How many members do we have?

[RLWP]

When I am representing CWDF at the National Users Forum this month, I would like to be able to say roughly how many members we are

 

This isn't actually very easy. For instance:

 

We have around 450 members on-line at any one time, each day, at peak times (8till10pm)

We generate around 550 new posts per day on the forum.

We have on average 6-8 new members join the site per day.

We have a total of just over 1,1 million posts

We have 15,888 registered members.

 

All good stuff, except the last figure.

 

That figure is the number registered account, created since we moved to the new software ten years ago, and that to register you must provide and valid email address.

 

However, I know includes every account; - active accounts, passive accounts, spam accounts, duplicate accounts, locked accounts, banned accounts, Willi GoFarr, James T Berk and Nobber, people who had one question, asked it and went away, people who only read the forum, but want to get emailed notifications, etc.

 

So, it isn't necessarily a fair reflection of the number of actual real, active members

 

Another way of looking at this:

 

We have registered members, of which;

7964 have made one or more posts

4080 have made 5 or more posts

2963 have made 10 or more posts

948 have made 100 or more posts

 

2613 have registered in the last year

2708 have posted in the last year

4473 have logged-in in the last year

 

If you take the website stats, currently the monthly figures give:

Unique visitors 90,153

Number of visits 188,235

Pages 2,488,573

Hits 12,682,340

 

 

Not all members log in to read the site. Not all members post, many just read without registering.

 

 

My feeling is that we have about 5000 active members, I would be comfortable saying that. What does anyone else think?

 

If you go to the members tab and click 'more search options' you can generate and play with the above statistics

 

 

Richard

 

p.s. Thanks to Daniel for these figures

[DeanS]

Every member boater, possibly has a wife/family who rely on this forum as much as the main member. They browse as guests, or rely on word of mouth sharing of what's been posted on the forum. I would say this forum is the Google of the Waterways.

[DHutch]

p.s. Thanks to Daniel for these figures

No worries at all Richard, as I say, I was going to post something similar myself, if more of a 'interesting to see some figures' type thread.

 

I will be interested in the replies.

 

Daniel

[junior]

It depends how accurate you want to be. If i was pitching to an external agency based on those stats, i'd be happy to say "we have between 3 and 5 thousand active members" as a headline. I'd probably try and add the total of registered users/accounts, but then elaborate that many can be spam accounts. It's also important to mention that you do not need to be registered to view and i would imagine that there is a high number of people who get their news/views/info from the forum without having a registered account.

 

If you are looking for one specific headline figure/number, i suspect that it will be difficult to be accurate.

[RLWP]

Every member boater, possibly has a wife/family who rely on this forum as much as the main member.

 

I agree with that, and the IWA, NABO, RBOA etc can claim the same thing. All of those will be represented at the NUF. So, I'd like to work out what would be a sensible figure for the number of 'users'* we have

 

Richard

 

*whatever that may mean

It depends how accurate you want to be. If i was pitching to an external agency based on those stats, i'd be happy to say "we have between 3 and 5 thousand active members" as a headline. I'd probably try and add the total of registered users/accounts, but then elaborate that many can be spam accounts. It's also important to mention that you do not need to be registered to view and i would imagine that there is a high number of people who get their news/views/info from the forum without having a registered account.

 

If you are looking for one specific headline figure/number, i suspect that it will be difficult to be accurate.

 

I know it can't be accurate, I'm very unhappy with 15,888 which is accurate, but wrong. I could easily use 948 as being guaranteed active members, and that's way too small

 

Richard

[john6767]

Looking at those figures, and assuming you just want one headline figure, I would agree that "5000 active members" would feel about right. That should cover active posters and active viewers. I think the number of posts per day figure is also one worthy of quoting as well,.

[Gingerbeer]

I would say that you have approximately 15,000 members and have an average of 7 new members per day. Like you said, some use more than one address and name and some have one name and others from the family just use that name instead of creating one of their own. I wouldn't stress over any exact numbers because they will change daily. Give a good account of what you believe it is given the information you have and stick with that.

 

Do you want them to know how many actual members you have or how many active? That is a big difference and the answer will depend on how much you want to sell this site as something people need and come to. Which sounds better in numbers? A visual always helps to get the point across so maybe a chart or graph explaining what the numbers relate to so comparisons could be made.

 

Not sure what they do at the National Users Forum get together but congrats for being the rep for it.

[RLWP]

To be honest, I want to have an appropriate level of credibility at the NUF

 

http://canalrivertrust.org.uk/news-and-views/news/canal-and-river-trust-national-users-forum

 

http://www.canalworld.net/forums/index.php?showtopic=58929

 

Richard

[Ex- Member]

If you deleted all accounts that have not used the forum for the past 6 months, then the remainder could be the figure. If you then did this for the next 6 months you would have a true reflection of actual membership.

 

You could always send an automatic e-mail to members who's accounts have been inactive for 6 months and ask them to log in to confirm their continuing subscription.

 

Like all figures and statistics though you can get them to say exactly what you want with clever manipulation, as government does on a daily basis.

[RLWP]

Like all figures and statistics though you can get them to say exactly what you want with clever manipulation, as government does on a daily basis.

 

This is one of the reasons why I'm asking this question openly rather than cooking up something clever by database analysis.

 

I wonder if we could generate the figures you suggest from the data that we have.

 

Richard

[DHutch]

This is one of the reasons why I'm asking this question openly rather than cooking up something clever by database analysis.

Which I fully support, and if nothing else, highlights the openness that is rife, if not exclusively or uniquely existent, within the site.

 

I agree with that, and the IWA, NABO, RBOA etc can claim the same thing. All of those will be represented at the NUF. So, I'd like to work out what would be a sensible figure for the number of 'users'* we have

Although to counter that, some groups such as the IWA have 'partner' or 'family' members, and we do not know how they count them.

- Obviously some families here have more than one account. The options go on!

 

If you deleted all accounts that have not used the forum for the past 6 months, then the remainder could be the figure. If you then did this for the next 6 months you would have a true reflection of actual membership.

Pruning the member database is an option, however as it effectively costs nothing to keep them (the 'members' table within the database is already several orders of magnitudes smaller than the 'posts' table) while I have considered attempting to assess how many accounts may be 'spam generate' and remove them, I do not feel I wish to remove dominate members who may return, or pester them we emails. Of nothing else, even emailing them all, is a fair task and load on the server, compared to keeping them.

 

 

 

Its also worth pointing out that, although it is very hard to determine one from the other, and hence there is some doubt about how many of the 16k registered members where registered by bots for spamming I am fairly firmly of the belief from what I have seen that the are quite defiantly the minority, if still perhaps a significant number. I'm going out on a limb here, but personally from what I have seen, I would put the figure between, say 5% and 25%.

 

Posts, and members with posts, on the other hand I would but well into the 90% figure, maybe in excess of 99%, as the number of spam posts in several orders of magnitudes below that of real posts, and almost all spam posts and associated accounts are removed from the data.

 

 

 

Daniel

[Ex- Member]

Although to counter that, some groups such as the IWA have 'partner' or 'family' members, and we do not know how they count them.

- Obviously some families here have more than one account. The options go on!

Quite possible looking back at your list of figures. My brain has to be in the mood for such calculatory work though LOL So won't have a hope in hell at the moment LOL

[john6767]

If you deleted all accounts that have not used the forum for the past 6 months, then the remainder could be the figure. If you then did this for the next 6 months you would have a true reflection of actual membership.

 

You could always send an automatic e-mail to members who's accounts have been inactive for 6 months and ask them to log in to confirm their continuing subscription.

 

Like all figures and statistics though you can get them to say exactly what you want with clever manipulation, as government does on a daily basis.

If you look at how many members there are who have posted in the last 6 months, I make it 1975, the number who have logged in in the last 6 months 2850. So the number logging in is not a massive percentage of the total, and I would bet if you required re-registration if you have not posted for 6 months then a very large number of that 15000 would disappear.

 

Interestingly if you qualify those figure to only count users who have made 5 or more posts (in total) the numbers are 1500 and 1656 respectively, ie much closer together.

 

So you could argue* that perhaps the 1656 number more accurately reflects active users.

 

* you can make the numbers say whatever you want really....

[DHutch]

I think one of the issues,

I would say that you have approximately 15,000 members and have an average of 7 new members per day.

I think one of the issues is, a large proportion of the people/groups that will be present will or do not have a very good working grasp of the demographics or even function of a forum. As well as, not really knowing how big canal world forums are.

 

At which point you can either come up with a nice firm, rounded, finger in the air, type figure, which they can understand, but can or may dispute. Or else you can hit them with cold, hard, indisputable, factual numbers, which they may well not understand, and can still dispute the relevance of, even if they don't know a thing about them or forums!

 

Some sites/organisations quote the number of 'paying members' along side other metrics, which is a nice figure, but as we work quite hard to keep the forum free for anyone to join, use, and read, we cant!

 

 

 

If I was going to say something, I would be more included to say something like;

 

"We have around (5),000 active users, (2.4),000 of which have (posted) within (a year) who contributing (550)(posts) per(day) with a total number of 1.1 million posts and 16,000 registrations in the last ten years."

 

Or a condensed version of that. And if you want, you can then point them to connect you, myself or look on the site for further stats.

 

One thing I am planning, which in a draft form or otherwise could be posted up before the the meeting, is a page of the static site on the forums history/stats.

 

 

Daniel

[Paul C]

One post, a year ago, does not an active member make!

 

I'd say the most useful metrics would be active posters in the past week, 2 weeks and 3 weeks. This doesn't take into account members/visitors who might/might not be logged in but read the forum. If you're going to include these, you might as well use page hits, which is a fairly standard internet way of gauging the popularity of a website.

[Tony Brooks]

I am very tempted to suggest that we should use CaRTs apparent methods of calculating visitors and multiply the number of members by the number of times they visit the site per year (I know that is not possible because some of them will remain logged on as I do).

 

I think I would then account for 365 x 3 visits (often more) = over 1000 visits. We should end up with millions

[Ex- Member]

Ok Trying to get brain ion gear. These stats below could actually say quite a lot, depending of course on the interpretation of a member. Personally I think that 6 months inactivity on any site would indicate that the user is no longer what could be described as a member.

 

 

We have registered members, of which;
7964 have made one or more posts
4080 have made 5 or more posts
2963 have made 10 or more posts
948 have made 100 or more posts

 

7964 have made one or more posts. So on this statistic when the posts were made is important as if the posts made are within the past 6 months then class those as a member.

 

I'm assuming the top figure is an amalgamation of the 3 sets of figures below. Meaning that the 4080 have that made 5 or more posts are also part of the 7964 have made one or more posts.

 

Interesting that that represents almost exactly 50% of registered members stopped posting which left 4080 that have made 5 or more posts. Then 75% of those continue to post 10 or more posts. Around 8% in total look to have become regular posters.

 

 

 

I would assess from this that of the 7964 have made one or more posts. Find out how many posted more than 10 posts in the past 6 months and that will be a fair representation. Of course though IMO based on my 6 month premise.

 

I meant to add LOLOL my ball park guess is 1935 members.

Edited September 1, 2013 by Julynian

[RLWP]

One post, a year ago, does not an active member make!

 

I'd say the most useful metrics would be active posters in the past week, 2 weeks and 3 weeks. This doesn't take into account members/visitors who might/might not be logged in but read the forum. If you're going to include these, you might as well use page hits, which is a fairly standard internet way of gauging the popularity of a website.

 

12,682,340 page hits per month is a tricky figure to work with

 

Richard

[Paul C]

 

12,682,340 page hits per month is a tricky figure to work with

 

Richard

 

Its as meaningful, and comparable, as any other statistic. You could only directly compare the forum/website to something equivalent, like another canal forum, or possibly another canal website. You can't really invent what you consider a "member" to be then compare it to eg paid membership of an organisation like NABO or IWA, for example.

[Tesla]

Personally I think the number of topics, posts and views per week or month are a good measure, the fact is this is a very active forum.

 

The number of views is perhaps of most interest, showing that a great number of people are following what members say, even if only a handful contribute? Like it or not, the posters can nfluence a lot of people over time.

[DHutch]

If you look at how many members there are who have posted in the last 6 months, I make it 1975, the number who have logged in in the last 6 months 2850.

So the number logging in is not a massive percentage of the total, and I would bet if you required re-registration if you have not posted for 6 months then a very large number of that 15000 would disappear.

I would agree that forcing anyone who had not taken active part in the site in the last year/6months would significantly reduce the numbers.

 

So you could argue* that perhaps the 1656 number more accurately reflects active users.

I would agree with that also, up to a point, which is that you/me/we and forums in general dont have a very good handle of what 'active' means.

- Even including bot sign ups is certainly not a passive process to register with the site and validate an email, certainly not to then go on to make a post, and or sign in once a year. And we are then still in the case of not knowing how many are reading the site without having been registered, which is the sole basis a static website can use to quote its user size.

 

At which point the thing I would be trying to do in the context is consider how 'active' the membership of the other organisations is.

 

For instance, I am a member of the IWA, having joined a long time back when booking the boat in for a national rally and ticking the 'sign me up automatically using this information' for whence my sole activity is to not remove the standing order from my account, and bin the quarterly magazine from time to time, often only unwrapping it from the packaging to place the two in the right recycling bin. In ten years myself and the boat have been to two national and one campaign rally.

 

 

One post, a year ago, does not an active member make!

But yet as said allowing a £20/yr direct debit, five years ago, does make you appear on the IWA member figures, as an example.

I'm also an equally (non) active member of the HNBOC, and the caldon canal trust, and several others.

 

 

Daniel

[Ex- Member]

I think reasonable to describe someone as an active user they must be making at least 1 post in 6 months.

 

With regard to viewers, my dad read the forum avidly but never posted and don't think registered, I guess many do the same. Problem is you also get bots and single visits. Also It's surprising sometimes when i'm searching for something on google and not necessarily boat related that canalworld forum is on the first page somewhere relating to my search, the forum must get 1000's of hits from this effect.

[DHutch]

I'm assuming the top figure is an amalgamation of the 3 sets of figures below. Meaning that the 4080 have that made 5 or more posts are also part of the 7964 have made one or more posts.

Yes. Although as said, you can generate most of the figures, and of those figures yourself, including if you with 6 (or 18) month figures.

 

Ok Trying to get brain ion gear. These stats below could actually say quite a lot, depending of course on the interpretation of a member. Personally I think that 6 months inactivity on any site would indicate that the user is no longer what could be described as a member.

You see again, while im almost more interested in talking about it than coming to a figure, I would disagree with that.

 

I have a lot of hobbies, and cant do them all at once, so tend to go in cycles and often ditch something for 6 months over winter, or a year, or 18months till the following summer. But would still very much consider that to remain a hobby, myself a member of the associated club, etc.

 

If you swap 'member' for 'active user' I may be more included to agree, but still think something like '3 or more posts in 18 months' is a fair than '1 minimum post in the last 6 months' but it still then leaves you to debate which figure is more representative if you trying to compare your site to another canal organisation or group.

[RLWP]

Also It's surprising sometimes when i'm searching for something on google and not necessarily boat related that canalworld forum is on the first page somewhere relating to my search, the forum must get 1000's of hits from this effect.

 

Yes, I searched for National User Forum and got a link to here first. The thread had been indexed by Google within an hour of me posting to it

 

Richard

[DHutch]

 

This doesn't take into account members/visitors who might/might not be logged in but read the forum. If you're going to include these, you might as well use page hits, which is a fairly standard internet way of gauging the popularity of a website.

12,682,340 page hits per month is a tricky figure to work with

Its as meaningful, and comparable, as any other statistic. You could only directly compare the forum/website to something equivalent, like another canal forum, or possibly another canal website. You can't really invent what you consider a "member" to be then compare it to eg paid membership of an organisation like NABO or IWA, for example.

It might be worth breifly at this point, while whe're talking about comparing ourselves to other websites, to bring up the 'UK waterways website' which ranks opted-in waterways websites. The site tells you to place its code on every page of your site, which almost all of the other sites and the only other forum listed do, however when we implemented this on our site our traffic was such that we unintentionally and unknowingly took down their site. After an exchange of emails with the owner, we now have the link just on the index pages (not even, the 'find new content' page) and still sit comfortably at the top!

 

 

 

Daniel