running amok....

WBahn

Joined Mar 31, 2012
29,929
I'm seeing in the XF admin panel that June 2014 had 1099 new members. We are at 1877 so far this month. The re-design plays a large part because there is a "Sign-up" button now on the main site, which gets 2/3rds of the site's traffic.
Of course, half a month's data is statistically irrelevant.
Need to look at year/year numbers just before the change and just after the change.
 

tcmtech

Joined Nov 4, 2013
2,867
The way I see forum numbers is more like a scene of an accident.

How many are just lookers and how many are actual participants?

Lots of lookers does not mean there was a substantial increase in participation when an event occurred. ;)
 

WBahn

Joined Mar 31, 2012
29,929
Thanks, Adam.

While determining what it means is always problematic, I would definitely agree that it is a substantial difference and is "probably" significant.
 

jrap

Joined Jun 25, 2006
1,125
I have no way of checking that @shortbus. We'll just wait a few months and see if the trend continues.

Thanks
The way I see forum numbers is more like a scene of an accident.

How many are just lookers and how many are actual participants?

Lots of lookers does not mean there was a substantial increase in participation when an event occurred. ;)

The way I see it, the first step in getting a lurker to participate is getting them through registration.
 

tcmtech

Joined Nov 4, 2013
2,867
So as the admin do you have any breakdown statistics that show the overall ratios of members to post count?

Say how many members there are in the 1 - 10 posts and those in the 10 - 50 and 50 - 100 groupings or the like?

To me that would give a fairly accurate breakdown of the overall real membership numbers and relevant overall participation.
 

WBahn

Joined Mar 31, 2012
29,929
So as the admin do you have any breakdown statistics that show the overall ratios of members to post count?

Say how many members there are in the 1 - 10 posts and those in the 10 - 50 and 50 - 100 groupings or the like?

To me that would give a fairly accurate breakdown of the overall real membership numbers and relevant overall participation.
You (as in any member) use to be able to get those numbers from the Members pages. Sadly you can't any more.

Like most forums, a very large fraction of people that have made a post only make that one post and then disappear.

Here's a post of how things split out back when we crossed 30,000 members:

http://forum.allaboutcircuits.com/threads/30-000-members-strong-aac-tidbits.71809/

I suspect things haven't changed much in terms of percentages. The median number of posts is probably still 2.

The top ten post counts total 134,624 and the second ten are just under half of that at 65,814. So the fall off is still very close to exponential.

If we subtract the OT and some other administrative forums that aren't tallied in the post counts, we have roughly 750,000 posts in the database. So the top ten account for 18% of all posts and the top twenty account for about 27%.

I suspect that anyone that has posted at least 100 times is near the top 2% of all posters and making your 10th post puts you in the top 20% or so.

While I'm guessing that the top 100 posters likely account for 1/2 to maybe even 2/3 of all posts, that is only part of the story. Much of vibrant content is driven by the people that post once or twice and the follow-up discussions that results among the regular posters. They provide the seeds that the rest of us nurture.
 

tcmtech

Joined Nov 4, 2013
2,867
So basically that says that when one or two big gun poster members here decides to leave they in fact carry far more post count and activity value than what the raw members numbers would imply. :eek:

I sort of figured as much. :(

To me this all looks very similar to when a church decides to claim anyone who ever signed the guestbook for having been there even once for any reason whatsoever as member so they can say they have over 20,000 members.
However in reality when the Sunday services roll around there are maybe 20 - 30 actual people who are consistently ever there and maybe 10 - 20 more that occasionally float in and out over the year putting the real supporting congregational membership at less than 50 and not 20,000.
 

WBahn

Joined Mar 31, 2012
29,929
Yep, the numbers can be misleading and can be used to mislead. The "correct" number can be hard to determine, in part because the number that needs to be used depends on what it is being used for. To use your church example, you might have very different numbers if you are trying to decide how many chairs the sanctuary needs versus how many newsletters you need to print.
 

JoeJester

Joined Apr 26, 2005
4,390
TCM,

If you visit http://www.forumdr.com you can see some metrics that can be used. Personally, I view a 92 day period. The last time I did that at AAC was in 2013. I was going to do it again at the same time in 2014, only jrap, who didn't know I was doing that, threw a curveball and changed forums and I lost the ability to grab the information. I never recorded the email addresses as they are none of my business and not germane to what I was doing.

During that 92 day period, there were 3008 members who visited the site. All had at least 1 post. Of the 3008 members, 1706 had a membership longer than 92 days. Of the remaining 1302 members, 1090 didn't last 30 days. Of that figure, 532 didn't last one day. If you look up "internet culture" you would find the 90-9-1 rule. Basically it states that 90 percent lurk, 9 percent are casual posters, and 1 percent post a lot.

One group of metrics common to some websites and forums are:
- Increasing the number of people to view your site
- Increasing the number of people to register at your site
- Increasing people to post in your forum ... raising the numbers in the 9 pct and 1 pct category.

Some forums won't let you see the attachments without registering.

So, how does one get the first step done? Content. Content that people search for and are interested in.

I haven't been to diyaudio lately, but that is a niche forum. They do not diversify from that theme. The only one I know that has a diversified, yet electronics niche, is EEVBlog and their forum. Dave Jones full time job is that site and the associated youtube videos.
 

Georacer

Joined Nov 25, 2009
5,182
While I'm guessing that the top 100 posters likely account for 1/2 to maybe even 2/3 of all posts, that is only part of the story. Much of vibrant content is driven by the people that post once or twice and the follow-up discussions that results among the regular posters. They provide the seeds that the rest of us nurture.
So basically that says that when one or two big gun poster members here decides to leave they in fact carry far more post count and activity value than what the raw members numbers would imply. :eek:
These two quotes reveal the delicate balance a forum administration must hold, in order to have a healthy and vibrant online community.
 

JoeJester

Joined Apr 26, 2005
4,390
Another metric I used at a small forum (< 500 members) in a very specific niche was posts per membership day. In that area, I categorized the posters in the following manner:
- Lurkers - those with zero posts
- One and done - those that were "ordered" to post as part of an initiation ritual
- casual poster - those who posts per membership day is less than the average minus twice the deviation.
- active poster - those who posts per membership day fall above the casual poster but less than the average plus twice the deviation
- hyperactive poster - those who posts per membership day is above the average plus twice the deviation.

I know Bill M doesn't like the hyperactive poster moniker, so I think in his honor, I will rename that category prolific poster.
 
Top