Mark Twain and the Need to Move Beyond Simple Metrics

Thoughts on exclusively relying on basic statistics.

I can’t speak for every nonfiction author out there, but I have always found that each book takes me in new directions and opens unexpected doors for me. Always.

To be sure, the same tenet has held true with The Visual Organization. When I reflect on the book nearly one year after its publication, I shake my head at some of my recent interactions, speaking gigs, conversations, and contact requests. Those who think they know exactly what the future holds are almost always wrong.

Moving Beyond “Metrics”

Relying upon simple “metrics” can maximize the chance of making the wrong decision.

Still, not all of these events and new relationships are completely surprising; many were downright expected. For instance, I’ve had many conversations with folks about contemporary dataviz. I’ve spoken to my fair share of executives about how they use data to do their jobs. Most use KPIs, dashboards, and other numerical ways to manage their business. The most dynamic of the lot, however, understand that relying exclusively upon quick overviews in the form of “metrics” is not only dangerous; it can maximize the chance of making the wrong decision.

Look at following charts. What do you see?

Click to embiggen.

(Thanks for the photo, @pickover and @peeterskris.)

You might think that the figures above are wildly divergent. In a way, you’re right. In another, though, you couldn’t be more wrong. The mean, median, and variance of each dataset is identical. You don’t have to be Nate Silver to see that basic statistics in this case mask very different stories and trends.

Simon Says: Data Must Be Contextualized

As Mark Twain once said, “There are lies, damned lies, and statistics.” It’s one of my favorite quotes, and that statement is as true today as when he said it more than a century ago.

KPIs and their ilk can certainly provide high-level overviews of the health of a business or department. Without context, however, those metrics lie somewhere between meaningless and incomplete.

Feedback

What say you?


IBM sponsored this post.

philanimated

Navigation

BACKRANDOMNEXT

YOUR AD HERE

Filed Under



Enjoy this post? Click here to subscribe to this RSS feed or here to sign up for my bi-monthly newsletter.


Submit a Comment

Your email address will not be published. Required fields are marked *