Hendrickson says, “Twitter is not a numbers game”
This week I was joined by over 1250 publishing colleagues in the Big Apple at the O’Reilly Tools of Change Conference. One session in particular that stood out for me was a workshop called “Twitter Scorecards for Publishers”, hosted by Mike Hendrickson of O’Reilly.
Most publishers in the workshop were still struggling to define where Twitter fits in their business. 1/3rd of the room identified Twitter as a “new tool” for their company.
In usual workshop fashion, we audience members were put to work; assembled into small groups and amongst many other questions, asked to answer this: what do we want to measure in Twitter.
Suggestions offered were: Followers, Friends, Total Retweets, Follower/Follow Ratio, Followed Back %, @ Mention Count, List Count, List Followers Count, Unique Retweeters, Unique Messages Retweeted, Follower Retweet %, Unique @ Senders, Follower Mention %, Inbound Messages Per Outbound Message and Update Count.
Depending on what your goal is with Twitter, you can pick and choose which numbers matter to you.
To build your own Twitter scorecard, ask yourself these questions:
What is my reach?
Are my tweets interesting and informative enough to build an audience? Do people spread my content through re-tweets? Are people adding me to lists and are those lists being followed?
What is my demand?
How many people did I have to follow to build my count of followers, and are my followers usually reciprocated? Hendrickson says that “Twitter is made to follow people that you find interesting, don’t do it for the game or the numbers”.
How am I at engagement?
How diverse is the group that @’s me? Am I broadcasting or participating in a conversation? Hendrickson joked that “if you only tweet about ice cream, you’ll probably get all 35 ice cream people following you. If you want a broader audience, broaden your tweets”.
What’s my velocity?
How likely am I to be re-tweeted? Do a lot of people re-tweet me or are they the same few followers? How can I re-word my tweets so that people find them interesting enough to spread?
What’s my activity?
What’s too much, what’s too little? Are my tweets effective in generating new followers, retweets and @ replies? Hendrickson notes, “if someone is only following you because you’re following them back, that person isn’t worth following”. He also talked about providing relevant content, rather than just brief commentary.
Thoughts from other publishers on measuring Twitter
To my surprise, in our workshop group, publishers agreed with my own thoughts: that it’s not necessarily about the amount of followers you have, as it is about the results you are getting. I can also argue with myself that you should have a decent amount of followers in order to attract new followers, as well. The point however, is this: if you want a better “score”, you should have less useless followers, and more engaged followers.
This follows the same rule we tell publishers to do with their email lists. If you have people on your lists that haven’t even opened an email in over a month, take them off your lists. Some publishers we know will even take people off their lists if they haven’t bought anything in six to twelve months. Why? It’s for the numbers. You might have a higher conversion rate when there are less people on your list.
It’s up to you to decide whether you prefer to tell people you have a list with 200k people on it with a 2% open rate, or a 86k list with a 60% open rate. Parallel those percentages with conversation rates and you have a few more bragging points.
Using Klout.com to rate your Twitter profile
Back to Twitter though. Hendrickson showed the audience a site called Klout.com, which basically rates your Twitter profile. It takes into account how many people you follow, how many people follow you, whether you get re-tweeted or responded to, whether you’re re-tweeting or responding, and factors like that.
When analyzing Twitter accounts from a compilation of all attendees at the conference, Hendrickson found that trade publications had an average score of 49, where over 30 is considered to be doing very well. Corporate publishers were around 37 (Microsoft, Google, etc). Hendrickson noted that corporate could do better if they talked more to their audience, asking questions and providing more customer service.
Network Score looks at the Klout score of each person who interacts with you to determine:
- The inﬂuence of people who @ message you
- The inﬂuence of the people who retweet you
- The inﬂuence of the people who follow you
- The inﬂuence of the people who list you
- The inﬂuence of the people who follow the lists you are on
Klout identifies Twitter accounts as one in four different types of “tweeters”: casual, persona, climber, or connector. Check out Klout.com and run your own profile to get more in-depth descriptions of what each of those mean in klout-speak.
Hendrickson noted, “the one thing to make sure you don’t do when you’re a climber is get spammy. Don’t broadcast, broadcast, broadcast, or you’ll be considered a spammer. Don’t overtweet your followers and climb too quickly”.
Hendrickson also noted that personal accounts have a more responsive and effective following than corporate accounts because with personal accounts you have the ability to “know” the person tweeting, whereas most corporate accounts don’t have that personal connection.
He told the audience that Tim O’Reilly’s Twitter account (@TimOReilly) has a much higher Twitter score than O’Reilly (@OReillyMedia). This isn’t to say you shouldn’t have a corporate account, but you should certainly have both.
So, are you following us on Twitter yet? We’re there at @Mequoda and we tweeted the crap out of TOC, so check out our tweets, and also check out the #toccon Twitter feed from the conference to catch up on all the talk that went on.
Also, what do you think about this Twitter grading business? What do you measure?
For more information on Twitter, join us for our Twitter for Publishers 2010 webinar on March 16th.