I decided to write a long post as nothing that took place at the Influence Scorecard Summit needs to be kept back – but I’m still trying to wrap my hands around what we came up with, in the first meet (we’ll meet again in second quarter of 2010, probably in New York, San Francisco or London).  Frequently, as  I write my ideas down in this blog, the ideas crystallize and evolve – maybe that’s what will happen, here.

The working group that attended the Influencer Scorecard Summit included Katie Delahaye Paine, Marshall Sponder, Ted Shelton, Jay Krall, Harish Rao, Diane Meier, Connie Bensen, Philip Sheldrake, though we had a few more who attended but were not present on day 2.

We also have an open Wiki for collaboration, thanks to Ted Shelton.

While the slides I’m presenting are in no particular order, we attempted to define, for our own working knowledge, what influence, engagement, advocacy, sentiment, ROI, transparency, authentic and trust meant; these were not absolute definitions, but those we could all agree on when referencing those terms.

We decided Influence was the power to affect someone’s actions – and that makes sense to me (I’ve often thought this, but it’s not always easy to prove – but we were not going to get caught up proving this, we just wanted to agree on what we call “influence”).   Engagement was defined as some action more than “0″; essentially, moving from one state to another (we didn’t define what “0″ was).   Advocacy we agreed was “engagement” with an “agenda” while Sentiment is taken to be an expression of an opinion.  We didn’t attempt to define Return on Investment for the time being.  Transparency we defined as the “default state” of deciding to allow all information to be publicly accessible (unless there is a specific reason to keep information private); we discussed the implications of companies operating with their default state of  public vs the default state of treating all internal information as  private (the norm).

It would seem to make sense, I think,  if your into Social Media, you’d want to have your default state of transparency to be set to “public” and only keep things “private” if there is a reason – and I’m sure this issue of “transparency” can be debated, one way or the other.

Authenticity we defined as “genuine” and “real” while “Trust” is made up of competence, integrity, reliability and authenticity.

We treated “Engagement” as a “breadcrumb”, a sign, trace, crumb to which there was “attention” applied – and we postulated “Engagement” could be “visible” or “invisible”; linking Engagement to Forrester’s Technographic profiles (via GroundSwell) seems to make sense to us, but we dropped out “inactives”.  We inserted the idea of “passion” without defining yet how it would be measured – though passion probably has to do with “acceleration” of “engagement”.

We didn’t come up with a formula (yet) though we probably could have, but it felt premature – and quite frankly, many in attendance felt that most formulas for Engagement contain “compound measures” which aren’t measurable/provable and end up creating more confusion and controversy without leading to much that’s useful  (in my own words).

However, at some point, we’ll need a formula for Engagement, based on our definitions, otherwise, we won’t be able to operate on the data we collect or refine (correct) the formula by applying it to real use cases using “transparent algorithms”.  I think it can be argued that the current formulas for Engagement are, or are not, transparent (I’m thinking of Peterson’s which would seem to be transparent, but also other others, that are harder to apply to an actual data set because some aspect is missing, and you have to put in a “0″, and that’s not what we want.   Maybe we could come up with a formula for Engagement, even in our meetings yesterday and today, but that doesn’t mean we should.

In defining “Engagement” (and we’re trying for a model that includes “offline” as well as “online”) we looked at bell curves (see below), we also postulated that “positive” or “negative” sentiment was at one side of the bell curve, or the other, and “neutral” was in the middle of the curve – and most social media data that’s collected, typically, will be in the middle of the bell curve (neutral) but what we end up caring most about is the what’s at either side of the bell curve – as those things impact business the most.

The particular bell curve shown above doesn’t really tie into sentiment – but a similar one could have been show that does tie in to sentiment (we discussed this though we may not have drawn out a diagram for it).

In defining “Influence“, I began to visualize a “map” or “diagram” and even though I give David Armano (and sometimes, Brian Solis) an hard time because I’m sick of pretty marketing diagrams that are next to meaningless when you try to operate on the information – I opted and voiced an opinion that we needed a map for Influence, and Philip Sheldrake will produce it by year’s end based on our notes (below).

The “map” of “Influence” includes “topical values”, “demographics” and “volume/reach”, but we defined reach as being subordinate to relevance, and my take is we’d only want to determine the reach an Influencer has/had with a “relevant” audience for the message.   I don’t think many, or any Social Media platforms really have reach calculated as best it could be – because the audience would have to be determined to be “relevant” to the topic query, first.

One example I can think of is a restaurant chain I do some work with – they want to reach the target audience (people who will come in to one of their restaurants and order a meal, hopefully, return often to repeat the meal) – the “reach” metric will be more useful if we can tie it to those individuals or groups most likely to do so – but no social media platform can really do that, today – therefore many of the definitions we came up with might not be fully implementable, today (though they probably will be within a few years).

We talked about “Network Effects” and equated it to “Word Of Mouth” and “amplification” of messaging; Harish Rao argued that there is a lot of “Network Theory”  and proven equations that could be overlay to our working definitions, even now, though it is probably premature to do so at this time.   We also discussed “potential” engagement vs. “actual” or “post” engagement (don’t recall what we cited as an example of it – probably Katie Paine could speak to it better than I can).

We looked at “Relevance” as topical, possibly obtained with Qualitative data, and this may be possible using Social Media Monitoring (I was thinking Crimson Hexagon, but they didn’t attend – wish they had – tried my best to get them to the table – that’s all I can do).  Credibility was defined by source and message – again, a practical example might be finding someone who has demonstrated knowledge in a specific area that’s relevant to what is to be influenced.

Finally, we left today with some action point, with my preference to create a real use case, beyond a hypothetical one- and for that, we’d need real data (we’re working on it).

Of course, we don’t have the formula for “Influencers” mapped out yet – though it’s strongly suggested by the charts I’ve presented.

There was a tension created by what we could achieve in a day and a half of intense meetings – and of getting caught in the weeds of details we can’t solve now.    I’ve been in roundtables and meetings before where “strategy” and “concept” were more important to get right – but I found those meetings unsatisfying because we could spend our time talking concepts and strategy and never get to do the second or third pass and come up with something that actually works.

It’s in my nature as an analyst and artist to want to “touch the data” and work directly with it – for me, it’s frustrating talking about ideas that are, as yet, un connected with the details and measures that might prove/disprove them.

Also, in giving a short version my presentation on The Future of  Monitoring Social Media that went over very well in London, on November 17th (below) ….

… I found a curious but interesting reaction when I described the typical analysis I do for clients and stakeholders in PR coming from someone who is a CMO currently – that it’s up to the analyst or social media person to question everyone, including the CMO, on what they want to achieve by having analysis done.

Personally, I find that attitude, offensive – it’s almost as if people are so busy, they can’t think and define, or figure out what they mean or what they want – and they want analysts – who are not often allowed to be decision makers, to figure it out, for them – meanwhile – even though that kind of back and forth is time consuming and difficult to accomplish on a regular basis.   I guess that’s why I like be grounded to the work I do – that I don’t want to get to the point where I’m talking about “boxes” and “concepts” that are “abstracted” from the actual things they measure – which to me, is somewhat, dehumanizing.

On the other hand, i found myself getting caught up in details where maybe, what was needed, was to agree on broad principles, first – so, I think there’s an interaction coming from different members of our working group, where this “tension” was actually constructive – even if, at times, it didn’t feel that way.

Overall, I’m glad we had our first meeting – though my head is still buzzing with the information and it’s going to take time and work for it to crystallize – but I hope I took the first step here and K.D. Paine will have a post up soon at The Measurement Standard site covering this Influencer Scorecard Summit gathering.