Twitter became the latest of the big digital media platforms to restate the audience estimates it reports to advertisers and investors. In a third quarter letter to shareholders, Twitter disclosed that it has overstated its monthly average user (MAU) counts for three years.
Twitter said the restatement was due to incorrectly recording “1-2 million users per quarter” from certain “third-party applications” to Twitter’s MAU count.
Twitter said the readjustment does not impact its daily average user counts, and added that due to its “data retention policies” it could not “reconcile periods prior to the fourth quarter of 2016.”
Twitter reported 330 million MAUs for the third quarter of 2017, and said its “adjusted year-over-year growth rate” is a 4.1% increase over the third quarter of 2016.
The revelation, coming at the end of a year of similar restatements by big digital rivals such as Facebook and Google, sparked a trigger of news reports, a well as a tweetstorm on Twitter itself.
How is it possible that off-platform users can affect the Monthly Active Users (MAU) but not the Daily Active Users (DAU)?
Surely if someone was active on Twitter at some stage during a month then they HAD to have been active on at least one of those days during the month.
Maybe they meant to say "does not SIGNIFICANTLY affect daily average user counts".
But consider the implications of such users not significantly affecting the daily data, but affecting the monthly data enough for Twitter to go public. For example there must be a lot of users who tweet/read tweets very irreguarly (at least once a month but not much more).
Now consider how big, say, TV could get its numbers to be if they included people who viewed TV for just one second each month. Thank goodness that TV is independently measure by a third-party, and the markets insistence on using 'average minute audience' as the currency. It's about time other media grew up and came clean and used validated third-party data reporting on 'average' audiences rather than pumped-up 'cumulative' audiences based on miniscule thresholds.