3 ms·

Doesn't it seem a little odd to qualify G+ usage by public data numbers? I get that that it's what you have available, but if you used that method you'd probabl

by cheald 12y ago

Doesn't it seem a little odd to qualify G+ usage by public data numbers? I get that that it's what you have available, but if you used that method you'd probably conclude that Facebook has a fraction of the users they actually do, too. I have no doubt that G+'s active usage numbers are weaker than Google would like, but it seems like quite the chasmic leap to declare that nobody's using it based on that.

dredmorbius 12y ago
Drunks, keys, and streetlights: it's where the light is. Yes, that's a weakness of the study (I'm the numbers guy), but it's also what's available. It also allows for direct comparisons with other platforms offering public posting. And in the face of a dearth of quantified metrics from Google, offers a data point.
- mmorett 12y ago
  And there's a reason why Google is hiding or manipulating their numbers. They're poor. If they were good, you know darn well they'd be bragging about them. So they play tricks and games to try to hide that data.
  - dredmorbius 12y ago
    Oh, quite.
dredmorbius 12y ago
FYI, I've extended the initial analysis to look at the publicly inactive profiles based on "views" numbers on profiles. https://plus.google.com/104092656004159577193/posts/RhnKkfTNPKR https://plus.google.com/104092656004159577193/posts/RhnKkfTN... Upshot: The average (mean) publicly active profile has 25 times the views of the mean publicly inactive profile. 34% of non-public profiles show no views at all. The median views is 693 -- two views per day if the profile's been active for only a year, less if longer. Follower counts are far lower for publicly non-active profiles. To the 35%ile they have one follower -- for publicly active, that's the 20%ile (overall it's interesting how small follower counts are per profile). Mean followers for inactives is 6.4, for actives, ten times that -- 66.4.

by cheald 12y ago

dredmorbius 12y ago

Drunks, keys, and streetlights: it's where the light is. Yes, that's a weakness of the study (I'm the numbers guy), but it's also what's available. It also allows for direct comparisons with other platforms offering public posting. And in the face of a dearth of quantified metrics from Google, offers a data point.

mmorett 12y ago
And there's a reason why Google is hiding or manipulating their numbers. They're poor. If they were good, you know darn well they'd be bragging about them. So they play tricks and games to try to hide that data.
- dredmorbius 12y ago
  Oh, quite.

dredmorbius 12y ago

FYI, I've extended the initial analysis to look at the publicly inactive profiles based on "views" numbers on profiles. https://plus.google.com/104092656004159577193/posts/RhnKkfTNPKR https://plus.google.com/104092656004159577193/posts/RhnKkfTN... Upshot: The average (mean) publicly active profile has 25 times the views of the mean publicly inactive profile. 34% of non-public profiles show no views at all. The median views is 693 -- two views per day if the profile's been active for only a year, less if longer. Follower counts are far lower for publicly non-active profiles. To the 35%ile they have one follower -- for publicly active, that's the 20%ile (overall it's interesting how small follower counts are per profile). Mean followers for inactives is 6.4, for actives, ten times that -- 66.4.