Andrew Lang, the poet, literary critic and anthropologist, famously said, "He uses statistics as a drunken man uses lamp posts - for support rather than for illumination."
The modern day version of this that I've heard used throughout corporations is that they use *research* like a dunk uses a lamp post - for support not illumination.
I have since developed this into my own analogy about the current social media research and measurement tools out on the market.
Research:
This is the ground you stand on. Research is often used to answer the question "Where are you?" and where do you want to go.The sad thing is that companies usually spend most of their dollars backwards looking not forward looking: "Where have we been?" versus "Where are we going?"
Measurement:
This is the lamp post. The lamp post has many levels from basic, "This is what happened", to "This is what's happening now" but by themselves they almost all stop there. Without interpretation and consulting (this is why I still have a job) measurement doesn't provide the "Here's where you need to go and this is how you get there."
Most marketers use this to justify the dollars they spent not learn where to spend their dollars next time.
Illuminating:
Eventually we need tools and statistical models to provide the right illumination. While some of us are better at interpreting trends and data (seeing in the dark) than others we are still biased, we still see the data the way we want to see it. This may seem unrealistic to some people but imagine analytics that recommended what you should do next in any given situation? This is broad and vague intentionally.
Take the insurance and credit card companies. They have predictive analytics about each of us. They know that if you pay your bills late you are more likely to get in a wreck. How do they know this? Mountains and years of data (literally, it's all stored inside mountains).
Imagine a set of tools that told you because a certain person book marked an average of 10 items a day using Diigo or Delicious that they were 5X more likely to be an influencer to a friend or family members purchasing decision than someone who shared the same amount and types of data on Twitter.
Now we could use primary and secondary research and analytics to figure out if that was true or not but there isn't a tool that has accumulated enough data to tell us that. And to do that we need more than just social media data.
We will get there eventually. This may sound a little about the Social Media Dystopia I wrote about before it doesn't have to be a violation of our privacy or creepy.
Don't Forget the Rest of the Lamp Post:
There is still a lot of money to be made on selling the lamp post and the ground we stand on.
As we race to provide illumination there is still a lot of value in knowing where you stand and having that lamp post to lean on. In fact you still need all of this to provide enough elevation for any illumination to be effective.
If, like me, this is a large part of your job, don't worry about being replaced. Companies will still need people to tell them which streets to go down. Having lamp posts lighting the way means we can move a lot faster.