Category: measurement

What’s the frequency, Twitter?

Tweet-birds-muralI guess by now we’re all aware that with Twitter, the more frequently we post, the greater the opportunity that our posts will be seen and acted upon.

But how much is too much?

This is something I’ve thought about since Guy Kawasaki brought it to my attention back in 2009. (See my post, To drive traffic, tweet and repeat, about Kawasaki’s test of tweet frequency.)

I’ve also worried about this idea of frequent and repeat tweeting. Like every other blogger with a PR or marketing background, I want to drive traffic to this blog. One way I do this is by sharing links to my blog posts on Twitter. I usually repeat the post a couple or three times, and generally I tweak the wording a bit in an effort to appear somewhat less lazy than a guy who automates tweets to churn them out at regular intervals. I don’t think I tweet too much. But I can’t help but wonder whether others see my approach to repeat tweets as obnoxious or spammy. I wonder whether followers ever notice those repeat tweets, and whether they think I’m posting too much about my own content.

When Kawasaki ran his experiment in 2009, he put eight hours of time between each repeat tweet. (He scheduled four identical tweets over 24 hours, then reported his results.) But a new experiment by Jade Furubayashi of SimplyMeasured makes Kawasaki’s test appear to be a model of restraint.

Furubayashi tweeted for one week in 15-minute increments, and another week in 30-minute increments. Her results showed that the more frequent the tweets, the greater the traffic from Twitter to her website.

So, frequent tweeting seems to drive web traffic.

But one item not addressed in Furubayashi’s experiment (and an issue raised in the comments to her post) has to do with the content of the tweet. Which also has to do with the headline or description used. (More about that in a previous post.)

What do you think? What’s the sweet spot for tweet frequency?

Image: Twittering Tweets Mural by cobalt123 on Flickr.

Another silly social media ranking

kredkloutLast spring, it was Education Dive’s list of the top university Twitter accounts, as determined by Klout score, number of followers and some kind of secret “subjective appraisal.” (I groused enough about that shoddy methodology last spring, so no need to rehash it here and now. The links are there if you want to revisit it.)

Now comes something via the Huffington Post called The Top 100 Best and Most Collaborative U.S. Colleges. And, just like Ed Dive’s approach with Twitter, this ranking’s methodology does not pass muster on many levels.

Once again, there’s a heavy reliance on an institution’s social media Klout score. As if that weren’t enough to raise skepticism, it relies on another ranking — one that is well-established but that comes under fire year in and year out: the U.S. News & World Report listing of the best colleges and universities.

  • HuffPo contributor Vala Afshar, who compiled the ranking, used this formula:
  • Pick the top 100 colleges and universities listed in the U.S. News & World Report ranking of national universities.
  • Look at their Klout and Kred scores.
  • Rerank the U.S. News & World Report top 100 by those social media influence scores.
  • Create an infographic that claims: “The very best schools are the most social schools.”

So. Where to begin the critique?

Let’s start with Klout and Kred. These tools are supposed to measure the influence of a social media user. And it may be true that individuals with high Klout and Kred scores may be more influential than those with lower scores, as this more or less balanced article suggests, I’m not too familiar with Kred, but I think Klout is designed for the individual, because it offers rewards from brands to those who attain certain scores and levels of activity. Chris Syme calls the Klout score an “ego metric” (see her comment on this post from a couple of years ago), and I agree. Aliza Sherman, in the article I cite earlier in this paragraph, says:

A high Klout score is like a Maserati or whatever the cool car of the day might be. It’s fun to flash around, but at the end of the day, it isn’t practical.

So I don’t think these vanity metric tools carry much clout or cred.

And what about the U.S. News & World Report rankings?

First of all, Afshar only looked at the top 100 national universities. U.S. News ranks many other types of schools — regional colleges, liberal arts schools, specialty schools, etc. So by limiting only to national universities — all Ph.D.-granting — Afshar excludes many from his list. For example, Williams College was ranked the top liberal arts institution by U.S. News, and it has a pretty impressive Klout score of 85. Meanwhile, one of U.S. News‘ top national universities, the Colorado School of Mines, was ranked No. 100 on Afshar’s “most social” list with a Klout score of 57.

So if you’re going to use Klout as a metric, don’t penalize ostensibly social media-savvy schools just because they’re not in the “national” category.

Finally, Afshar’s ranking suggests that an institution’s high rank in social media equates to high levels of collaboration. I don’t buy it — for the same reasons I didn’t buy Ed Dive’s list of the best higher ed Twitter accounts last spring. Look at several of the schools on either list, and you’ll see that much of the communication is one-way. That doesn’t sound very collaborative to me.

One commenter on the HuffPo site wrote, “[I]f you calculate the correlation between U.S. News ranking and your social rank, you get a coefficient of around .32, which indicates a weak to middling relationship. In other words, it is difficult to make the claim that ‘the very best schools are also the most collaborative.'”

Difficult, yes. But as we tend to see more often these days, not impossible.