My Photo

Enter your email address:

Delivered by FeedBurner

My Other Accounts

Flickr LinkedIn Twitter

« Upcoming Events | Main | It's not Whack-A-Mole »

May 29, 2008

TrackBack

TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a00e5501a78c5883400e552a558d38834

Listed below are links to weblogs that reference Community Velocity: A Metric for Measurement:

» Measuring Community Velocity from Servant of Chaos
When you are involved in a community you can really get a sense of its health. You know when it is active and you know when it is in decline. Think about the social networks that you use, think through [Read More]

Comments

Feed You can follow this conversation by subscribing to the comment feed for this post.

Rawn Shah

Hi Rachel,

I'll ask around in some of our communities to see if they can get the data, and also float the idea to the metrics workgroup to see what they think.
-rawn

Rachel Happe

Rawn - that would be great - would love to get your team's input on this.

samantha

Cool. Thanks for the post.

Steve Kuhn

Hi Rachel.

Hi Rachel.

Do you think it would be worthwhile to have a variable for consumption (reads), in addition to your variable for production (posts)? I'm thinking that there are a lot of lurkers in any organization, reading but not posting, but obviously the lurkers and the organization get benefit from the consumption of communication in social media.

Marco ten Vaanholt

Rachel,

A few comments:
- Addressable market might be an ambiguous term. Is it total amount of internet users ? Is it total amount of people in an already existent ecosystem, or is it also total amount beyond the existent ecosystem... how would you capture that ? Needs some more eplanation or definition for me to become concrete

In your formula, are you adding all the numbers up ? or do you put weight factors for each in?

You also are not including any quality numbers either somehow.. you can have a super big community ( e.g. yahoo, 50M+ members, billions of posts and page views, but how valuable is the community going at this speed ? - needs some thoughts )

What does active mean ? Needs further explanation

what does velocity mean ? Needs further explanation

Hopefully I gave you some more thoughts

M

JC John SESE Cuneta

Great post!

I agree with M. As a CM myself, I'm always point out that community success is not about the number of members but quality members. Are these quality members active? What is an active member anyway?

It's hard to quantify a Community specially if you are talking to a your bosses who still think traditional strategies.

Kim Kobza

To share some first impressions:

(1) The type of community for which the measurement applies would be very important for the statistical accuracy of the metric. So for instance, we might measure the behaviors differently in closed, open or hybrid networks. The metric described above would probably be most appropriate in an open community. My thought is that metrics expressed as ratios would be most accurate in closed communities, or at least, most capable of measurement for exactly the reason that you described - it is easiest to define the market.

(2) The number of posts should be coupled with some measure of views to provide a true measure. Let me use an example. If you go to http://ratemyspace.hgtv.com/ and search on NASCAR. You will find 13 posts that have 20,000 plus all time views. If similarly you go to racing one at http://community.racingone.com/galleries/mostviewed you will find 43 pages of posts that probably don't have 20 or 30% of the total posts on rate my space.

That being said, both communities are strong for different reasons. If you measured all of the blog posts, etc., on racingone, you would get a number in the many tens of thousands.

The concept of velocity, and I would argue sustainability is very important. Are members, or even unique visitors, consistently active? A community is not successful is there is no repeatability. So the ratio is on the right track.

Maybe the way we do this is to develop a series of ratios.

One of the key determinants in this set of measurements is going to be how many calls to action are developed, and in what way are they delivered? The interesting thing to know would be whether the relationship between calls to action and community activity is linear and direct, or exponential and parabolic.

We need to give this more thought. The industry clearly needs measurements of interactivity.

Other key measurements:

(1) How often do members, invite other members or participants? How often do they recommend people or data within the community such as videos or other forms of content? Would measure how viral a community is - this seems like a very key concept. A viral index. We will work on that.

(2) Also, there should probably be a trust index. How do we measure trust?

(3) Frequency of visits in relationship to content contributed would also be another measure.

(4) One of the items that we look at is the percentage of professionally produced content to user generated content.

We'll work on some of these indices and contribute. K.

Rachel Happe

Wow - great feedback and the comments bring up some really important points. Here's what I'm understand as the other considerations:

- Quality
- Counting views as well as contributions
- Repeat activity vs. one time activity
- Viral/members inviting others
- Trust

Great additions. One thing to keep in mind is this metric absolutely cannot do everything and I don't feel like the issue of quality and trust can really be incorporated. Do they need to be tracked? Absolutely...but it is so variable what that means across different communities that I think it is impossible to capture in a benchmarking index.

In regards to measuring viewed items - I think that matters more or less for different types of communities and in my mind, a robust community must have increasing contributions by members - otherwise it is a broadcast channel. And I think it is important to track view rates but not a key indication of a real community. Feel free to disagree with me on that point.

The viral nature is also important - but again to a lesser or great extent depending on the type of community. And, all communities don't have an 'invite' feature so...I'm not sure it is universally trackable. The increase in membership month over month should account for this somewhat (although not directly).

Repeat activity vs. one time activity is an interesting addition - how would you incorporate it into this metric? Suggestions?

Some definitional responses to questions:

- Active member - a member who has done something (posted, rated, etc) during the period being analyzed
- TAM (total addressable market). The number of members you estimate you could have in the community if everyone in the market joined.(the market could be total employees for internal communities, could be millions of consumers in a brand community). Hard to estimate in some cases but I think worthwhile in understanding penetration percentages.

This metric is intended to be something that every community can calculate both to see their own progress over time - with regards to increasing reach & participation - and a metric that can be used to benchmark communities against a group of other communities. That issue necessitates having a fairly simplistic set of measures that everyone can get or estimate and a normalizer (hence the TAM). I didn't add weightings so the absolute activity (total active members) and the concentration of activity (posts per member per period) are currently equally important.

Thanks everyone for the great comments and contributions. They bring up questions of whether we can develop some other metrics for reach, virality or quality (any suggestions on how to measure that?)

Rachel

James Young

Do you have any guidance on what the progression should look like on a time line? I mean, .01 at what point in the life of the community? .3 at what time?

Rachel Happe

Hi James -

That is a great question... unfortunately I don't have enough data to say (and it likely varies by type of community). For large enterprise communities ramp will be in the range of 18 months - 2 years.

Rachel

The comments to this entry are closed.

My Other Project

Twitter Chatter

    follow me on Twitter

    People You Should Know