My last blog post summarized why I deleted my Twitter account and started fresh. To follow-up, here’s why Twitter ranking tools are meaningless–with a comparative focus on HubSpot’s Twitter Grader application.
According to HubSpot’s blog, the grade is algorithmically determined by six factors:
1. Number of Followers: More followers leads to a higher Twitter Grade (all other things being equal). Yes, I agree that it’s easy to game this number, but we are looking at measuring reach and I did say all other things being equal.
2. Power of Followers: If you have people with a high Twitter Grade following you, it counts more than those with a low Twitter Grade following you. It’s a bit recursive, and we don’t get carried away with it, but it helps.
3. Updates: More updates generally leads to a higher grade — within reason. This does not mean you should be tweeting like a manic squirrel cranked up on caffeine and sugar. It won’t help either your Twitter Grade or your overall happiness in life.
4. Update Recency: Users that are more current (i.e. time elapsed since last tweet is low) generally get higher grades.
5. Follower/Following Ratio: The higher the ratio, the better. However, the weight of this particular factor decreases as the user accrues points for other factors (so, once a user gets to a high level of followers or a high level of engagement, the Follower/Following ratio counts less).
6. Engagement: The more a given user’s tweets are being retweeted, the more times the user is being referenced or cited, the higher the twitter grade. Further, the value of the engagement is higher based on who is being engaged. If a user with a very high Twitter Grade retweets, it counts more than if a spammy account with a very low grade retweets.
My recreated @ariherzog account is ~40 hours old. After following who I used to follow, I removed those names from the older account to bring the number to 4 (which you can see below). Still with me?
With the logic understood, take a look at the following two screenshots of my new and old accounts:
Twitter grade of @ariherzog - Click to zoom
Before continuing, note the rank on the left and the overall grade on the right. The presumption is higher ranks and grades are good indicators you would follow me as a result. Personally, I don’t look up tweeps’ grades to determine if I follow; I look at tweet content and follower/following ratios; but that’s me.
Twitter grade of @ariherzog_old - Click to zoom
My new account has also considerably dropped among Newburyport graded users from the top three to 25th place. If you care about that sort of thing.
I’m more productive, if about 100 tweets are any indication. I’m replying to people more, retweeting less, and watching the flow of tweets less–because of other activities keeping my eyeballs away from the Twitter stream. I’m clicking on external links more, leading to more RSS blog subscriptions appearing in my feed reader. I’m more focused. And, I’m manually blocking spam-like accounts, e.g. those with usernames like @pam25heart or @twittermachine24.
I’ve always agreed with Danny Brown and Mark Drapeau that ranking tools are meaningless. I hope the above comparison helps explain why it shouldn’t matter how one is ranked as much as what one tweets. Someone may be ranked at #1 with a grade of 100 (which I’ve had in the past) but if the content is irrelevant to you or annoying to you, what’s the point of the rank?
Related articles you may enjoy:
{ 10 comments }
Twitter: DannyBrown
October 2, 2009 at 6:43 PM
Bang on, Ari. They’re a fun way to show friends if you’re into that kind of thing, but mostly they’re irrelevant for the reasons you’ve mentioned here, as well as they don’t really take into effect context of the content. I could blast out 10,000 messages of meaningless drivel (some may think I do already…!) and that would put me higher than someone who’s sharing tweet after tweet of solid content.
Until that can be looked at, and the algorithms actually make sense, they’re good toys to play with. But you wouldn’t use a toy to win a client or impress a potential employer/employee, would you?
.-= New from Danny Brown: Something New, Something Old =-.
Your last question is enigmatic, you realize, Danny, right?
Disclaimer: I’m the developer of Twitter Grader.
You are indeed right — the results in this case don’t make sense.
What you’ve stumbled on though is a limitation of Twitter Grader particularly as it relates to how it credits older accounts for tweet and retweet activity. It’s something I’ve been wanting to fix for a while, but have not come up with a decent solution yet.
You’ve got a good test case here. One quick question — did you rename your account and then create a new one, or did you start the new one “fresh”.
Thanks for trying out Twitter Grader — and apologies that the results are so strange.
Correct assumption. I changed @ariherzog to @ariherzog_old, then created @ariherzog.
Thanks for stopping by and your thoughts!
Another thought… why is my new rank only 539,000? Shouldn’t it be closer to 3 million or such? Or is it that high because I’m following (or followed by) people with high ranks themselves? Further proof of the silliness IMO.
A couple of points:
1 The Twitter Grade is a relative score, so it’s not that surprising that you’d have a decent grade/rank because there are so many users that have very few followers and very little engagement.
2. Having looked at things a little deeper, I’m surprised that you’re that surprised with the new vs. old grades. Your old account has a bunch of followers and had decent engagement at the time. Why would your new account deserve a higher grade?
Twitter Grader is designed to measure your authority and reach on twitter. In that regard, I’m not sure that the software is behaving all that poorly in the case of your new vs. old account. Based on the data available do far, your old account does deserve a higher grade (at least for now).
Umm, err, uhh, I meant lower rank. Higher number. My presumption is new Twitter accounts start at the lowest levels, aka grades of F or N/A or such. The algorithms determine higher grades and higher ranks, right? So, why would a 48-hour account have a rank of some 500K when there are 4M users in the system? Shouldn’t a new account be much farther down?
Twitter Grader’s algo is just silly, it changes too frequently to give any credence to one’s ranking. Some days I have 99.99 others I’m down to 99.93 but if I look at my rank in GB as location I can be in the top 40 one day and nowhere to be seen the next day, then in and out again. I actually once refreshed the page and saw that I’d gone from 11000th ish in the world to 993rd…next time I was 3407th…pointless.
.-= New from David Bradley: Stroke, NIR, Guests =-.
Is any online algorithm tool not meaningless?
Well, they’re all meaningless to everyone else. The only meaning they ever have is if one’s own rank is high and you want to persuade others of how worthy that status is
Comments on this entry are closed.