Is DAU an important metric?

EA revealed last week during its earnings call that it would no longer report DAU and MAU metrics in quarterly earnings statements. EA’s CFO, Blake Jorgensen, explained the decision pithily with:

“We’re going to stop talking about monthly or weekly or daily active users because I’m not sure what anyone can do with that information.”

The decision to stop reporting on user counts is certainly an interesting one that defies industry norms. But withholding these numbers from shareholders and other interested parties is not unprecedented; Microsoft stopped releasing XBox live MAU metrics in 2019.

It’s easy and sensible to view any selective suppression of metrics as an approach to hiding bad news; after all, public companies theoretically are rewarded when they share good news. But DAU (and MAU, and WAU — any simple binary measure of user engagement with the product over some period) is a tricky metric in that it tends to be singularly focused on as a proxy for the overall health of a product or service. Especially for mobile apps, analysts — and often, generalist analysts that don’t specialize in a given vertical — have been trained to equate slowing DAU growth or DAU deterioration as a sign that a company is failing.

It certainly could be true that a company occupies troubled waters if its DAU is shrinking, but it’s not necessarily true, and it’s tough to make a nuanced case as to why that is when interested parties (read: public markets investors) are reflexively discouraged by DAU values that are declining, or flat, or even increasing-but-not-to-expectations.

Part of that nuance is a recognition that not all users are created equal: as I point out in The only growth metric that matters, a product could increase its DAU and not actually grow, at least in the long-term sense. Consider that the Day 0 retention value for every single cohort is 100%; if a company wants to inflate its DAU numbers, it can simply acquire very large cohorts of cheap traffic, every single member of which will add to DAU on the day they are acquired. Run such an acquisition program for a week each month, and MAU numbers are miraculously increased whereas revenue likely stays flat and marketing expense increases but potentially not substantially.

What’s more important than the mere presence of a user in a product on some day or in some month is what they actually do within that product and how often they come back. A user could interact with a product every day or just once per month and contribute the same value to an MAU metric; very many non-paying users could interact with an app multiple times per day habitually or just once ever and contribute the same value to a revenue metric. User counts alone don’t reveal anything about the commercial health of a product.

In Monthly churn is a terrible metric, I visualized the impact that changing retention rates can have on overall DAU levels; if cohorts are being acquired that are less retentive than previous cohorts, at some point, that degradation in retention will impact top-line DAU. But again, because of the disproportionate impact that DNU (daily new users) has on DAU versus cohorts any older than Day 0, aggressive acquisition can mask underlying problems with retention.

Moreover, no competent management or marketing team would operate in such a way as to specifically maximize DAU independent of all other metrics; DAU increasing might be a symptom of a larger, successful strategy, but the fact that DAU increases in and of itself reveals nothing about marketing efficiency, or the product’s revenue trajectory, or retention and engagement. If management wouldn’t look to a metric in assessing its strategy, should anyone else?

That said — investors need to look at something in evaluating a company’s performance over time, and DAU is easy to understand. Eliminating DAU altogether is probably a worse signal than whatever the change in DAU may have implied. And for internal analysis, DAU provides an obvious “What?” starting point from which to ask the important “Why?” questions. I’ve never seen a company ignore DAU in their internal reporting outright, since DAU should be fairly predictable given retention rates in a way that makes flagging emerging problems very easy. So while DAU shouldn’t be looked to exclusively as an indicator of health, it’s a good starting point for investigating problems and understanding the dynamics of a business.

Photo by Nathan Dumlao on Unsplash