Tagged: twitter

Social media mystery: the case of the 5000 follower intern

So we were hiring an intern.  Hiring might be a strong word since we’re not planning to pay this intern, but I digress.  So I’m reviewing resumes and sometimes if somebody looks promising, I check out his or her Twitter account.  It is at least in part a social media internship, after all.  So one candidate comes to the top of the pile and she has over 5,000 followers on Twitter, no small feat.  Or so I thought.

5,000 followers is neither here nor there by itself, but this person was following only 30 people and had under 100 tweets in just a few months on Twitter.  Reading the tweets, I was not super impressed: almost all were links to various marketing and social media articles, no added commentary, no @ replies, no real value add from information or entertainment.  How do you get 5,000 followers with 100 mediocre tweets in a few months?  I was pretty sure there was only one possible answer: cheating.

A phone screen confirmed, the intern candidate had paid some shadowy service to deliver followers.  It was, she said, an experiment to see if having more followers would make it any easier to get noticed or get her message out on Twitter. Purists – and even semi-purists – can be horrified all they like, but here’s the thing: it works.  5,000 followers and 100 tweets over a few months gives you a higher Twitter Grader score than mine with about 1,000 followers and 5,000 tweets over more than four years. Maybe that says more about Tweet Grader than it does about buying followers.

So if you’re still in the habit of judging people by the quantity of their following, or using follower quantity as the raw materials of social media scientific inquiry, or to create dubious metrics like TweetGrades, beware, not everything is as it seems.

And the intern?  No decision yet, but I give her points for scientific method.  If she’d done a double-blind study, I’d be willing to double her salary.

The state of the twitter economy

I’m not sure what’s more narcissistic: binging oneself, checking your follower status, or reading your own blog’s back catalog.  That’s a topic for another post, but while committing a minor sin of onnetism I discovered a post from last November that bears revisiting some six seven months later.

After attending a meeting of Boston Media Makers, I set out to estimate the market value of my twitter stream.  You can read the gory details here, but the upshot is that the Magpie service seemed to value my twitterish at about $15 CPM. I wonder if that figure has gone up or down, and why.

Here’s the formula: I used magpie to get an estimate of what they’d pay me, then using followcost and some guesstimation, I figured out what my audience was, and derived the CPM.  Back in November, Magpie offered me 69 Euros a month, and I was tweeting about 5 times per day to 252 followers.  Interestingly, today Magpie quoted me only EU 23.49, but I now have 632 followers and tweet about 3 times a day.  These figures suggest a CPM of about $3.75, quite a drop.  What’s changed?

Magpie's estimate Tweet frequency via Followcost

Well, I am tweeting less – to the relief of many – and that might make me less attractive to advertisers.  But I have more than double the followers (so my total theoretical impressions are up), and my twitter grade is up and my percentile rank is up, too.  (In November I was #10,546 out of 255,406 for the 4th percentile, and now I’m #44,613 of 2,276,191 which is the 2nd percentile)  So why is my Magipe CPM a quarter of what it was half a year ago?

Twitter grader stats

Well, gentle readers, as  you may have noticed, I didn’t really buy the ad valuation last time out (my estimate was a lot closer to diddly) so the fact that it’s gone down should please me.  But here’s the thing – it’s still too high by a huge factor.  Back when I had 250 twitter followers, I could tweet a link and around 20 people would click on it.  Pretty sweet.  Today, with over 600 followers, I can tweet a link and about 20 people click on it.  Based on grader’s estimates, the twitterverse is about 10x larger in terms of number of users now, but the results that I get – and by extension, what I figure an advertiser would get – in terms of clicks is pretty much the same.

I suspect that this bottoming out of the Twitter ad economy (which, by the way comes from a whopping sample size of one) is partly a coming around to reality and deflation of hype, and partly a change in the way people use Twitter. Follower and following numbers are up, and use of applications such at Tweetdeck to manage these larger streams is also way up.  These applications let users group and manage their Twitter friends, and thereby reduce the number of tweets that are actualy read.  This, and the fact that the applications remove from view the actual Twitter UI, suggests to me that the prospects of anybody making money with Twitter advertising – including Twitter – are dwindling.

DUCKIE SAY RELAX

I’ve done it over 2,000 times., and I don’t think that’s immoderate for a man of my age.  I’ve done it on a boat but I’ve never done it with a goat. Sure, when you’ve done it as many times as I have, it may seem almost routine, but I’ve noticed that some people are having some anxiety about their First Time, so here’s my advice for  your first twitter experience:

Relax. Just do it.

Seriously.  If you’re sitting on the twitter sidelines trying to figure out what it’s about or how your company can use it, you’re missing the point.  Just get on there and say something.  It’s social media, after all. Follow some people, @ them a bit, get your feet wet.  No ideas but in things! I promise you’ll be able to figure it out after a while.  (Here’s a hint: ask for help on Twitter, tweeple love to help)

Some people say that Twitter is the most important thing since sliced bread. I doubt it. But if they’re right, why are you wasting time worrying about it when you could be living it?  And if it’s not so important, what have you got to lose?

So if you’re still fretting (yes, I’m still talking to you, G*****, and you too, L*****), here’s a handy tip: you can erase your embarrassing tweets. Permanently and forever. Just click the trashcan next to the goof – see it on the right?

The first time doesn't have to hurt

The last thing people need these days is something else to worry about.  I suppose that might be an argument for ignoring twitter entirely, but if you can’t do that, I urge you to just jump in.  There will be plenty of time down the road to laugh at ourselves for being so foolish or faddish.

Are avatars authentic or effective?

I was engaging in some micronarcissism (that means looking at my Twitter page) the other day when I chanced to notice that most of the icons – or avatars if you prefer – were faces, most of those photographic.

The old New Yorker cartoon said, “on the internet, nobody knows you’re a dog” (the cartoon showed an actual canine using a computer) yet here are some people using (I assume) their real faces for their online presence.

Before pondering the implications of that, a brief geektour of the numbers:

I classified twitter pictures into four types:

  1. Faces (photographic) – to the best of my ability to tell, photographs of one person’s face
  2. Faces (illustration) – faces but not photographic, includes illustration and overtly manipulated photos such as “obamifications” (which should be called “Faireifications” or perhaps “Obamanations”)
  3. Corporate or personal logos
  4. Other (body parts other than faces, bucolic scenes, pictures of animals, etc.)

Some Twitter avatarsOf the 36 icons pictured in my little “Following” bloc,

Faces/photo: 28 (78%)
Faces/illustration: 1 (3%)
Logo: 4 (11%)
Other: 3 (8%)

Of the top 50 Twitter Elite in the USA (via Grader)

Faces/photo: 39 (78%)
Faces/illustration: 4 (8%)
Logo: 4 (8%)
Other: 3 (6%)

The results are pretty consistent these samples.  Faces are in. Photorealistic ones, especially. I’m not sure if that has changed over time or if it’s always been the case.

Not that there’s anything wrong with that, but why?  I’m thinking that there’s a general movement in social media for authenticity and transparency, that you should say who you are and be real.  There’s a lot of software in our brains devoted to recognizing and understanding faces, and we seem to like to use it.  Faces humanize online experiences.

But let me take the contrary position for a moment.  Shouldn’t your online avatar or chat icon stand for you in a communication and marketing sense?  Isn’t it a small ad banner that you can use creatively?  And shouldn’t you at least attempt to stand out in the crowd or cloud?

On the one hand, if I don’t already know who you are, seeing that you’re a middle-aged white guy with unfortunate facial hair doesn’t add much to my online consumption of your updates.  On the other hand, once I start reading those things, seeing that photo might add depth or credibility to your online presence, and then I even stand a chance of recognizing you in person.

Here’s a post (that I found via a tweet from a logo avatar) about adding your photo to LinkedIn, which seems a lot more straightforward.  On LinkedIn, like Facebook, you’re definitely supposed to be you.  On Twitter or blogs, you could be a character, a brand, a team, all sorts of things.

What the duck?As a guy who uses a duck (you can sometimes still find my old icon, a rasterbated photo) for online imagery, I guess I could be accused of hiding.  But that icon serves pretty well:  it’s easy to recognize,  related to my online brand, consistent across social media sites, seldom changed so consistent across time, and pretty good at standing out in the crowd.