The Tom Sawyer Effect

There has been a lot of chat this week about social media measurement and ROI. There’s always a lot of talk about social media measurement and ROI, but this week Twitter, the blogosphere and London Digital Week has been gushing with it. Or rather squeezing. Trying very hard, but not really getting much more out than we had before. Sorry…

Econsultancy warn us that some companies are better set up than others for social media engagement and think that we will hit  upon better ways to measure as brands get more dependant on them. Which is a bit of a cop out, but valid to point out that social media is essentially the new customer service and without it most retail and service brands will just die. However, an earlier post on direct approaches for measurement was more useful…

Essentially with all of these conversations we just seem to be covering the same old ground, just gradually welcoming more people into it. But at least as we do so we should move away from the current situation. Last year we were working with a global measurement agency (that really shouldnt be allowed to remain nameless) to evaluate  a pan-European integrated campaign. They were suggesting using a standard value for every site, regardless of where it was on the site, whether it was editorial or multimedia, how targeted it was, how many people responded or what they thought.

At the Imagination event How was it for you? a panel mostly formed of content and entertainment leaders (channel 4, covent garden opera, mobile content and digital content) kicked over the dust for an hour – we need to measure all of the areas our brand is being discussed, we need to relinquish control of the content etc etc – before we started getting somewhere.

Louise  Brown, head of new media for Channel 4 surmised that people only have a certain amount of time to spread across all activities, let alone media. And so if we can measure how much of this time you have managed to grab you’ve got a universal benchmark for measurement.

The trouble is, this exercise is almost for its own sake. Time might matter to a broadcaster, but for a retailer, or a service, an event, even a charitable campaign, what matters is who signs up, who goes to the event, who buys. And it is difficult to tell at the moment how much of this is driven by chat.

She made another interesting point around this, which is that for Skins the users who follow them on more than a couple of their channels – broadcast, website, facebook, bebo my space, twitter, messenger – are so involved that they will do any request made of them. We didnt find out what these requests were, and of course most people dont have such a great asset as Skins to play with, but interesting that depth of engagement can lead so directly to power of response.

essentially there are two take-aways from the whole debate for me:

1) that each campaign needs to be evaluated on its own merit. or rather, on what provides the most merit (or value) to the brand. the whole idea of a cost per acquisition needs to be extended to a cost per engagement, per recommendation, per visit etc depending on what the likelihood of a conversion

2) that once we start to implement these more robustly measured programmes we need to encourage everyone to share, as a comparison with a similar brand or campaign is the best benchmark besides direct returns. and essentially it will always be these competitors who are vying for your audiences eyes and paws.

There are some places where this sharing is already happening, most notably the monthly meet-up and wiki Measurement Camp and  ex Forester analyst Peter Kim’s wiki, which has now expanded to a third list to accommodate nearly 1,500 examples…

Then for the first bit we are talking to the lovely people at E.Life to trial to put together specific and relevant criteria blending traffic, engagement and discussion and overall sentiment changes as well as any measurable actions. As ever, more to come, but for now more links of varying usefulness can be found on my delicious page…