DAU is not a strategy

Daily active users remain the polite number. They go up when you send a push. They go up when the app opens because a child sat on a phone. They go up when a QA device is left on a desk overnight. None of those movements is a strategy. They are a count of processes that reached onCreate, or the equivalent, and survived your filter — if you have a filter.

In the Product Signal Atelier we ask a blunt question in week four: if DAU fell twelve percent and every meaningful action stayed flat, would you celebrate the “efficiency”? Most rooms go quiet. That silence is the point. DAU is a volume knob on presence. Presence is not value, and it is certainly not a reason to hire, to ship, or to buy more ads.

What we put in its place depends on the product. A marketplace might lead with a completed, paid, non-refunded order with a named seller and buyer. A subscription reader might lead with a content event that lasts longer than a bounce threshold you wrote down, not one you inherited from a template. A utility that people open once a quarter should not be shamed for a low DAU; it should be instrumented for successful completion and for the absence of support tickets.

There is a practical objection: leadership grew up on DAU. Fine. Put it in an appendix. Lead with the contract metric. If the appendix and the lead diverge, write the paragraph that explains the divergence. That paragraph is App Analytics. The big number at the top of the slide is often its opposite: a refusal to choose.

I am not arguing for fewer charts. I am arguing for a first sentence that could survive a sceptical finance partner and a new engineer on the same afternoon. DAU rarely can. Activation, defined, owned, and stable for a quarter, usually can. If you cannot name that event yet, that is the work — not a prettier DAU widget.