facebook moves beyond numbers, towards attention metrics

when i was reading today’s release from facebook, i was instantly reminded of a similar announcement that compete made and that i reported on four months ago. compete had announced that they were moving to a new metric (called attention) for calculating popularity on the web.

attention is a time-based measure, so logically the more time we spend on a site, the more attention we give it. we think of attention as a pie-chart - its finite - so the sites that are increasing in attention over time are performing well along this metric.

facebook has provided more details on an earlier announcement about their new way of measuring the popularity of applications that are ranked in their directory.

currently, the focus is on total number of users, but going forward it will be based on user engagement.

we define engagement as the number of users who touch your application every day (measured from midnight to midnight each day).

these touch points are:
- canvas page views
- link clicks in fbml
- mock-ajax form submission
- click-to-play flash

The number of engaged users is calculated by putting all of these touch points together.

it’s good to see that we are slowly moving beyond absolute numbers and towards interaction-based measurements. as for how this is going to affect facebook applications, the ‘zombies’ application, one of facebook’s most viral add-ons went from 3 million users to 90,000 users in one day.

Technorati Tags: compete, attention, metrics, facebook, engagement

One thought on “facebook moves beyond numbers, towards attention metrics

  1. blaine garrett

    Wow. This is an interesting read. I have seen little marketing that actually analyzes this in depth. At most I have seen hit/pageviews in conjuct with demographics, but really detecting how much time and attention people give to a site is important.

    I think one additional metric blogs and article content sites should figure out is time spent on an article. There could be some determination of how long it *should* take to read a given article, and then with the time stats, determine how much attention people actually give the article (full read, skim, bail out after first sentance, etc). Just a thought.

    Reply

Leave a Reply

Your email address will not be published. Required fields are marked *