Search Engine Journal recently compiled a very helpful list of social media metrics that are ideal for community managers who are trying to measure the performance of their activities.
One metric that seems to be missing is a rate or score of responsiveness. Historically we’re used to discussing click through rates, conversion rates, cost per acquisition or cost per impression.
We’ve been experimenting with a few response scores ourselves and a score that seems to work can be calculated by dividing two data sources: volume of response generated and volume content created.
At any point in time a community manager will be tweeting, blogging and producing other types of content. This content then receives a response in the form of retweets, sharing, coverage or an increase in brand or content related conversation.
Comparing output Vs. Response using a score starts to give us an idea of how good our content is. If the content is good then the score should be high!
How could this thinking be applied to a real situation?
Let’s say we know a community manager called Justin. Justin works for London Pies Inc. London Pies Inc make the most sublime pastry covered treats on this side of the M25.
This week Justin created:
- 2 recipe videos
- Left 8 comments on food blogs
- Tweeted 50 times
- Uploaded 10 photos
In total Justin created 70 pieces of content.
Whilst trying make sense of how this week’s content performed in comparison to last week Justin dips into:
- The company’s Buzz monitoring tool
- Tweet deck to count RTs and mentions
- YouTube and Flickr to check views
Once a response score is applied to weekly figures we can start to indentify the weeks where the content has been exceptionally good:
Just by looking at the response curve, generated by weekly scores, we can see that week two and Justin’s last week have generated the best response because they have low volumes of content generation and high volumes of response.
This could also be a useful way to track the merits of different types of content. For example, John could look at the proportion of content created on YouTube, Flickr, etc. and compare that to the overall response rate. It may help him figure out what type of content is most worth his effort for the response achieved.
Interesting thought… But I think that’s a good method for valuating the relevance of the content you create. I mean, week 2 your responsive rate per content was 340/40 = 8,5 and week 4 it was 300/70=4,26. So the content created on week 2 was much more interesting for users.
well.. afterall you can use the inverse function to see the relevance of the content and this function to see the response curve. But in order to better predict outcomes, I think each analysis complements the other. Or even… if you could detail the types of contents created you could compare it to the average response rate per piece.
there’s a lot to talk about this subject hehe
good luck!