Community Data Question


(outofthebox) #1

Hi @richard_millington,

Thank you for the fantastic blog post today on community member projections using actual data. Very helpful too - you’re going to save people tons of money and generate accountability in this space.

I have a few questions to help me understand and utilize this information in a more accurate way.

One question (knowing that this data is an approximation/aggregation) - you have in, say, Year 3, a total of 750,000 visitors coming from direct search traffic to the community. That’s nearly as many visitors as the company’s homepage (1.25M). It is also a 15x increase from Year 1. Do you find that to be a pretty normal rise in search results?

Does that lead to an implication that a community manager should highly prioritize identifying, empowering, and supporting community members who can make mega quality posts that drive incredible search traffic? Or is this more typical for high volume communities that have a ton of questions (e.g., customer support) and they win the long tail? Or…?

Alternatively, is the more important number that on average, whatever the search traffic looks like, you’re going to have a .3-.7% conversion rate of those search visitors who register to join the community?

And, similarly, however the website, newsletter, and social numbers are doing for the rest of the company, we should expect about 3-5% of the loyal followers of a company will be interested in joining the community?

The other number that stands out to me as a relative constant in this space is to expect an annual churn rate of 75%. Is that defined as the number of people who post in one year who do not post the next year?

Thank you for leading the way and raising the bar for all of us.

(Richard Millington) #2

Hi @outofthebox

The short answer to all of your questions is "it depends "

But I’l try to give a slightly more helpful answer too.

It varies wildly depending upon so many factors like the quantity of content the community is producing, whether the market is growing/mature, how popular the organization is within that market, how the community is setup, etc…A 15x increase from pretty much nothing is reasonable, but what you really need are some idea of how popular the trends are at a meta level first.

This really depends on your goals. Mega quality posts can drive search traffic but so can hundreds of long-tail search posts too. Customer support communities are probably leaning to the latter.

You have to define importance here. Conversion rates can vary from 0.01% to 10%. So you need to put in place what that benchmark looks like.

Again, depends on the company and the product. A telecom company will be lower than a video game, for example. Also, joining is only a small part of the real battle. Retention is the challenge.

You might want to check in with @Bas_van_Leeuwen about what an overall average retention/churn rate looks like. I suspect 75% might be a bit low here. But then some people can reappear after a year or so.

(outofthebox) #3

That’s very helpful!

Do you see any consistent correlation between # of active members and total # of posts?

Similarly, between # of ‘super users’ and # of super user posts? (Perhaps this is definitional in some communities, a super user posts more than X times/year?).

For popularity at a meta level, is the best data in Google Trends? Or…?

(Richard Millington) #4

Correlation? Absolutely…at least up to a point.

Yes, popularity based upon the size of the topic - keyword analysis etc…

(outofthebox) #5

What do you see as the normal range for number of posts for an active member?

(Richard Millington) #6

It depends what you mean by a normal range. If we discount lurkers, then I think a typical IQR is something close to 1 to 15. But this varies wildly by the type of community you’re building. Customer support will be a lot lower.

Here’s some client data.