Online community metrics: numbers you need to pay attention to

by Martin Reed on 15 December 2009 in Articles

metrics for online communities

I’m surprised people haven’t pulled me up on this yet – I often state that member count isn’t a reliable indicator of the success of an online community; but nobody has asked me what numbers are better to use. Here, I’ll offer you some suggestions.


This is a tricky one. On the one hand, an increasing number of new members implies that your community is attractive to outsiders. You need to bear in mind though, that the faster you attract new members, the more diluted your community can become. A sudden increase in new members can actually damage an online community – so be careful.


Without activity, you have no community. There are many ways you can measure contributions to your online community; number of posts per member, overall posts per day, number of posts per discussion thread, etc.

Generally speaking, the more contributions to your community, the more value it gives (and receives). However, not all contributions are equal – is a 5 word post worth as much as a 100 word post?

Don’t assume that the higher the number of posts per day, the better your community is – those posts could be abusive or contain little in the way of value.

Bounce rate

Your site’s bounce rate shows how many of your visitors leave without clicking through to any other pages of your website. For example, if the bounce rate of your homepage is 50%, it means that half of your visitors leave your site as soon as they arrive.

You need to be careful here, though. Remember that not all traffic arrives via your homepage – so don’t rely solely on the bounce rate of your homepage as an indicator of how attractive/relevant your community is to visitors. Monitor the bounce rate of the most popular landing pages of your site. Also bear in mind that a sudden influx of traffic will almost certainly increase (worsen) your bounce rate – for example, if a page hits the homepage of Digg, you’ll typically see a huge influx of traffic to that specific page but very little in the way of visitor exploration of your site.

I’d suggest that if your bounce rate is consistently above 50%, you need to work on the site – either you’re not explaining the community’s purpose, you’re not offering anything unique, you’re advertising in the wrong places, your community is too hidden, or you’ve gone for form over function. I don’t think you could go more basic than the homepage for Just Chat – and its bounce rate is only 20%.


Be careful on this one – measuring the health of your online community with pageviews can easily be a deceptive ego stroke, like member count. If you want to double your pageviews overnight, you can simply edit your forum configuration to hold 10 posts per page instead of 20, or edit your blog configuration to display only 10 blog comments per page instead of 20.

However, as long as you’re aware of the potential pitfalls of using pageviews as a metric, it can still be useful. Not only do pageviews reflect how engaged your visitors are, they can show you where any bottlenecks may be. Are there a high number of pageviews for a specific section of your site? Look to reduce them – bring more content onto the page.

People don’t like clicking ‘next’ over and over again – that’s why you’ll see people jump into a forum discussion thread at page 30 with a comment like, ‘I couldn’t be bothered to read the previous 30 pages, but…’. Sure, even if all the content is spread over fewer pages, you’ll still have people skip over the content, but you’ll be more likely to see them skim some of it. It’s all about making it easier for the visitor.

Time on site

Pretty simple this one. Basically, the longer a visitor is on your site, the more engaged they are. You want this number to be high, and to increase over time (as the amount of content in your community increases and relationships get stronger).

Repeat visitors

Your community depends on its members returning to get involved in discussions. Fresh blood is good, but your community won’t get anywhere unless people stick around and form relationships. The more repeat visitors your online community sees, the healthier it is.

Visitor loyalty

This is a good measure of just how addictive your online community is. If visitors are only dropping by once a month, then it isn’t interesting enough. If people are visiting multiple times per day, they’re hooked.


People don’t recommend rubbish to their friends – they only recommend stuff that’s good. Every time someone uses your ‘tell a friend’ form (you do have one, right?), it’s a vote of confidence in the community. Make sure you are measuring the use of this form.

Do you offer a link for visitors and members to share content from your community? The more content is being shared, the more engaging and interesting it is.

Newsletter response rate

Most email newsletters are done wrong – they are used as a way of broadcasting messages to subscribers. Instead, they should be seen as conversation starters. Don’t send emails from a ‘donotreply’ address – use a real, live email address. Encourage people to hit the ‘reply’ button. You want to encourage conversation, not discourage it.

If people are simply reading your newsletter and not clicking any links or getting in touch, then something is wrong. Either your community is boring, or your newsletters are.

Incoming links and online mentions

Buzz is good. The more people talking about you, the more attention your community will attract. Not all mentions are guaranteed to be positive, though – so don’t just count a mention as a vote for your community. If you do something bad, you might get thousands of mentions; is that a good thing?

I’d keep track of all mentions and incoming links and divide them between negative, neutral and positive and see how they trend.

* Be wary of playing the numbers game *

In this article, I’ve listed 10 ways you can use numbers to measure your community’s progress – but at the same time I’ve shown how they don’t always reveal the true picture. When it comes to online communities, numbers aren’t everything. A community is about relationships – and relationships are subjective, not objective. Don’t forget this key fact.

Share this community building advice


Similar Posts

Previous post:

Next post:


Ann Arbor December 15, 2009 at 4:18 pm

Your first point about registrations seemed a little counter-intuitive to me, but then I read on and found out what you mean. Makes a lot of sense. Like so many things in life and on the internet, there is a balance that must be kept.

Mr Woc December 15, 2009 at 5:28 pm

Hi there

Some good stuff on there, I have recently been working on my bounce rate too, interesting to see yours is only 20% which is excelllent !

Does the bounce rate only show the people hitting and leaving ur homepage though if your using google analytics ?


Nick van der Vaart December 15, 2009 at 6:05 pm

Excellent piece of research. Thanks.

Peter Renier December 15, 2009 at 6:26 pm

Hi all

Statistical data is imperative to understanding how well your website is doing. Google Analytics is still the best (IMO) system out there.

@ Mr Woc – I think bounce rate is a visitor landing on a page (any page) and leaving that same page. – of course the page has to have the tracking script…

Greg December 15, 2009 at 6:42 pm

As always a very informative and genuine post!

I hear people talk all the time about how “big” and “successful” their community is but fact is their registrations and number of visitors is not a reflection of a succesfful community at all. Visitors come and go… Registrations roll in all the time… But are they actually building a community? No!

Martin Reed - Community Manager December 15, 2009 at 9:24 pm

Mr Woc – As Peter suggests, you can view the bounce rate for any page you like in Google Analytics via the ‘Content’ category.

Daryl Wilkes December 16, 2009 at 5:35 am


another excellent article. I didnt have a tell a friend form………..I will have one VERY soon!

thanks again.


jms December 16, 2009 at 8:22 am

Martin – I’m not sure any of these are good metrics, they might be: but it very much goes back to the first point on your previous post – – Why are you doing it? The metrics should start from there.

By way of example – Dell recently published some information on why twitter had been a success for them and it very simply came down to orders. Yes they had people following them but their measure of success was in $ revenue generated by referral.

Martin Reed - Community Manager December 16, 2009 at 10:51 am

jms – I agree that deciding on what metrics you’ll use should take place during the planning process. If you don’t think any of my suggestions are good metrics, please could you suggest some of your own? I’d love to hear them.

jms December 17, 2009 at 6:18 am

Martin – I think the point i was trying to make was that they were not good per se – they may be good in the context of specific objectives… i.e. if part of your objective of using sm is to get people to sign up for your newsletter then newsletter signups are great.

I was thinking later on about the same thing – somebody had commented that a safety releated facebook campaign had only got 500 friends and was therefore not a success – my argument would be that the purpose of the campaign was to change a particular behaviour in a specific group – the measurement should focus on the level of change not the number of friends. If the campaign is repeated its the number of people whos behaviour has changed that should be the metric not the number of people who friended the group or read the posts… (obviously th more readers the higher the probability but…)

I hope this clarifies.

Nicole Price December 17, 2009 at 8:35 am

This should be treated as a Bible for Community blogs. Every single item is important from the point of view of the visitor and that is what makes it important.

Jonathan December 18, 2009 at 3:48 pm

The numbers are important and generally a crowd attracts a crowd, that is unless it’s the wrong crowd that you are attracting in the first place. Personally my experience has been to encourage high quality contributions by offering prizes for the best comments, as this way new members will seek to emulate this, rather than posting one comment threads and replies!

Angela Connor December 19, 2009 at 10:37 pm

Good advice Martin. This is such an important area and the importance of specific stats can vary from community to community. If it is easy for trolls to create new profiles when they are booted and you don’t ban IP’s for instance, your registration numbers could be skewed. Of 10,000 members, how many could actually be recreated troll ID’s? I pay close attention to the number of blogposts created daily and monthly and the comments and image galleries. I am really trying to increase the number of image galleries and trying various tactics to increase submissions so measuring my success is crucial. Again, great post.

Martin Reed - Community Manager December 21, 2009 at 11:34 am

Angela – Good point about trolls and spambots; here’s a good stat for you. The email penpals section of Just Chat would have 7,000 more ‘members’ if I didn’t manually vet all new applications and delete the spammers and scammers.

I love that you’re focussing on image galleries – photos are hugely undervalued but extremely valuable in online communities.

Megan December 22, 2009 at 1:32 am

Time on site, repeat visitors, and contribution numbers are probably the most important “numbers” to look out for.

David January 6, 2010 at 5:30 am

Free tools like google analytics and statcounter can help a site owner understand visitor metrics. As mentioned in the article it’s quality what counts over quantity

Beth January 12, 2010 at 8:50 pm

These are very good measures.

I am wondering what the success rate might be of folks attempting to start a community, and actually succeeding.

And, once the community has been established what might be laudable goals to set.

I mean, what we have is a goal of setting up a community. OK, I get that.

But, then I ask myself the next question. For what purpose? How long is a community’s average life span? What are good examples of people who have started communities who have transition it to something that can support a full time effort if that is the goal.

I am not talking about someone with lots of cash to spend. There are believe it or not about 5% of the population who can lay $50,000 down to start an online community anytime they want.

These are the people who own 95% of all the stocks on the New York Stock Exchange. These are the same 5% who own 40% of everything ownable in the US.

So, I am asking, as a poor person with only my fingers and a machine, what are my chances of starting a community, and staying with it long enough to do the members and myself any real good?

Has anyone tracked this sort of thing?

Martin Reed - Community Manager January 13, 2010 at 2:29 pm

Beth – In 2008, Deloitte found that 35% of the online communities they studied have less than 100 members and less than 25% have more than 1,000 members. All this despite the fact that close to 6% of the businesses studied had spent over $1 million on their community projects.

I think amateurs are better at building online communities than corporations, and for many reasons. Often, the amateur has the passion – they’re motivated by sharing and relationships rather than money. Secondly, amateurs can take on more risks when promoting and building the community. Thirdly, amateurs aren’t restricted by as many rules and don’t have to jump through hoops to get approvals for their strategy or ideas.

You don’t need money to build a successful online community – you just need passion, enthusiasm and a strong work ethic.

Mike February 2, 2010 at 11:46 pm

Totally agree with time on site. I own a community that was fairly popular but the time on site was a bit low. I did a bit of investigation and it turned out that some users were extremely immature and turned off a lot of visitors. Even though they were amongst my most active posters I had to give them an ultimatum. It worked a little (not entirely sure if it had an effect.) I’m still working to improve this metric.

Brian February 9, 2010 at 8:35 pm

I think you’re right that one needs to examine a larger number of metrics to determine the health of the online community. The “time on site” metric is very basic however does not lie. New registrations and also total membership are true metrics. If you are charging a membership fee each month then your revenue and trend are indicators.

steve Millerton February 12, 2010 at 4:31 pm

The two metrics that i focus on is bounce rate and time spent on the site. For me this is the best way to find out what visitors are doing once they reach the site. Obviously the longer they spent on the site the better chance of making a conversion

Sean March 1, 2010 at 3:56 pm

Mike: I had the same situation. I allowed members actually to begin writing content for a site I worked at. All was good till this one member, who wrote five great articles before this, stuffed hidden links into new content. I couldn’t believe it. All the links were pointing to non-relevant sites so I had to confront the guy. He denied it….. so weird since you can just look at the html for the article. I had to use the ban hammer!!

Kate June 22, 2010 at 7:46 am

Nice work!
Found a lot of familiar points, but it was interesting to read about ‘Newsletter response rate’ and ‘Endorsements’.
Thank you!

Jean Martin July 30, 2010 at 4:46 am

Very informative article. I think that this list of metrics is universal for measuring all kinds of internet activity, being this community , an online shop, a newsletter, a blog etc.
I always use Google Analytics to get statistics for my internet projects.

{ 1 trackback }