Can you post some statistics on how exactly the number 200 was chosen, how does it perform, a graph that shows how many pageviews/revenues come from users with rep X ?

  • Why is this a feature-request? Any reason why I shouldn't retag this discussion?
    – C. Ross
    Commented Jan 29, 2010 at 15:08
  • This leans more toward [support].
    – random
    Commented Jan 29, 2010 at 15:14
  • @random Looking at the discussion below, I fail to see any support.
    – C. Ross
    Commented Jan 29, 2010 at 17:51
  • When the alien bursts out of the guy's chest, you can't say that the guy himself started out as an alien. @cro
    – random
    Commented Jan 30, 2010 at 4:14

3 Answers 3


It was chosen using a scientific process similar to this (all characters are figments of Atwood's imagination):

Evil Jeff: We need advertising to keep our four-chambered stomach full, Jeff.
Jeff: It's a necessary evil, I know.
Good Jeff: So let's reduce advertising for common users.
Evil Jeff: Noooo!
Good Jeff: The idea that advertisements might be reduced later will encourage participation through selfish idealism! Evil Jeff, you must agree that it will work.
Evil Jeff: I suppose. At what point should advertisements be reduced?
Interruption Duck: Quack
Jeff: How about 200 reputation points?
Good Jeff: Sounds good.
Evil Jeff: I'm late for an appointment with Dogbert.

Long story short, it's a number someone pulled out of his unicorn.

  • Good story, but I would like to see some real statistics. It's very relevant to stack exchange sites.
    – ripper234
    Commented Jan 29, 2010 at 14:00
  • ouch...sounds painful..
    – Matt
    Commented Jan 29, 2010 at 14:01
  • 5
    You seem to think my story isn't true. Well, OK. But it is based on the truth. That's practically the same thing.
    – Welbog
    Commented Jan 29, 2010 at 14:02
  • I haven't said that - I just want to see the statistics - even if they weren't involved in the original choice. How many pageviews are from users with rep < 200 right now... ?
    – ripper234
    Commented Jan 29, 2010 at 14:30

I'm pretty sure it's an arbitrary number without any scientific basis as Welbog points out, but it turns out to be a good number. By far, the majority of StackOverflow users are one time users who don't even have an account and are dragged here via Google and we want to have maximum advertisement on those users without sacrificing user experience for StackOverflow contributors.

The threshold is meant to distinguish SO "users" from "one timers" and it's pretty successful at doing that: only ~25k of registered accounts (about 20 percent of them) have 200 reputation which means the threshold is not low. It's pretty easy to get 200 rep in a day or two -- this means it's not too high.


It has been said some times that the largest group of visitors are non-users through Google.

Of course, this is nothing else than yet another appearance of the Pareto principle (the 20-80 relation for the friends). If this holds true, then the bulk of the traffic comes from non-users that visit the site, that might have a greater chance of clicking on the ads, as they haven't seen the same ad over and over again. Also, you generate goodwill from users that drive the content, that generates the non-user traffic. It's a win win.

Considering that there are now 183849 registered users (so we are not counting unregistered visitors, that are in the number of a million), and there are only 21423 users with >= 200 rep points, there is 8.5 times more people looking at the "extended" advertising than the "reduced" advertising.

  • Your avatar so purdy.
    – random
    Commented Jan 29, 2010 at 14:39
  • 'users with <= 200 rep points' should be '>=' Commented Jan 29, 2010 at 16:17
  • @Carlos: you are correct, fixed.
    – perbert
    Commented Jan 29, 2010 at 16:29

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .