Jonny Bentwood of Edelman recently unveiled an example of Edelman’s approach to measuring influence on Twitter, utilizing the Edelman-branded TweetLevel tool. The list of Top IT Analyst Tweeters, initially published on July 21, consists of the top tweeters among the IT analysts who are active on Twitter. These analysts include individuals from large and small firms, consultants, SMEs, and even some firms and brand names.
A Curt Monash post, So who is an analyst anyway?, contributes a fine discussion of analyst services for technology vendors and buyers in a professional world that includes newer venues like social media. Monash also explores the notion of what qualifies a tech professional as an analyst. One point from Monash has particular relevance to the Edelman TweetLevel influencer work:
You need flexibility in how you deal with influencers. Everybody is different
I conjecture that Edelman / Bentwood view this list of top tweeters as a test case / work in progress where they are fine-tuning calculations - and assumptions - while processing reactions, analyses and feedback. They deserve props for jumping into the fray by publishing a list that results from evaluating a class of tweeters using their “influence” meter (TweetLevel). With so many calls for metrics to evaluate the effectiveness of many aspects of businesses, here’s a starting point for trying to determine levels of influence. It may evolve and mature into a useful tool -- if tested, tweaked and nurtured thoroughly.
How do the calculations work?
The backbone of calculating the ratings of IT analyst tweeps is TweetLevel. TweetLevel can take daily influence temperatures that are tied to Twitter presence, based on 4 high-level attributes (as explained by Bentwood below):
1. Influence (What you say is interesting, relevant and many people listen)
2. Popularity (How many people follow you)
3. Engagement (You actively participate within your community)
4. Trust (People believe what you say)
Per Edelman, what constitutes a “top tweeter”, and potentially a top influencer, is described in the overview that explains the calculations on the TweetLevel site. The formula behind each TweetLevel rating for anyone on Twitter looks like this:
Jonny Bentwood comments on the calculations:
…as it happens the formula is all about being transparent. I can’t stand measurement tools that purely focus on popularity so have tried to be as open as possible about the inner works of how tweetlevel works. There is some real science and hard-core maths behind this if you want to dig deeper.
The full Bentwood post provides much more detail on the calculations, weightings of scores and methodology.
Anyone who tweets may visit TweetLevel to calculate current ratings based on the four attributes of Influence, Popularity, Engagement, and Trust. TweetLevel also provides comparison profiles to people included in an individual’s ecosystem of “friends”.
Personal Snapshots and Fluctuations
When first published July 21st, the list stood at 1000, but two days later Bentwood added at least another 150 individuals, and then re-ran the TweetLevel ratings, which of course reset positions for many individuals. This change suggests the possibility of continuous updates for this list, perhaps on a daily basis. Such a list is dependent on the daily fluctuating nature of TweetLevel ratings, in turn based on the daily fluctuations of each participant’s Twitter activity. If maintained in this fashion, the Top Analyst Tweeters list would be in continuous flux, to a certain extent.
I’m in the list and was surprised by my inclusion. I’m a one-woman show for my consultancy, and while many years of a software industry market intelligence practice qualifies me as an analyst, I never really thought a lot about whether or not I’m an “influencer”. For a small consultancy, the Edelman list may help with visibility – to connect to more colleagues and potential clients. But am I a “top influencer” due to my presence on Twitter, or even through my professional writing and interactions? – I’m asking this while looking at the many very talented heavy-weights in the Top Tweeters list. Thorough study of Michael Wu’s marvelous work on influence would be quite helpful here.
After Bentwood replaced the original list of 1000 with the second version, and as I began to understand more about the nature of TweetLevel-measured “influence”, I decided to track my “TL influence” temperature over several days:
Like a game of musical chairs, the individuals in this list could continuously jostle one another for somewhat different positions relative to one another. Barring some major career-changing event (positive or negative), most of the participants may continue to hover in predictable plus/minus ranges near where each one is currently on the list, if Bentwood does not continue adding new individuals.
Analyst Influence On A Daily Basis?
How often should these ratings be rerun and published? Is that a helpful activity or just muddying the waters?
The flux of the TweetLevel approach is problematical but also to be expected for a work-in-progress. With this list of top IT Analysts on Twitter, we are not tracking a relatively shorter-lived entity like a Justin Bieber, where day-to-day popularity seems to matter. Bieber is currently the #1 TweetLevel “influencer” of all Twitter users, a position he has held for a while; Bieber’s overall influencer score is 95. The person in the #1 position on Bentwood’s list of top analyst tweeters is Jeremiah Owyang with an influencer score of 76.6.
Instead of tracking pop culture figures, Bentwood’s list of Analyst Tweeters covers highly-qualified and respected technology professionals whose influence and credibility emanates from a lot of work that occurs outside Twitter. Edelman’s TweetLevel measures Twitter attributes that are also impacted by influence earned elsewhere, sometimes subtly, from other actions and interactions involving these analysts. Yes, Twitter presence and worthwhile tweets are part of the story - but many people may follow these analysts on Twitter because of who they are and what they write in blogs or accomplish in client projects, not because of what they do each day on Twitter.
Does a particular analyst really have less influence during a day or even week of less activity on Twitter, than during the many days and weeks of constant activity? Not really. Call it “residual kharma” – just because a well-known analyst that is a frequent Twitter contributor has taken a week or a month off, it doesn’t mean that all of the previous tweets suddenly have no influence. Nor does it mean that the anticipation of future tweets doesn’t carry influence across the break. Only if the analyst / influencer is absent from Twitter for a long periods of time, is influence likely to ebb a bit, until the analyst begins anew with tweets.
Vetting and Influence: Interested Parties
Who will be looking at a list like this? Most likely: buyers, vendors, other analysts, AR professionals, writers, consultants – pretty much anyone looking for guidance, assistance and forward-looking thinking regarding different segments of the technology industry.
® While the notion of influence may matter more to AR professionals and vendors, a lot of people will still consult a list like this to understand more about whose work they want to follow or whose experience they want to enlist. People and companies who are more attuned to Twitter may take this list more seriously, while others may think it trivial.
® Does this approach help vendors identify analysts with influence? In a very limited fashion, depending on the tweeting ability of each participant, there is a certain element of influence associated with analysts on Twitter. But to really assess “influence” and “reach”, there are lots of factors, including significant qualitative ones, that have to be evaluated as a whole. Context of expertise for each analyst is the starting point – goals of the vendor also matter. There are great analysts who are influential but don’t make a lot of noise on Twitter.
® From the buyer’s POV, are these fluctuating ratings helpful or distracting? A list like this will draw buyers to look at certain analysts or consultants, but as for vendors above, many factors come into play to help buyers to decide on tapping a particular analyst for assistance.
® AR Professionals are definitely interested in influence - take a look at a recent #archat on Twitter. I can see this list as particularly useful for introducing “new” players who are sharing information and opinion on Twitter. AR maestro Carter Lusher of Sage Circle has posted his take Technobabble 2.0’s Top Analyst Tweeters provides useful insights into visibility, but does not show who is influential where he concludes:
Bottom Line: Analyst influence is a complex condition that is the result of the interaction of business objectives, visibility, opinions expressed, client base, and so on. While a generic, mechanically-created list can provide useful input into a formal analyst list management process, such a list cannot measure actual influence.
Influence is still largely subjective and ephemeral. Reliability, credibility, expertise, integrity, top-notch work – all are probably more easily measured – but not necessarily as characteristics of Twitter presence. And reliability, credibility, expertise, integrity, and top-notch work are essential qualities of technology analysts and consultants who really make a difference for their clients and readers. To determine the ability of any analyst as a fit for a particular purpose requires looking at each individual’s body of work, knowledge and talent, original thinking, relationships to significant entities, and then evaluating how these attributes map to client needs and objectives.
Additional thoughts regarding influence, analysts who tweet, tech industry analyst practices:
§ Michael Wu’s series on Influencers
§ Twitter-based #archat on Influence, Analysts and AR professionals – transcript for the conversation
§ Vinnie Mirchandani - DealArchitect: More tenets for the next-gen analyst
§ Ray Wang: 7 Tenets of Building a Star Analyst Firm
§ Barbara French & Gideon Gartner: Advisory Industry Competition: Pushing Past ‘Business as Usual’ Part 1 Part 2 (also read the comments in each Part)
Related posts:
I have a confession to make: I am a Hybrid
About the author: Julie Hunt is an accomplished software industry analyst, providing strategic market and competitive insights. Her 20+ years as a software professional range from the very technical side to customer-centric work in solutions consulting, sales and marketing. Julie shares her takes on the software industry via her blog Highly Competitive and on Twitter: @juliebhunt For more information: Julie Hunt Consulting – Market & Competitive Intelligence Services
