Twitter?

If I were to invest in Twitter, I’d want to see a user metric that told me how many users are engaging with Twitter as a habit. Something like: how many users visit the app or website 5+ days in a week? Is that growing? Of course this will be a smaller number, but trying to find the biggest number shouldn’t be the game here; it’s having a number that’s useful.

If you’re building a consumer product that you want people to engage with over and over, day after day, you need to measure whether or not you’re accomplishing that. Measuring MAUs doesn’t do it.

Other notes:

1. I’d always expect a Q1 jump in active users for a service like Twitter. For many, the new year brings new resolve to engage or re-engage with social media. I saw this behavior both at WordPress.com and Medium. So it’s no surprise to me that for the last two years, Twitter’s Q1 has seen the biggest quarter-on-quarter increase in active users. This pattern doesn’t bode well for Twitter’s growth through the rest of the year.
2. The disclaimers around measurement (slide 3) of MAUs are also worth reading. The metric isn’t just bad because for its inability to capture actual core usage of the product, it’s bad because it’s really hard to measure well.
3. An early Twitter employee told me once that they wished the company had had better user metrics from early on, because once you adopt a bad one, it’s really hard to shake it. Lots of echoes of that sentiment in Ev Williams’ own post about metrics.
4. I can’t find a reliable statement of how Twitter’s “Average MAUs” are calculated. I’ve reached out to their Investor Relations team to ask, and will update the post if I hear back.
   Update: I heard back, here’s the definition: “We define MAUs as Twitter users who logged in or were otherwise authenticated and accessed Twitter through our website, mobile website, desktop or mobile applications, SMS or registered third-party applications or websites in the 30-day period ending on the date of measurement. Average MAUs for a period represent the average of the MAUs at the end of each month during the period. MAUs are a measure of the size of our active user base.”