Catching Up with Scantily Clad Analytics Emperors

Don’t let tales of others’ analytics heroism scare you — the data race is far from over.

Time to panic!

Every organization boasts awesome analytical prowess — except yours. Infiniti Red Bull Racing deftly analyzes gigabytes of data each race to boost performance. Hollywood filmmakers mesh emotional data collected at every heartbeat with complex algorithms and data-based decision making to craft optimal stories. They’re all doing it. And making it look so easy.

But despite public declarations otherwise, deep down, you may know that your organization isn’t using data well. You fear that an insurmountable disadvantage has developed based upon the gap between the analytical capabilities that you know your organization has and what your competitors seem to have.


In the well-known fairy tale by Hans Christian Andersen, everyone incorrectly believes others can see the Emperor’s fine clothing. Having been told that only the wise can see the cloth, everyone fears that they alone lack the ability to see it and are uncomfortable with publicly admitting this is the case.

Similarly, I suspect that our impressions of the analytic capabilities of others may not be entirely accurate. Why do I think so? Because reporting bias and social desirability make it likely that reports of analytics capabilities are, at best overblown.

Consider: positive reports of analytical capabilities get attention. We all like epic tales where analytics reveals hidden insight and creates value. It can and it does. But we see far, far fewer stories about organizations that are not using analytics at all, or that are using analytics poorly. Our impression is formed by the reports we see, not by the reports we don’t.

Moreover, with the current popularity of analytics, organizations have incentives to highlight their analytical capabilities so others will view them favorably. When analytics stories are told around corporate campfires, data sizes and capabilities grow with each retelling — they’re probably as elastic as the fish that got away. The stories just sound better with more data, more complicated tools, and fancier models. With the current emphasis on analytics, no one wants their organization to be data-inept — the “fool” of the Hans Christian Andersen tale who can’t see the miracle cloth draped on the Emperor.

The good news is that, if you know your organization needs to improve its analytical capabilities, you may not be that far behind everyone else. Despite our distorted impressions, many others are in similar situations, relying only on managerial intuition or basic spreadsheets instead of on complex prescriptive models that use massive, streaming, unstructured data.

But while it’s not time for panic, it also isn’t time to rest on your laurels. At the risk of mixing up fairy stories, take a page from Aesop: if you’re truly in a race between the tortoise and the hare, and you’re the tortoise, understand that your competitors may or may not be caught napping at any point in the race — but whether they are is irrelevant. Your goal is to finish the race — so you have to keep moving forward at the best pace you can.

Starting the transition to becoming data-oriented can seem daunting. But it doesn’t have to be. Here are two ways to make your organization’s first step towards analytical prowess and catch up with the speedy hares — who may not be as far ahead as they seem.

Reduce the barrier to getting started

Even the best-intentioned people won’t take an analytical approach if it is too hard. For organizations that are lagging, the status quo is the path of least resistance. VisiCalc didn’t invent algorithms; it just made them easy for anyone to use. What can organizations do to make it easier to get started?

  • Provide examples of analytical models built using your organization’s data with your organization’s tools. It is much easier for people to create valuable models by extending and modifying; starting from scratch is tough.
  • Codify analytical knowledge (such as data descriptions, guidelines, examples, templates) in a collaborative tool like a wiki. Quickly modifiable repositories not only prevent duplicate efforts, but also allow efforts to build on each other.
  • Predict the most useful data for your organization and clean, prepare, or format it for your organization’s tools. You’ll sleep better knowing everyone isn’t looking up how to import the data again and again.
  • Anticipate organizational concerns with data use. If there are rules or restrictions associated with some data, spell them out so that people have realistic expectations about what can be done. Siloes and restrictions may be unavoidable; better to know before plodding forth.

Augment individual skills

At the core of projects are people applying skills to problems. But without analytical skills to apply, there is not much hope of analytical growth. What can organizations do to promote building individual skills?

  • Curate content to make it easy to access and build on. It is true that there are vast resources on the internet to help people learn skills. But the vastness is overwhelming, particularly when building initial skills. Reduce the search space by providing content specific to your organization’s tools, context, and data. With collaborative tools, this content can grow and mature to provide a rich, organization specific resource.
  • Encourage running with scissors (or at least jogging). Misuses of analytics tools are inevitable as people begin to learn. Give people some time to experiment, take risks, and mess up — then use the experience to add to the organizational learning by including it in your collaborative tool.
  • Exploit differences. Within your organization, people have a variety of skills and backgrounds. As people figure out cool skills, provide a forum to spread them. For example, a brown-bag series can both showcase tools and build a culture of learning.