> Twitter has said in its SEC filings that it estimates that fewer than 5% of its monetizable daily active users are “false or spam accounts”
Is this different to total percentage of false or spam accounts? The monetizable daily active users seems like a very specific subpopulation of total twitter accounts.
Yes, it is crucially different. mDAU is defined in the SEC filings which you can easily check. Basically it boils down to, there's say a billion twitter accounts. A shitload are inactive, or obviously bots, or joke accounts, or actual alts (don't double-count), or API users, or people using Tweetdeck, or doing all sorts of other things that Twitter can easily detect are not resulting in eyeballs on ads. Twitter boils down the number of "users that use Twitter clients that display ads" down to "just" 230M/day. Twitter then claims that, of those users, who are mostly just every day people logging in, maybe liking a post, scrolling a bit, seeing an ad on a real Twitter client -- of those users, when they sample them, they find that <5% of them are bots (like headless browsers and whatnot). That is obviously a completely different measurement than just overall bot accounts, or how often you will personally see bots posting, or how many of your followers are bots. Totally unrelated numbers.
Musk knows this but still knowingly confuses them to try to trick people into thinking Twitter is lying about that 5% number (since it's pretty obvious there are many bots on twitter)
Is this different to total percentage of false or spam accounts? The monetizable daily active users seems like a very specific subpopulation of total twitter accounts.