Daily and most active – the new Facebook metrics
Facebook made the announced metric change to "application engagement" from "application install" yesterday. As Fred Wilson noted, they reacted quickly to the market need for real information about the number of users as opposed to installs.
Facebook has two new metrics — daily active users, and most active applications. Daily active users is the raw data — how many users actually use the application. Most active applications is a percentage metric — the percentage of users who have installed it, who are using it today. Inside Facebook notes that the two most active applications are games.
This is a good change. It will:
- Discourage the development of silly one-use widgets like Vampires and encourage the development of applications that people find genuinely useful. I myself have tons of those silly apps installed… it's a defensive mechanism to ensure that they don't fill my notifications queue with garbage.
- Favour applications that produce engagement — productivity, entertainment and games, for example.
One metric that is missing is a duration metric. If a meaningful advertising model is to be developed for facebook, a valuable piece of information would be how long the user uses the application.

August 30th, 2007 at 9:20 am
I don’t think it will discourage people from developing apps like Vampires, which is ranked in the top 50 for daily active users. Combine it with its sister apps, Zombies and Werewolves, and together they make the top 20, or about the top half-percentile of all Facebook apps. Just look through the most active users list and you’ll see silliness trumps productivity by a wide margin.
August 30th, 2007 at 9:32 am
I think you’ll start to see a split, Gary. Vampires, Zombies and Werewolves have only about 5% of users continuing to use them. They’re one-hit wonders that will be difficult to monetize. Top Friends, by comparison, has 20% repeat use. It’s much easier to make the case for advertising there.
August 31st, 2007 at 6:10 am
[...] measurement. They’re missing Attention (how much time was spent on a particular widget app, Alex agrees) and Velocity (did the application get shared and spread among a network, and Influence, who share [...]
August 31st, 2007 at 6:15 am
I think they got their terms of the measurement attributes confused, as they’re really measuring Interaction, not engagement.
details here:
http://www.web-strategist.com/blog/2007/08/31/facebook-to-rank-applications-by-interaction-not-engagement/