Blog

Feb 09

Posted by Ken Yeung

Posted in Marketing, Social Media

Just How Influential Are You? Secrets Revealed About Twitter Influence Tools

For a lot of people, the one thing we know what Twitter is used for is conversation. For businesses to possibly get at least a hint of usefulness, we’d need to look at metrics. Now with social media, the metrics we look at and analyze are different from other online metrics – which is to be expected. In this area, we’re going to be looking for influence, because that’s one of, if not THE, main measurements of assessing the return on investment. Sure, with Twitter you could track all your links using Google Analytic tracking, but keep in mind that not all tweets will have a link. What’s more, you’re not interested in links. You’re going to want to be interested in evangelists and people who you believe are really passionate about your product. Once you’ve found them, you’re going to need to gauge their influence over others. So how exactly can you do this? Well currently it doesn’t seem possible with Twitter directly, but there are several tools that can help shed some light on this area for you.

In this blog post, I’m going to look at three well-known Twitter influence analyzing tools: Klout, Twitalyzer and Twitter Grader. Each one will be dissected to look at their offerings and whether it’s really interesting enough to look into further. I’ll also be exploring the applicability of each as well.

WHAT ARE THESE TOOLS?


Klout is a social media influence tool that not only focuses on you personally, but looks at brands and industries to help assess influence. In a way, Klout will show you your social graph and assess just how influential you are along with who is influencing you.

Twitalyzer is dubbed the “serious analytics for the social media expert” which may lead some to believe that it’s going to provide you with much more relevant analytics to make you have a superior web 2.0 strategy. Twitalyzer will offer you more visualizations, easy-to-read statistics, integration with Google Analytics and is free.

Twitter Grader is a tool straight from the guys at HubSpot. Perhaps one of the early influence metrics out there, Twitter Grader assesses your influence by location and gives you a percentage as a score. It too is also a free tool and will display word clouds in your past tweets along with other pretty interesting metrics.

SO WHAT’S THE BIG DEAL WITH THESE TOOLS?

After using each of these applications multiple times, I’d like to think that there are some differences amongst all three – some major and some minor – that would give pause to any company interested in using either one.

In the first place, if you’re talking simplicity of style and information, then I would go with Twitter Grader over any of the other two. Why? It’s because you’re just asked to provide a username and it spits out a score with no explanation or description of what it means. From my understanding and observation, it seems that quite a few people have received percentage scores of over 90%, but then what really separates me from, say, Chris Pirillo, Brian Solis, Jeremy Toeman, etc.? It seems more irrelevant, right?

The one redeeming feature with Twitter Grader is that it gives you a ranking as well – so my ranking is 12,037 out of 6.1 million people. Now, that’s only 6.1 million people who have entered their username into the search query, NOT all the Twitter users in the world. To have that number would be truly impressive. But going back to simplicity, Twitter Grader offers us the option of looking at a more geographic setting so we can see who the top “influencers” are by city, state or perhaps even country.

If you want more metrics and statistical analysis, then perhaps you might be interested in Twitalyzer. Out of the three tools described in this post, I’ve seen more metrics provided on Twitalyzer than any other. You’re going to be able to measure the impact of a person’s influence, their ranking in terms of engagement, influence, generosity, and clout. I think that this specific tool would be perfect for large companies who want to assess their influence and report back to their supervisors, then Twitalyzer would be a good metric to have. If you just want to know the basic information, then you might want to stay away from this because it’s too much for what you’d like to do. I do like how Twitalyzer has links to definitions so you’ll be able to easily understand what each metric means instead of wondering whether that 5.0 is a good or bad number.

But what if you want more topical or categorical influential data? Then that would belong to Klout. While all three of these tools clearly understand how to assess influence in the real-time stream, one thing that separates Klout from the other two is the fact that they understand how to segment all the influencers out in the stream into different buckets of information. So why is this important?

Imagine that you’re a business that is interested in seeking out influencers to review your product or perhaps help you promote it. If you’re a consumer electronics giant like Sony, Boxee or Panasonic, you’re going to want a way to filter out all the noise to get to the information. Through Klout’s algorithm, they’ve already done the work for you. You’re going to be able to sort through a list of influencers with respect to consumer electronics, technology,etc. A preview of one such list is to the left showing you the top influencers that talk about technology according to Klout, along with their Klout score – almost like it’s Digg for Twitter, only there’s no voting to increase the score.

ALL THE NUMBERS ARE DIFFERENT – HOW CAN WE TELL WHAT’S IMPORTANT?

One thing that you’ll encounter is that the numbers offered on each of these tools are completely different from one another. Their algorithms can vary across the board so you will never have the same results. In order to properly evaluate someone’s influence, you should look at not just the number, but in fact who they are, the types of content they are producing (tweets, blog posts, video podcasts, etc.) as that will be a good gauge of what you might need. Each of these numbers should be taken with a grain of salt – measuring influence isn’t a true science and is often subjective.

To show you the difference between the three, below is a chart with the information on several popular individuals on Twitter:

  • Jeremy Toeman is an avid user of Twitter and is the founder of Stage Two – understands and tweets about gadgets & consumer electronics.
  • Adam Burg is a semi-frequent user of Twitter.
  • Chris Pirillo is a well-known Twitter user and tech “geek”.
  • Rafe Needleman is a member of the media who reports on technology trends & uses Twitter pretty often.
  • Deb Schultz is an analyst and uses Twitter somewhat frequently.
  • Veronica Belmont is an Internet celebrity, host of Tekzilla on Revision3 and has a huge Twitter following.
  • Ken Yeung is a prolific Twitterer and social media enthusiast.
Twitter
Grader
Twitalyzer
Klout
Jeremy Toeman 99.2% 2.2% 30
Adam Burg 89.0% 0.1% 18
Chris Pirillo 100.0% 86.3% 89
Rafe Needleman 100.0% 2.2% 48
Deb Schultz 99.8% 1.4% 41
Veronica Belmont 100.0% 65.1% 64
Ken Yeung 99.8% 5.0% 56

As you can see from the table above, there is not a single uniform measurement out of three social media metric tools available on the Internet today. So what do these influence number really mean? Like I said, it’s subjective. IF you choose to use these for a benchmark project, I would recommend being uniform in your execution. Do not choose to use Klout and then use Twitter Grader as that will skew your results and leave you with questionable reporting. It’s similar to web analytics – a unique visitor could be calculated differently from one analytic tool to another, leaving you with inconsistencies that may not paint a complete picture.  In the end, there aren’t any easy solutions on assessing one’s influence. You’ll need to really dig deeper to find out how motivating that person can be in persuading their friends & family to do something.

Tags: , , , , ,

4 Responses to “Just How Influential Are You? Secrets Revealed About Twitter Influence Tools”

  1. Jeff Katz 10 February 2010 at 9:49 am #

    Ken,

    Thank you for reviewing Twitalyzer as part of this post. I want to expand on a couple of points that you made.

    Since our team comes from the web analytics space, we appreciate your reference to how we all collect, define and calculate Influence in a similar manner to the WA vendors do with measures like Unique Visitors and Visits. That said, I should note that with our 2.0 release last month, we made some changes to our metrics. Impact has supplanted what was previously calculated as Influence. You can see here what are metrics and measures mean: http://www.twitalyzer.com/help.asp#definitions

    Regarding “more topical or categorical influential data” Twitalyzer has that and then some. Start with the Benchmark report as an example. You can filter by location and/or Tag. Tags are attributes that are either auto-discovered by our system (via hashtags) or manually added by users. For example, I may want to see who are the most influential people on Twitter using #measure, I will see this http://www.twitalyzer.com/list.asp?u=twitalyzer&f=&l=&t=measure I can drill down to just people in Portland – http://www.twitalyzer.com/list.asp?u=twitalyzer&f=&l=Portland,+OR&t=measure

    Of course that is just one of many ways organizations and individuals can sort and filter through our data.

    – Jeff

  2. Israel Kloss 11 February 2010 at 11:14 am #

    You didn’t include TunkRank? We measure influence by how much attention
    your followers actually give you at http://tunkrank.com. That’s different than the algorithms that any of these services use. Would you be interested in beta testing it?

  3. Ken Yeung 16 February 2010 at 3:16 pm #

    Hi Israel, I didn’t include TunkRank because I wanted to compare just some of these tools. That’s why there are three companies working in Twitter influence tools that provide a similar benefit to their customers. There are several other services that have also been left out of this as well. I would be interested in checking out TunkRank to see how different the service is.


Leave a Reply