3 ms·
There's an interesting question here about what extent inflating engagement metrics such as MAUs/DAUs is fraudulent. There are lots of subjective edge-cases - d
by avnfish 4y ago
There's an interesting question here about what extent inflating engagement metrics such as MAUs/DAUs is fraudulent. There are lots of subjective edge-cases - do you count multiple profiles as multiple users? Does _any_ engagement qualify as activity? How aggressively do you discount your metrics based on assumed botting rates?
Twitter has overstated it's DAUs previously by 1.9m per quarter[1]. If this has a direct impact on the price, is this therefore fraud? I think we wouldn't be in this situation if the rules were clearer.
[1] https://www.theverge.com/2022/4/28/23046170/twitter-miscounted-daily-users-three-years-q1-2022-earnings https://www.theverge.com/2022/4/28/23046170/twitter-miscount...
- skywhopper 4y agoThe thing is, this isn’t actually easy to figure out. There isn’t even one right answer for how to do it. And if Twitter revisited their analysis and found they were off by one percent, IMO that correction shows good faith, not attempted fraud.