From the Lab
Find your ideal customer evangelists
Today one of our twitter peers asked us how she could find people of influence for a given region. As I researched the question and solution I found that there is actually a name for what is probably the end-game for this exercise:
Influencer marketing- is a form of marketing that has emerged from a variety of recent practices and studies, in which focus is placed on specific key individuals (or types of individual) rather than the target market as a whole. It identifies the individuals that have influence over potential buyers, and orients marketing activities around these influencers.
As I will demonstrate it is possible, using some freely available analytics tools, to narrow the list of influencers. However, as our own Social Network sage Matt Thomson pointed out, the analytics tools fall short in becoming truly useful and further click-based research is really needed to establish credible, influencers who may become evangelists.
Given the current landscape I will choose Twitter and blogs as my universe from which to collect people. In this exercise let’s say that I wanted to find people in Colorado who have a large sphere of influence in the digital world. So let’s begin to find some influencers.
I want to start off with a list based on Twitter users for two main reasons:
- - Hyper-active Facebook users and bloggers also seem to have accounts on twitter
- - Twitter ranking tools are more predominant making it easier to build my initial people-of-interest list.
Influencer Search
There are a number of twitter ranking or grading sites. TweetValue provides a monentary evaluation of a twitter profile. TwitterGrader provides a “grade” for twitter users and TwitTown provides a list of the top 150 users. TwinFluence analyzes a twitter profile to derive reach and trending measurements. All of these “analytics” tools have one problem, they are not very searchable. Honestly though they server their purpose which is to rank users based on whatever formula each site thinks is best. Fine. But how can we use this?
TwitterHolic will provide the top 1000 twitter users and unlike some of the other sites it displays the location. So, using the old “Next Button” I iterate through the pages grabbing all the users in Colorado:
42.
Darth Vader (darthvader) Empire, CO Followers: 19,484 Friends: 4,015 Updates: 250
201.
zaibatsu (zaibatsu) Denver http://thedrilldown.mixx.com/ 6,494 6,386 1,732
227.
Joshua Allen (fireland) Denver http://www.fireland.com/ 6,075 61 372
255.
Lucretia M. Pruitt (GeekMommy) Denver, CO http://www.geekmommy.net 5,791 5,083 26,702
397.
Jeff Walker (JeffWalker) Colorado http://www.productlaunchformu... 4,371 44 786
I abbreviated the list a bit but you get the gist. From this list I will use my logical deduction skills and determine that Darth Vader is probably not a good person to contact. Although he is very good at gaining friends and influencing people I think his desire for evil may create a rough relationship.
Zaibatsu on the other hand is a good suspect, although, per his website, he was banned from Digg. Good for him, great for publicity.
Joshua Allen is outright hilarious and you should read his blog/tweets. That said he is probably a good person to talk to. If nothing else we would get a laugh.
Number 255 is fantastic. Her blog already has emblems for an “Eleven Moms” community so she is definitely in the know and aware of her circle of influence. Along the same lines Jeff Walker looks like a good candidate.
If the list were longer or if I wanted additional information I would use TwinFluencer to trim the list or provide additional ranking metrics. Since I only have 4 users I don't really need to perform any additional trimming.
No matches? Well if you have no users showing up out of the 1000 then the next step is to use a tool like Monitter and search for keywords in tweets with X miles or kilometers of a city or cities in your area of concern. This should give you a good idea of who is active. If you monitor the results for a time you will should be able to get a good starting list, maybe even better than by starting with high "ranking" twitter users.
Influential Blogs
Now I have some active people with large followers now lets see what their digital properties are like. There is a great blog here which discusses various ways to measure a blog’s influence.
Using Alexa we can get a feel for the amount of traffic a blog is receiving and where the traffic is originating. Here is the Alexa Chart for number 201’s blog, thedrilldown.mixx.com:
[caption id="attachment_583" align="alignnone" width="300" caption="Blog Hits"]
[/caption]
Sadly, in order for Alexa to display the metrics the site must be in the top 100,000. Usually this is a pretty high-order for individual blogs. What do we do then? Well, one rough and quick measurement is to use Google and find out how many sites link to each person’s blog/site. The syntax is: “link:<url to check linking to> for example: link:http://www.fireland.com/
Site Links General Search hits
www.fireland.com 250 21,300
http://www.geekmommy.net 369 4,170
http://thedrilldown.mixx.com/ 204 3,950
While at google we might as well just search on each person to see what happens, I put the search results count for these general searches in the last column.
Conclusion
From these results I would conclude that GeekMommy is the best candidate to tap on to become an evangelist or catalyst for some marketing or pr campaign. She has many more updates, friends and linkback than the second place finisher, Fireland.
Well that was the non-automated, quick and dirty way to try to find influential social network users in a given regions. Yes the process leaves a lot to be desired and is not very robust but it should give you an idea of how much research is still required to gleen usable, actionable information out of the social networks. If you are intersted in hearing about some of our social networking analytics tools and ideas let us know and give us some information on what you would like out of such tools. We only build things that people ask for.


![Reblog this post [with Zemanta]](http://img.zemanta.com/reblog_e.png?x-id=23be4af1-2e1c-4a16-b93a-183923711fd1)