Tom H. C. Anderson - Next Gen Market Research™

Interview on Influence with Social Media Citizens

I was interviewed by Giedrius Ivanauskas on behalf of both Social Media Citizens and Social Marketing Forum earlier today. I found his questions rather interesting, as they’re not the type of things I typically get asked by market researchers. It was sort of interesting thinking about the questions both as a researcher interested in social media influence and as an ‘influencer’.

I’ve posted the first part of the interview below. You can read the full interviews at Social Media Citizens and on the Social Marketing Forum.

…

G: How would you define a social media “influencer”?

T: Anyone who is active on a social media channel of any sort is an influencer. Conversely, they can also be influenced to some degree.

Obviously some have a much wider reach and are also more well respected (likely to be listened to). Bloggers have the added importance of having whatever they say be ‘evergreen’, well perhaps not always pertinent, but whatever they’ve said can come up in a search engine result anytime. So to me, and more and more savvy marketers, influence is defined on a sliding scale.

This is what services like Klout.com are attempting to measure in a somewhat simplistic way.

G: How do you measure the influence?

T: Depends on the specific media source of course. If it’s a blog, actual traffic as measured on sites like compete.com, Alexa.com (neither usually too accurate) etc. are ok. However, both seem to show much lower traffic than Google Analytics and other first party trackers. Additionally what you have to keep in mind when you look at a site like Compete.com is that depending on the provider and host of the blog it may not be giving you just the statistics for actual readers of just that single blog, but the stats for everything on that specific domain.

I’ve seen some of the blogs in my industry get an unfair boost because of this. If the blog is hosted as part of the same domain as the rest of the information and service offering of a large company, trade directory etc. then often this other traffic not pertaining to the blogs popularity can get counted. Sometimes even their survey traffic benefits their traffic and SEO (relevant for issue for me as I’m in the market research industry).

My blog is totally separate, for better or worse, so I don’t benefit for these errors in measurements. Though I sometimes wonder if I should have run my blog off of www.andersonanalytics.comdomain rather than www.tomhcanderson.com to take advantage of some of these scale benefits. However I realized that having them separate allowed me a lot more freedom in terms of content etc.

As for Twitter, well clearly Followers are a pretty good indicator of influence. Arguably, even better than followers on twitter is # of times listed. A combination of these two metrics is better of course. After that, re-tweets etc. However, I find that these secondary level metrics such as “re-tweets” contains a lot of errors. Some of the accounts with the most followers, such as my account on Twitter is often not captured in the public timeline.

This can happen for various reasons. Suffice it to say, that the most influential on Twitter are sometimes even more influential than you think, because a lot of the ‘re-tweets’ etc. do not appear in the public timeline and Twitter search tools and therefore is not counted for these most influential users by services like Klout.

Engagement, is certainly important. PostRank is trying to measure this across a few channels. They and many other services connect to several social networks including Twitter, Facebook and even blogs. I know they are working on adding LinkedIn currently.

The most influential of us though a are clearly not active only on a single network, but across multiple networks. Just like a luxury product goes well with another luxury product, so do different social media channels

This way they take advantage of both Reach AND Frequency.

G: Who/What is influencing you on Social Media Properties?

T: It’s not really so much who as what. I’m influenced by what I find interesting. Wired and Mashable are certainly interesting sources for information.

When I first got into social media, specifically when I was trying to understand Twitter, Guy Kawasaki was an inspiration to me as well. He gave a lot of helpful tips on his blog. However, as with all of us, we share certain tips and keep others secret

G: What are the key mistakes when targeting influencers and how to avoid them?

T: Depends on what you’re targeting them with. But generally I would say you need to appeal to them. If you want them to do something, you need to give them an idea of what’s in it for them, without being offensively obvious about it. The fact is that you need their help more than they need yours, so you have to be political about it.

Secondly, a canned message rarely works. If you truly want to get them on your side, you better let them know that you actually are a fan and know what they are about.

G: Do you think targeting influencers is overrated or is it an important part of Social Media Strategy?

T: Of course it’s important and it will become even more important as social networks continue to connect us all.

Putting aside the super influencers for a minute, we also need a strategy for everyone else. The Long tail of influencers if you will. We’re not quite there yet, but I think in the same way that it makes sense to target super influencers with information that lets them know you are a fan or similar to them, text mining and data mining will allow us to customize this type of messaging for masses in the future with less time invested.

Not exactly sure what it will look like. It may be a new version of Phishing and Spamming or “Spear Phishing” as Jeff Jonas, Chief Scientist at IBM calls it. He views this as something that will first be used by criminals, but I’m guessing marketers will be there just as fast.

G: What are the trends that you are following at the moment and feel most passionate about?

T: I follow Text Analytics and Data Mining in general, also in terms of social network analysis Link Analysis, CRM Analysis are also of great interest. Web tracking is another area. Finally, widgets and other applications …

Feel free to read the rest of the interview Q&A here.

@TomHCAnderson