KDPaine & Partners proposes new un-standard for social media measurement

It seems that almost all of my conversations lately have been about the need for some sort of standard metrics for social media measurement. My response is always the same as it has been for the last 2 decades of measuring communications: There can be no standard because there is no standard goal for "communications." Yes, there may have been a standard way to count eyeballs in print and television, (and the accuracy of that can be debated) but just reaching eyeballs is seldom the goal for a good strategic communications program, and the goal always drives the selection of your measurement tool. If you're the PR Manager for Mothers Against Drunk Driving, your goal is members, contributions and ultimately fewer drunk drivers, and fewer deaths. If you're Southwest airlines, it might be ticket sales. If you're the National Association of Manufacturers, the goal is influence and access to Capital Hill. If the goal is to improve relationships with your customers, you need to measure engagement. So establishing a standard metric for such diverse goals is clearly a waste of time. Each communicator needs to decide for him or herself what success means in terms of his/her own organization.

However, when it comes to judging media, if one accepts that each organization has a slightly different approach to success, KDPaine & Partners is proposing a non-standard standard for media measurement if you will. We call it the Optimum Content Score (OCS) . The good news is that it works for both traditional and social media and enables you to track progress over time and against the competition or peer institutions. This is how it works:

You define what the perfect article, TV spot, or Radio Mention, Blog Posting , Tweet etc. looks like. Presumably it would feature your brand prominently, it would contain a key message, it would position your brand favorably on whatever issues you care about etc. etc. Not all criteria are applicable in all situations, so you pick and chose from a list that might include the following:

 

1. Visibility – where is the brand mentioned
2. Dominance – how much is the brand mentioned, does it dominate, or is just mentioned in passing
3. Messaging – does it contain one or more of your key messages? Or does it contain the opposite – a negative message or something you don't want to see in print
4. Positioning – does it position your brand favorably on issues of importance to your audience?
5. Who is quoted – are your thought leaders quoted or included?
6. Tonality – does the article/item/posting leave the reader more or less likely to do business with you, support your cause, do whatever it is that you want them to do
7. What type of conversation is taking place – there are 27 distinct types of conversation that we've identified. Understanding the nature of the conversation is critical to knowing what to do about it. Here are the 27: Presumably the most desirable would be expressing support or giving a shout-out, but again, it depends on the purpose of the program.
   1. Acknowledging receipt of information
   2. Advertising something
   3. Answering a question
   4. Asking a question
   5. Augmenting a previous post
   6. Calling for action
   7. Disclosing personal information
   8. Distributing media
   9. Expressing agreement
   10. Expressing criticism
   11. Expressing support
   12. Expressing surprise
   13. Giving a heads-up
   14. Responding to criticism
   15. Giving a shout-out
   16. Making a joke
   17. Making a suggestion
   18. Making an observation
   19. Offering a greeting
   20. Offering an opinion
   21. Putting out a wanted ad
   22. Rallying support
   23. Recruiting people
   24. Showing dismay
   25. Soliciting comments
   26. Soliciting help
   27. Starting a poll

        

        

Once you've determine what constitutes "Optimal content" you also need to decide what constitutes the opposite – the type of story you really wish would never appear.

You can then rate each item, posting, article or transcript according on a scale of + 1 to -1 depending on its content. You then average your score for the month/week/quarter (whatever timeframe is most appropriate). You will also need to do the same for your competitor's items. That way you can quickly see who is doing better or worse depending on your industry.

 

 

 

 

Comments

[anham]

I've been advocating this kind of un-standard (great concept) for quite some time now. I'm glad you put a name on this, as this might help counter the hype surrounding standards for social media measurement. In addition, standards should be seen as something of the past in a time and age when there are so many communities, media, tools, platforms, etc. To each brand, to each campaign, a unique set of metrics and indicators. However, I think our industry should agree on mutually accepted standards and metrics, such as the 27 ones you use.

Anthony Hamelle
linkfluence

[Davezilla]

28) Schadenfreude
There is no shortage of troublemakers who are not criticizing, not showing dismay, and not making a joke, but simply raising hell in a conversation for the sheer joy of it. Most YouTube, Slashdot and 4Chan comments are made by this type of person.

29) First to post!
Is this the digital equivalent of staking your flag in the soil? Is it conversation, self-verification or merely annoying? ;^)

I was going to add "Policing" (e.g., reporting a bad member on a social community), but I suppose #13, "Giving a heads up," covers that.

Here's a question, Katie. Do you consider spamming and legitimate advertising to be the same? I can't imagine clients would want to be lumped in the same group as those using unethical (or illegal) methods of marketing. How about link-baiting or trolling? Splogging? Still under #2?

(BTW, #28 and #29 — while true — were simply for your amusement. I generally involve myself in #16: Making a joke). ;^)

[Jeffrey Kahn]

I enjoyed the article and the list of 27 possible interactions. Just a observation for you to consider would be; that often we Humans are not as organized in our thinking. This is to say that multiple items on your list are often mixed into one communication for non-obvious reasons.

[Jeffrey Kahn]

I enjoyed the article and the list of 27 possible interactions. Just a observation for you to consider would be; that often we Humans are not as organized in our thinking. This is to say that multiple items on your list are often mixed into one communication for non-obvious reasons.

[Marcos Fargas]

Thank you from Spain.
The 27 types of conversation helps us to make more professional our online marketing work.

http://redescolaborativas.blogspot.com

[WebUrs]

Thanks for this post, it inspired me to write about measuring conversation in the social media space..... nearly impossible it seems

http://commetrics.com/?p=43

[Michael Blowers]

The concept of a notional match against 'perfection' holds real value in concept terms for measurement.

My only concern would be creating a perfect 'ideal' for a competitor or contemporary. For example, organisations in the same market probably won't have the same key messages. Instead could you create a sector-wide ideal against which you can compare? Hope this makes sense.

[Lindy Dreyer]

Sounds similar to the net promoter score, but I like the thought that must go into determining optimal content.

You were fabulous at BlogPotomac, BTW. Look forward to learning more from you. ;-)

[Janet Johnson]

I'll join your un-standard un-coalition for measurement!

First of all, it's stunning to see you've been able to devise a list of 27 types of conversation... upon a quick perusal, I wasn't able to add to any of them, which is exciting enough.

Then your 'sentiment factor' of +1 or -1 is something I've done in conjunction with an SEM firm successfully for a client, where we measure positive, negative and neutral coverage for Attensa (the client) and their two top competitors.

Our monthly reports have turned into something extremely actionable from a PR perspective.

You've simply taken it to a whole new level.

Cheers!