PropertyChat 2018 stats

Now that we're well into the new year, I thought it would be interesting to look at some stats from 2018.

I've decided to do two separate sets of stats each year:

1. "calendar year" stats at the end of the year, looking at what happened during the previous year
2. "anniversary" stats in June, looking back over the full history of the site

So let's start with some basics.

1. Posts per day in 2018 (28 day moving averages)



... we had a strong start to the year, but I think uncertainty with APRA and the obviously cooling property markets around the country are having an impact. We're not getting anywhere near as much "where to buy" discussion, but are getting quite a bit more "property market economics" discussion - which is to be expected.

Let's compare that to the last two years:

Posts per day 2016-18



A definite drop from previous years. I would think the number of new members asking questions about where to buy has dropped significantly.

 

2. Posts per month in 2018

Another view of the same data.



December is always quiet on PropertyChat.

 

3. Posts per year



Sharp drop off in 2018 - I'm a bit surprised it dropped so much actually!

Interestingly, visitor traffic to the site is actually up quite strongly in 2018.

We're getting more visitors, but they aren't posting as much. Lot's of people waiting on the sidelines to see what happens I expect.

 

4. New user registrations per day in 2018 (28 day moving average - "valid" users only)



Fairly consistent throughout the year really.

Let's compare to previous years:

New user registrations per day 2016-18



Registrations have actually been fairly strong this year.

 

5. New user registrations per month in 2018 (valid users only)

Same data taken monthly

 

6. New user registrations per year (valid users only)



Only a 7% drop from 2017 - still pretty good.

 

7. User status 2018 (all users)

• Total registrations up until the end of 2018: 18,144
• Valid: 15,294 (84.3%) - can post
• Banned: 1,221 (6.7%) - rule breaches or spammer
• Invalid: 1,378 (7.6%) - can't post (account deactivated; email bounced; account not activated)
• Removed: 251 (1.4%) - old invalid or banned users who have never posted and who have been purged from the database

 

8. Current user status 2018

• Total current users: 16,923
• Active users: 7,908 (46.7%) - can post and have posted
• Never posted: 7,386 (43.6%) - can post, but never have (lurker!)
• Email bounced: 653 (3.9%) - email has bounced back as undeliverable
• Incomplete registration: 492 (2.9%) - user has not confirmed their email address after registering
• Never logged in: 199 (1.2%) - user registered, but did not confirm their email address and never logged in again
• Spammer: 181 (1.1%) - was banned for spamming
• Banned: 70 (0.4%) - user was banned for rule breaches
• Email changed: 34 (0.2%) - user changed their email address, but never confirmed

 

If it has been purged from the database, then yes - the username is available for use.

It wouldn't matter, since we only purge users who have never posted.

 

Most people are quiet achievers

ta
rolf

 

I did the queries again only including members who had made at least 100 posts, but the results were almost identical in their shape (just slightly lower overall, naturally).

I then looked at "top" users only:

1. I calculated 80% of the total posts, which is about 515,000 posts
2. I ranked all users who have posted at least once by their post count (highest to lowest)
3. I then ran a cumulative total for each user showing how many posts they have contributed to the total (again, starting with the user with the highest post count)
4. I then went down the list of users and found the number where they have collectively contributed 80% of the posts.
5. That number works out to be about 450 users - who have contributed a minimum of around 200 posts each
6. In other words, the top 450 users have contributed 80% of the posts to the site

... so I did the queries again including only users with 200 posts or more - but they were still largely the same shape are previously.

Some of the peaks were a little lower - indicating that those peaks were driven by new members, but the overall data largely follows the same pattern.



Each of the peaks I've highlighted are slightly lower (ie distance between peak and trough) than the corresponding peaks in the original chart.

To me, the patterns indicate to me that new members posting drives a lot of the discussion - someone asks a question, and then the more established members contributed a lot more posts as a result.

I think the main reason for the drops this year will be the lack of newer members posting questions - thus the established members simply don't have as much to post about (there's only so many doom-and-gloom threads in the economics forum you can contribute to).

Interestingly, when I looked at an extreme example and only looked at the top 10 posters - there was still a broad pattern there.

So how about we flip it around and look at the bottom 80% of posters (ie exclude the top 20%)

The top 13 members have contributed 131,851 posts which is 20.5% of the overall total.



This chart is quite a bit flatter than the others - however, I'm not sure this really proves anything either way.

As I mentioned - most posts come as a result of a new thread that has been started - someone starts a thread and then people respond and discuss. On rare occasions, a thread may continue for a long period - but the posts in those very long threads is actually only a very small percentage of the overall thread count.

I think new threads is actually going to be a more meaningful metric - so I'll work on that next.

 

Definitely on the todo list.

 

A lot of lurkers who are just here to learn.

There's also a lot of people who create an account, visit a few times and then forget about it and never come back.

Or perhaps they were looking for some specific information and signed up because they found something useful - but then didn't have any other questions and so haven't been back.

There's lots of different reasons why people don't post or become inactive.

You'll find this happens on most sites (forums in particular) - the "headline" user count is made up of mostly inactive members who never interact with the site, and haven't logged in for years.

Even Facebook is full of people who never or rarely post (my wife almost never posts on Facebook, but does check fairly regularly to keep in touch with what's happening with her extended family). I have plenty of friends who used to post but got sick of Facebook and simply don't use it anymore. Many younger people have abandoned Facebook for alternative platforms like instagram, snapchat and so on.