5

I understand GFH is a new beta, and I should stop looking at the site stats, sheesh and all that.

Nevertheless I am interested in seeing the statistics on the Area51 page as they change over time. While GFH may not be ready for primetime on day 90 it still might be useful to see how the site is growing, the pace of various statistics, the effect of a promotion or conference visit, etc.

Is this possible?

1
  • 3
    Now you wish that you had known to ask that question 50 days ago when you could have set up an external tool to capture the data!
    – Fortiter
    Nov 30, 2012 at 3:47

4 Answers 4

Reset to default
3

No, I'm sorry. That is not currently possible. The moderators appointed to your site (and later elected) will have access to some general analytics to help convey the overall health of the site to the community, but the specifics are not generally shared with the public. That will likely change as we can build more comprehensive tools, but there are currently no historical tools built into the Area 51 analytics.

3
  • It sounds like better stats are on the roadmap. I hope so. It would enhance a community's ability to manage its growth correctly.
    – user47
    Nov 30, 2012 at 21:35
  • @JustinY It's going to be a part of what I generally call community education — having more visibility and insight into where there is and is not generally enough activity and attention.
    – Robert Cartaino
    Nov 30, 2012 at 22:08
  • @RobertCartaino - you are director of community development, so I assume you have access to the 'general analytics to help convey the overall health of the site to the community'. Could you do that - ie convey to us our overall health? Something simple like - 'you are doing fine' would alleviate any fears anyone has.
    – Duncan
    Dec 3, 2012 at 4:06
5

One stat that is visible over time is people/visit/pageviews. Here is a sample from Oct-9 thru Nov-28:

enter image description here

enter image description here

enter image description here

I also kept some pdf's of the area 51 genealogy home page, but I didn't notice it never kept the front page (so I don't have the Q/day) but I do have:

+--------+------------+----------------------+-------------+
|  Date  | Avid Users | 90day Avid User Pace | Total Users |
+--------+------------+----------------------+-------------+
| 10-Nov |         39 |                   65 |         446 |
| 15-Nov |         40 |                   64 |         455 |
| 19-Nov |         42 |                   65 |         463 |
| 24-Nov |         44 |                   66 |         483 |
| 29-Nov |         45 |                   64 |         495 |
+--------+------------+----------------------+-------------+
6
  • 3
    To let you know, the spike in October was due to two October 22nd blogs posts: Dick Eastman and Randy Seaver: blog.eogn.com/eastmans_online_genealogy/2012/10/… and geneamusings.com/2012/10/…
    – lkessler
    Nov 30, 2012 at 4:02
  • 2
    This is exactly the kind of reason to obtain and analyze this data - to figure out what messages sent out to the community actually get heard, and what the overall effect is upon the site in the hearing.
    – fbrereto
    Nov 30, 2012 at 7:06
  • In the interest of maintaining a Proof Standard, we should assert that the spike was "associated with" two blog posts. Any claim of causality would require more evidence.
    – Fortiter
    Nov 30, 2012 at 7:07
  • @Fortiter you are hardcore. It's awesome.
    – fbrereto
    Nov 30, 2012 at 7:09
  • 2
    If we can attribute an order of magnitude increase in visits (on one day) to a particular message that would be interesting. Of more value is the fact that the apparent medium-term impact of whatever caused it has been NIL.
    – Fortiter
    Nov 30, 2012 at 7:09
  • @Fortiter: I don't have access to StackExchange's server logs to prove that particular spike, but other postings made by Dick and Randy when they used my bit.ly links for Area 51 which I can track, have given me definitive information about traffic generated from their site.
    – lkessler
    Nov 30, 2012 at 14:32
3

Whatever data is available needs to be interpreted cautiously. We could become very enthusiastic about strategies that apparently have strong data support that actually show very little impact.

The graphs provided by Duncan show that something was phenomenally successful in driving traffic approximately 40 days ago. What the graphs do not show (because you need to extract other data to do so) is that the additional 2000 visits produced 18 new user registrations over that three day period. Of those 18 people, ONE has been "seen" on the site after 25 October.

Initial visits (eyeballs) are great. Registrations (clicks) are better. Regular users are what builds the site. Then we can move on to consider activity levels (voting, editing, asking, answering).

It is too easy to get excited about essentially meaningless statistics just because they can be gathered.

1
  • +1 excellent observation on the users the spike attracted. That's why search engines and organic growth seem best. They attract already interested people.
    – lkessler
    Dec 1, 2012 at 16:33
1

It is possible. Starting today, go to the GFH page at Area 51 and copy the stats you want into a spreadsheet. Do that again tomorrow. And the day after. ...

It should take you only about a minute a day if you really wanted to do it.

4
  • Ah, but it requires discipline, something which I lack - just ask my wife! I would much rather have a computer do a mundane job such as this anyhow.
    – fbrereto
    Nov 30, 2012 at 7:07
  • 3
    @fbrereto: If you're interested in the information provided by a mundane task, it will become a task that is no longer mundane, but one that is looked on with anticipation.
    – lkessler
    Nov 30, 2012 at 14:33
  • .@lkessler Touché!
    – fbrereto
    Nov 30, 2012 at 17:41
  • That actually is what I was doing -albeit only occasionally. I have not idea why my 'print to file' only did pages 2 onwards so I unfortunately never recorded some of the data.
    – Duncan
    Dec 3, 2012 at 3:59

You must log in to answer this question.

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