This is how I measure social stuff
by asi
I’ve been thinking quite a lot about measurement of social marketing activities lately and I think I might be on something. I’m far from solving all the challenges of measuring ‘digital’ and even further to suggest ways to measure the effect of digital on the bottom line (e.g “Driving Purchase Intent” etc.). But I think we’ve developed an interesting combination of areas and metrics that can potentially tell a good story on the effectiveness of our social activities and might give a good indication on ROI (although attaching VALUES to these metrics and benchmarking are still very problematic) .
This has been developed through the work on the cadbury’s going fairtrade blog and other HyperSocial projects. There is actually nothing original in this measurement model. It is built on existing methods and tie together the things we are all talking about for some time now, so credit and thanks to all of you that contributed through conversations, blogs, wikis etc
This model fits best to social projects like company blogs, social networking presence (i.e. facebook fan page) or even the good old microsite and can work on both permanent projects and more short term, campaign-based activities. You just need to tweak and attach the right metrics that tells the right story on your project. And that’s what it exactly does – it tells a story that help you and your client to evaluate what you are doing. Next step is to start building KPIs into these metrics but that’s for another post.
We look at four interrelated areas:
1. Reach: Simply put, we are talking – is anyone listening?
This is the most basic area of measurement and one which is the easiest to attach value to (and the one that naturally client cares the most about). Under reach you measure the relevant metrics of traffic.
* Unique visitors,
* Returns,
* pages views
* A deeper level of performance under reach is conversion to friends and/or RSS subscribers. These are your loyal people.
2. Engagement: Are we interesting? Are we doing something people want to interact with?
This category measures everything that happen on-site and indicates that you are doing something interesting or fun. Look for metrics that indicate that people like what you’re doing:
* Dwell time – do people stay long enough or are they shooting off quickly?
* Conversation rate (comments per post or posts on wall etc.)
* Clicks/UV’s if there are deeper areas
* Uploads / downloads
* Other things you wish people to do – plays, interactions (e.g poll participation, votes etc)
3. Influence: What is the impact we make on the wider web?
Here we simply ask – are we interesting enough for people to talk ABOUT us? Are they sharing us with their friends? The words Buzz and spreadable give me a seriously bad sensation in the back of my throat, but are to common to snub for the sake of this post. You can think of engagement as metrics for ‘owned’ vs. influence/impact as metrics for ‘earned’. Look for any metric/indication that tells you that you create impact off-site in relevant communities. We have to be really honest with ourselves as to who is talking about us, in which contexts and is it part of our objectives ((does this creates positive impact on the brand or on the agency that created it? ).
* Trackbacks – a behaviour in decline but still a good indicator
* Sharing (through Twitter, Social Networks, blogs, forums)
* Mentions on mainstream media (earned, not paid for)
* Social bookmarking (delicious, digg, reddit etc)
* Better visibility on search
4. Sentiment: Are we making people happy? Do they like us?
Here we measure both quantitatively and qualitatively how are we affecting people. This should be measured both on-site and off-site (you can theoretically have very positive comments on your own space but some nasty, cynical conversations outside in the wider web – you want to measure both). Simply count and analyse all relevant traces people leave behind or conversation they create elsewhere:
On-site: (usually fits with the metrics for engagement)
* comments,
* email they sent you
Off-site:
All conversations about you and mentions of you (fits nicely with the metrics for influence -blogosphere, twitterverse, forums etc)
Simplez hey?
Obviously each project deserves a relevant set of metrics depends on your objectives but i think this framework gives you a god start. As i said this is just a first step. Attaching agreeable values and attaching reasonable KPIs is still a bitch and in many cases are still founded on guesswork. But for a long term projects, you can have an overall targets and as you go along you learn about the natural potential of your project. Try to figure out your baseline and work from there – more on this later.
So what do you think? Your feedback is hugely appreciated.
Hey Asi,
I think one of the only things I would question is the relevance of dwell time – given the rise of tabbed browsers (With IE now having it as standard as well as the rise of Firefox), tabs may be opened and not checked until later, could this skew that particular measure?
It’s awesome to see similar types of things in terms of measurement popping up at the same time – this kind of measurement stuff is pretty much standard in every idea we pitch to clients, love the synchronicity of it all.
Great post and thanks for putting all the stuff in one place
Excellent post Asi!
I’m currently revising for the Interactive Media Strategies exam and this list will provide a great starting point for the ‘evaluation of online campaigns’ section.
I’ll be sure to share this with fellow students!
Hey,
Really great post and excellent attempt to pin down something that really is so hard to measure.
I think, however, there is a bit of a gap, which is action. Surely the point of doing any of this stuff is to effect the bottom line (increasing sales – either from new customers or existing users) so I think there is some value in looking at sales volumes (and values where appropriate) and seeing if there’s any correlation. Obviously this is massively dependent on the type of product or company, but there are ways to monitor where sales came from and I could see potential examples of how this can be linked back to social media campaigns.
nice post, man.
the bit on owned vs earned reminds me of the way a football player’s contribution is sometimes measured. I.e. Goals and Assists.
I’ve always preferred players who were all about the assists:
John Barnes, Peter Beardsley, Paul Scholes
Thus ends my light metaphor
Keep up the good work.
This is super-useful. Thanks for sharing.
[...] Asi Sharabi has created a great 4 step guideline for this: [...]
great post
working in media this is what we are looking at. But the really interesting thing is when you take brand data by week and co-relate it against the difference between running a TV or press campaign and someone downloading, embedding or passing on something
then social media will get its real value
[...] initiatives will be benchmarked against other marketing spends, we will be chasing our own tails with limited success. Moreover, as long as budgets will come straight out of marketing budgets we are bound for the same [...]