-
I find these metrics... doubtful and the impact they describe is statistically ambiguous. After 3 days 80% of what? Compared to what? journalismfest/918364571232063488Permalink On twitter.com
❤️ 2 Favorites
Mood -1 🙁
-
In my investigation of Facebook metrics I noted that traffic continues to propagate only if an article receives traffic >=50% of day beforePermalink On twitter.com
Mood 0
-
After day 3 we could be talking about an 80% decrease only applied to ~13% of the total traffic an article received on the first day.Permalink On twitter.com
Mood 0
-
If an article received 1000 hits day one, 500 day two and 250 day 3, this impact would drop day 4 from 125 to 25...Permalink On twitter.com
Mood -1 🙁
-
But that's likely too simple, the impact is probably even lower, projected over anticipated traffic over multiple days.Permalink On twitter.com
❤️ 1 Favorite
Mood 0
-
And then you still have to take Facebook's capacity to project traffic accurately as a given, which also seems doubtful...Permalink On twitter.com
Mood -1 🙁
-
Though we do *know* how companies like Facebook often attempt to collect these types of stats... By A/B testing.Permalink On twitter.com
Mood +2 🙂
-
Though that would imply Facebook knowingly distributed Fake News with no warning.Permalink On twitter.com
Mood -7 🙁
-
Also, reporters have gotten good at noticing that sort of thing, I think it is more likely FB is projecting impact over miniscule long tailOn twitter.com
Mood +3 🙂