Underscoring the complete uselessness of social media rankings
Posted: May 19, 2010 Filed under: Work | Tags: branding, corporate social media, mashable, rankings, Social Media 20 Comments
Apparently, everyone loves lists and rankings. From where I sit at IBM looking out over the world of social media, I’ve come across a million of these rankings. What companies are the best at social media? What brands are the most active on Twitter? What company has the prettiest avatars on Twitter? And so on.
And every single one I’ve seen – even when IBM scores very high – seems superfluous and shallow. Even rankings from respectable analysts, like Charlene Li, are usually far off the mark (that’s for a future post if anyone is interested). It’s impossible to effectively look behind the scenes and evaluate how a company is performing in social media without understanding their unique approach and what they are trying to accomplish.
So, if you are going to rank anything, you better be damn well sure that your methodology is rock solid and you are absolutely thorough in your research. Otherwise, you end up looking far more foolish than when you began.
So what got in my craw today?
A company called NetProspex claims to have created the “first comprehensive look at social media activity among employees of the nation’s largest corporations.” (I originally stated this was Flowtown’s report, thus making the correction). Yet another ranking. It only caught my eye when colleagues on Twitter pointed out the Mashable post on it. So, after enduring far too many of these rankings, I’m finally speaking out: Rankings are worthless and do everyone (except the report publisher) a complete disservice. I’ll talk more about the disservice later. But first a bit on Flowtown’s “rankings.” (Click on the image above for an enlarged view).
See IBM there? Way down at the bottom. Tied for 48th. With GE. A company that doesn’t yet allow all employees access to Facebook in the first place (no disrespect meant for my GE friends, of which I have many, but the perspective is valid). Such venerable brands like NCR, Pactiv and BMC Software apparently have more employees active in social media than IBM. Really? Or even take Dell, who by all accounts (including my own) has done a fabulous and innovative job in managing social media. But it’s been a top-down effort, with controlled access and clearly defined accounts. But this is measuring the usage by employees. Dell still doesn’t have a policy for all employees to blog (Richard/Lionel, correct me if this has changed).
Again, I want to point out that NetProspex isn’t measuring effectiveness. Many of these on the brands on this list very well may be more effective that we at IBM are (though I think we have a compelling argument). But this list measured volume. So, on that point, consider the following:
- 200,000: current IBMers on LinkedIn, according to their own data… (By the way, it’s the LARGEST single community of corporate employees on any social network platform anywhere)
- >1,000: current IBMers actively blogging externally. (some on ibm.com, but most off domain).
- >3,000: current IBMers active on Twitter (my own estimate… but consider that Eric Andersen almost immediately quickly reached the 500 limit when he put together this list of IBMers. If no limit, it would be 6x larger).
- 75,000: current IBMers on Facebook.
- 100,000: current and former IBMers on IBM’s Alumni Social Network on ibm.com and LinkedIn.
I can’t see any possible way to justify the facts above with NetProspex’ rankings. Silly.
OK, enough with this report. I’ve given it enough press.
Finally, I want to get back to my point earlier about how these endless rankings are, ultimately, doing us all a massive disservice. Why do I think that? Well, primarily because it turns all of this into a race. And as companies try and keep up with each other, they start to apply the tactics that helped company “A” get to the top of the list without realizing they have a completely different business model and culture. It always ends awkwardly. I talked a lot about the impact of corporate culture on social media previously here on this blog. It’s worth revising that in the context of these lists to ask whether we should put a final nail in the coffin of endless rankings. If only.
********
I made some updates above to clarify that this report was apparently conducted/published by NetProspex, not Flowtown.
Adam Christensen 
Thank you. This is the response I wanted to show to my friends in other firms. May I add that it does not even make sense to have Apple up there. The methodology looks flawed since Apple doesn’t even encourage folks to be social.
– B
I reblogged that ranking chart the other day but I was shaking my head in disbelief and had many people comment to me about the wonky rankings. I agree with you, these things fly around like buzzing flies but show no real value or meaning without specific context or actual case-studies and data.
Mark here from @netprospex.
Appreciate the feedback.
The purpose of our report is to bring attention to the fact that employees and (more importantly), your customers, are “on” social media. That’s where the conversation is taking place. So if you want to engage with your customers, you had better wake up and engage with them where they are (on the social networks) rather than continue to shout at them via traditional media.
I agree that it would be really useful to take it further and measure the effectiveness of social media. That’s a really subjective thing so it would be helpful to hear everyone’s thoughts on how they would measure and rank that.
Mark, thanks for the comment. I can appreciate the desire you have to raise the visibility of social media.
My issue is that the research you are publishing is flawed and doesn’t make sense. From your methodology: “We used 100,000+ NetProspex contacts from the nation’s largest companies and analyzed their social media activity based on data provided by Rapleaf.”
So, your only source was people already in your database. Is your database a statistically representative sample of corporate America? That is akin to me publishing data on who the largest cell phone provider is based on the phones I find in my own house.
I’d be perfectly fine if you put your claims in context (that this is a survey of YOUR customers or whatever it is). But by stating this is the “first comprehensive look at social media activity among employees of the nation’s largest corporations” you put yourself up to a pretty high standard. And it fails.
Again, I’m don’t really care that IBM scores low – no one will really remember this tomorrow – but I want to make sure for future reference more rigor is applied to reports like these before publishing them with claims about how well brands are doing in this area.
I understand your desire to get publicity, but publishing deeply flawed research as a means to do it doesn’t seem the right approach.
Thanks Adam – terrific post …. which bring our social media approach back into perspective (and the right one…) @delphrb
The 100,000 contacts for the report are from the largest corporations in America (per the Fortune ranking) and were selected at random from our database of over 13 million business contacts. Our database is the most comprehensive, accurate and up to date source of business contact records available.
Furthermore, the data was normalized. That’s why we published an index number rather than a raw (non-normalized) value. Normalization eliminates bias.
Thanks Mark. That still doesn’t explain how the largest single employee population active on social media ranks so low. Something is wrong in there somewhere. I’d go back and check your algorithms and/or your database… It’s giving you some REALLY skewed numbers.
To be clear, for the ranking we normalized the data so that organizations with high employee counts did not benefit and organizations with lower employee counts were not penalized.
Adam, I was wondering what you’d find useful in terms of a ranking and/or report that would help accelerate social media use in businesses like IBM?
We’re working with other Fortune 500 companies to help them leverage social media and would be interested in speaking with you about the interesting things you’re doing to help IBM.
so what is your study measuring? “…takes a look at the social network usage of employees, and ranks the top 50 organizations according to overall activity.” Are you measuring percentage of employees engaged? So whether you are measuring total employee numbers or percentage of employees engaged, neither syncs with the results. Take GE for example. Are you saying they have the same percentage of employees active in social media at IBM?
Adam, I’d be happy to take you through the methodology in more detail — do you have time for a conversation about it?
[…] experts and experimenters, from Gadi Ben-Yehuda to Rita J. King to Adam Christensen (the latter got up in arms this week over a survey that pegged Microsoft as “most social”).Samurai. Because […]
[…] experts and experimenters, from Gadi Ben-Yehuda to Rita J. King to Adam Christensen (the latter got up in arms this week over a survey that pegged Microsoft as “most social”).Samurai. Because […]
Adam, great post. I’m also a fan of Luis Suarez, your colleague in Gran Canaria. And, before I forget, thanks for the tip on Lao Szechuan. Maybe I’ll drag the gang up there tonight!
On a semi-related note, I’m co-authoring a book about enterprise social media strategy for the nice folks at John Wiley & Sons. Do you have time for a brief telephone interview later this week? Or we can do the interview via email, whichever you prefer. Thanks in advance.
Cheers,
Mike Barlow
203-209-6061
Those are good observations, Adam – and a clever segue into BD discussion, Mark…which is what this posting’s all about, obviously.
[…] Underscoring the complete uselessness of social media rankings (adamchristensen.com) […]
This blog was… how do I say it? Relevant!! Finally I
havge foind something that helped me. Kudos!
Skype has opened its website-dependent consumer beta for the world, soon after
establishing it generally inside the Usa and You.K. previous this calendar month.
Skype for Web also now supports Chromebook and Linux for instant text messaging communication (no voice
and video nevertheless, individuals need a plug-in installation).
The expansion in the beta provides assist for
a longer set of spoken languages to help strengthen that overseas usability
Awesome website you have here but I was curious if you knew of any community forums that
cover the same topics discussed in this article?
I’d really love to be a part of group where I can get advice from other knowledgeable individuals
that share the same interest. If you have any recommendations,
please let me know. Kudos!
For newest information you have to pay a visit world-wide-web
and on world-wide-web I found this web page as a best
web page for most up-to-date updates.