One thing Social Media “gurus” are experts at is making up definitions for words that already have established meanings. I bet old Merriam & Webster are pretty pissed about this and rolling in their graves. Wait are they dead? Alive or deceased, it’s pretty ridiculous for people to think they can change the meaning of something just because the context is new or unique.  This is exactly the case with one of my favorite social media buzzwords “Engagement.” Yes, you could logically argue that in the 15th century, when the transitive verb first entered into colloquial speech, that it’s creators weren’t thinking, “gee we ought to put a clause in here for how this relates to internet marketing.”

No of course they didn’t. They couldn’t predict the future, and they were marveling over the recent discovery of my favorite portion of the food pyramid- chocolate. But I am not going to judge them because-hey-even though they didn’t throw us a media tailored definition,  the standard serves as enough to easily judge any type of social media related meaning we could throw at it. Look at the beauty of M&W’s fine work:

Engage: to hold the attention of: engross <her work engages her completely> b: to induce to participate <engaged the shy boy in conversation>

Now you could argue that anytime someone liked something a brand produced that they were technically “engaged” with the brand.  Let’s be blunt here- my specialty.  Do you walk around telling people, wow I am really engaged in those boots right there, I have to have them? No you don’t. Okay. Well.  I know you’re yelling at me in your head saying- “boots aren’t content, stupid!” I ask you this, when you read a blog, watch a youtube video, or listen to something on the internet how often would you feel comfortable using the word “engaged” to describe your experience? And further more how often do you share that content that you read, watched, ore heard that did not “engage” you? I would bet the family farm that it’s probably around 85% of the time.

Now I’ve heard & read time and time again that engagement is best measured in RTs, reblogs, shares & etc. I don’t agree.  RTs don’t show engagement. Why?

1. How many times do we retweet something solely because it has a good title? I’ll admit I’ve done it more times then I am willing to say. We ALL have. If anyone says they haven’t they’re a liar and -if you believe in some sort of god-he knows when you lie…

2.  A retweet doesn’t show any actual engagement with the content.  It just shows that you know how to copy & paste, or if you’re using more “advanced” means of tweeting, it means you know how to click a button. Clicking a button is not engagement, it’s showing you’re about as smart as the average monkey.

3. That said, even if you did read the content and like it and retweeted it for all the right reasons, did you really engage? Do we measure engaging with a TV commercial by how much people liked the content?  No, we don’t. There have been plenty of awesome & amazing commercials that have generated little engagement. I know several ad agencies that can tell you all about this.

This leads me to what I would like to call my REAL Social Media definition for engagement: (ta da)!

Engage: to hold the attention of: engross <her work engages her completely> b: to induce to participate <engaged the shy boy in conversation>

Duh. You should have seen that coming.

PS. if you want to “engage” with this content leave a comment, start a discussion, or etc.

PPS. If you don’t want to “engage” and just want to share this, please do. Even though it isn’t a good measure of engagement, it is a good measure of popularity.  Just like an insecure teenager, I don’t mind sitting at the cool table.

Everyday on twitter, facebook, linkedin, etc I see posts on how to measure Social Media.  Usually while reading these posts I have to resist the urge to find a knife and go on a “people who cheapen my skill set in the social media space” killing spree.  For the most part the measurement solutions proposed in these articles lack context,  double count metrics, and fail to isolate an overall measurement goal. The result is equations (usually to measure ROI or some variant of ROI) that have enough holes in them to strain the pasta for my Sunday spaghetti dinner.

Yesterday I got into a nice discussion with  Olivier Blanchard about just how twisted many of the these measurement suggestions are and how they actually gain notary in the space. The conclusion was simple: people are making a quick buck off of people who don’t understand what measurement really is. Or in my candid and oh so witty manner of speech- people are getting played. In general, measurement is facing the same issue much of the social media world is facing- everybody’s “expert”.

So holding all this in the back of my head, I tried to understand how people get guided so off course by both these articles and their own better judgment. At around 5 o’clock last night it hit me. Adding qualitative aspect to a previously primarily quant based world has thrown some people so far for a loop that are willing to accept complete gibberish as a viable marketing solution as long as it has words like “tweets”, “likes”, and “posts” built into the equation. Additionally, you have an influx of tools to measure social media that don’t really MEASURE,  they MONITOR.  Did you see how I caps locked those words? It because we’re getting to my thesis: Social media monitoring is not social media measurement.

Why do people group these two very different things it one haphazard mess? A) because they can B) because people are dumb enough to listen to them c) because there’s a confusing mass of a no mans land in the middle that I like to call “reporting”.

So what really constitutes measurement and what types of analysis are simply monitoring? Well it just so happens I created the chart below to help clear up the misconceptions. Note you won’t see the words “engagement” or “volume” in this chart. This  focuses solely on the outputs and ignore the inputs ( ie: variations of metrics & types of  content).  Why are we ignoring the inputs? Because any good data nerd knows that the inputs should be guided by your companies internal goals and not by some one with a white board and a video on youtube. I could jabber on and on  about what each of these is and why they fall where they do, but this post is already too long. I’m going to let the image speak for itself.  In entries to come I will be analyzing various proposed measurement methodologies, pointing out the potholes, and giving credit where credit’s due. Thoughts? Comments? Bitter rage? Let me have it!

Special thanks to @devintrix &  @norcross for  listening to me babble last night about this.

I know this video is old, but I remembered today how angery it made me. This is me calling it out.

I’m a data nerd and as much as I hate to admit it the first thing that happens when I see a video like this I get pulled into the numbers. I immediately begin to back track and comprehend how they came upon the random slew of statistics.

Then inevitably I get angry and begin to realize that these statistics are either immeasurable of completely ridiculous. Take from the piece above. “India has more honors kids than America has kids.” How can you measure that? Does the US or India have a way to measure honors kids? And if they do does there definition of an honors student equal the same thing? And as my mind begins to spin off into a tizzy of possible measurement scenarios, I realize the slow drip quicksand has caught me-out of context data vomit.

It’s hard not to fall into the trap; our society routinely practices data bulimia. We choke down stakes of facts and figures and just regurgitate them out often with little planning. It hard not to with all the new ways to collect data; even the most useless facts seem alluring. Did you know that Mosquito’s are attracted to the color blue twice as much as to any other color?

But numbers are useless without a well framed purposed and without context statistics are not actionable. For example, let’s say Fred’s Fish Bar does a survey of customers and finds delivery to be the least satisfactory service in the restaurant. They spend a large portion of their next quarters budget revamping the delivery service only to see no change in results.

This is the set up for the classic contextual faux pas. Digging deeper into the data find that while delivery WAS the lowest scoring service, it was only marginally lower. Also, the survey was only administered to people who visited the restaurant for a sit down dinner-those who may not find as much value in delivery. Also, they find that while family dinning wasn’t the least satisfactory service, it was significantly lower among women. Perhaps the investment would have been wiser spent creating a family friendly perception of the restaurant targeted at women. But, who knows because this is all hypothetical- ie I made it all up.

But the issues addressed in the fantasy example all too easily plague real world businesses. How do you get data that isn’t just compelling, but actionable? It’s simple ,build a frame work. Take time to invest thought into the following 3 questions:

1. What are you looking to measure, and why?
2. What do you plan to do with results of your research?
3. Can this metric be backed up by any additional data?

So lately I’ve been seeing a slew of articles floating around the twittersphere about whether or not teens tweet. Here are just  few of the many:
Do teens hate Twitter?, Teens don’t tweet… or do they?, Why Teens Don’t Tweet, Are teens the driving force behind Twitter?, & Why Do Teens Shun Twitter?

Now I could go on and on about my personal opinions on who I think is right, but that’s not what this entry is all about.  From all the chatter around these articles it obvious that the general population is unaware of how much bias there is in survey data collection. In fact if you can find me a completely unbiased survey- I will buy you a pony. Yes, I will buy you a brand new pony.  All. For. You.

To understand why it’s impossible to prevent some type of skewed result, I probably should explain what types of things create measurement bias & affect survey results. For example,  a user can be feel pressured the give the response the surveyor wants to hear, that’s a response bias.  Additionally, many  surveys rely  only on the population that chooses to respond, which leads to voluntary bias & sometimes because certain groups of people choose not to respond it leads to under coverage bias.  Plus all those people who choose not to respond to the survey can create a non-response bias. If you want a more in-depth look at these bias..  click here

Now back to the whole teens twittering debacle. Why are so many of these studies have conflicting results? Well, it’s that bias beast I just outlined. Since you can’t force anyone to take a survey and it’s difficult to create a representative sample of the Internet population, it’s likely one or more of the previously mentioned bias are dramatically impacting the results in each survey. So in some case the results show teens are twittering.  In other it shows they’re not. And in some there’s not clear results. It doesn’t matter how big the name is of the company who preformed the study, there”s likely to be some type of flaw to point out.  Why do you think Neilsen’s in such hot water for how they measure TV? It’s the harsh realities of survey measurement.

So at the end of the day until companies find a method other than self report to determine a twitterer’s age, you’re going to have to take everything you read with a grain of salt.  It’s the facts of life,  and the facts of survey data.

Today I was bumbling through my new followers for the day on twitter and attempting to return the favor to those who appeared not  to be spammers. I noted to myself that my follwers  list was inundated with screenames such as   “sexxyblond4fun” , “brtnyffedvids”  & “cheapslutontwitter”.  Each was accompanied with an equally as  scandalous avatar of a girl giving her best “i’m sooo naughty ” face.  After seeing bikini after bikini as I cleaned out my account,  it hit me.  Spammers are sexist.

How come I don’t have any spammers following me with names like “nakedbradpitt” or “sexyboytoy”.  Why isn’t my followers list filled with attractive men with their shirts off and wearing a “come hither” look? If men get to look at hot,  sassy & supposedly desperate for clicks women such as this,   I want the same.

I want abs. I want oiled pecks. I want boyish looks. I want no shirts. Is that too much to ask?  Heck, I might even be persuaded to click one of those spam links. I mean Brad Pitt can make me do things I would, in normal life,  never ever do.

Call me shallow, but every girl I know wouldn’t mind if an attractive man followed them around even if it was virtually. The way I see it,  you spammers are really ignoring us females.  You’s be suprised what a hot male avatar might be able to get us to do.