The View From The Data
I wish I could give you data on the number of times I’ve heard or read “data-driven” in the past month — but I stopped counting at a gazillion. (That’s completely mathematically exact, of course.)
With cloud wars heating up to support anticipated increases in data, and as an executive, Board member, and advisor to organizations focused on leveraging this data for better decisions, I’m constantly reminded of the benefits and pitfalls of a “data-driven” focus:
The Glass at 50% Capacity
We often hail data as a superhero — complete with amazing feats of strength and power, bringing truth, justice, and world peace. Data implies objectivity, a fact-based measure upon which to make decisions. Numbers seem indisputable, unbiased. If you can measure something, you can assess its value to each constituency and agree upon appropriate action. You can make automated decisions with “rules-based” approaches, creating efficiencies. You can uncover new insights, usually unseen or non gatherable with the human eye, providing opportunities to create and improve.
Like a superhero constantly saving the day, it’s hard to imagine you’d reject the help. And that’s the danger. Because data can be red kryptonite.
The Glass at 50% Capacity — Again
Many organizations that strive to be data-driven miss the mark, and actually do themselves more harm than good as they over-rotate on the importance of data in their decision-making and product offerings:
Data cannot cure a dysfunctional culture.
Jeff Bezos is famously quoted as saying, “The great thing about fact-based decisions is that they overrule the hierarchy. The most junior person in the company can win an argument with the most senior person with a fact-based decision.”
In theory, sure. But in practice, it’s often bullsh-t. Why?
Bezos’ statement assumes that people agree upon what they’re deciding in the first place. That’s sometimes not the case. When people seemingly ignore data when making decisions, it’s usually not about the data — or the current decision at hand. People ignore the data or argue “irrationally” because they disagree with the questions being asked, the answers they might get, and/or the objectives they’re each attempting to achieve. Even if they do agree, they may opposed what to measure and why. They might be privy to different information, which colors their perspective. And behind all these dynamics are all the inevitable unconscious, unstated, and clear motivations of those involved — and the implications for each of them.
Organizations that can’t build true alignment around those motivations are doomed — no matter how “data-driven” they are.
Data != strategy (Data is not strategy.)
Some organizations get so obsessed with data, they forget the reason why they started collecting using it in the first place. With all the tracking, measuring, and monitoring, the metrics sometimes become ends unto themselves. By measuring everything, along with measuring the right things (hopefully), organizations invariably measure the wrong ones, sending conflicting messages and potentially causing the misalignment referenced above.
Even if organizations have the right metrics now, as they evolve and the competitive landscape changes, the old metrics may not serve them as well as they once did. At advertising-driven Internet companies, for example, typically monetization follows footprint; the community or network often need to be built out in critical mass before you can achieve sustainable revenue generation. The success metrics must follow that shift as well. Data-driven organizations that don’t evolve their strategies may discover they’re not measuring for success.
Similarly, slicing the data to rest on laurels will not have companies resting on their laurels for long. So congratulations on being the market leader in striped blue socks sales for 18–24 year olds who also eat provolone cheese at least once a week. What’s your margin on that? Will you make it up in volume?
Data can dull empathy for the individual.
Sometimes data can open people’s eyes to the extent of a problem, or signal strength. People hail data and big data systems as a way to create personalized experiences for people. Indeed there are many examples of where personalization has, at minimum improved the product experience, and in the best case scenario, has the opportunity to improve or even save someone’s life.
On the other hand, wearing a big data filter can cause decision-makers to affect individuals while distancing themselves from the decision more easily. (Just because they may be the “right” decisions shouldn’t make them easy.) Layoffs are a lot more difficult if you know each individual personally versus basing the decision on lists and statistics. While we lament hearing the numbers of people killed in conflicts, seeing individuals affected or hearing their stories is often exponentially moving. Put another way, a picture is worth a thousand data points.
The Story in the Data; the Data in the Story
To make data work in an organization, we need to acknowledge the knowledge in the numbers, the qualitative in the quantitative, the stories in the statistics. Reconciling data and story doesn’t undermine us. Rather, it frees us to manage more effectively:
What’s your success story — by the numbers?
Data neither will fix your problems nor ensure your success. Focusing on high-quality measurement stories requires strategic vision and discipline. It takes a thoughtful, confident, and committed leader to achieve “story-data equilibrium” — and to inspire and align others to do the same.
Manage the story and the data with an open mind.
In a blog remembering and honoring Andy Grove, Ellen Konar distinguished between research to prove a point versus research to make a decision. You (and your stakeholders) may not always like what the data is telling you, which can be unsettling — and which may require you to change your narrative and rebalance your story-data equilibrium. Leaders who understand their organizations’ value, their success measures, and their true north are much better positioned for success.
Be “data-story driven” not just “data-driven.”
It’s easy to say things like “garbage in, garbage out,” and “If you torture the data long enough, it will confess to anything.” It’s a lot harder to walk the walk about the facts and opinions and stories behind data-driven results. We need to be “data-story driven,” not just data-driven.
For example, after years of data collection, DataArts (where I’m on the Board) now brings data fluency to a stereotypically unexpected domain — arts and cultural organizations. Through its platform, DataArts helps organizations understand the data they could gather and how and why they should, and what metrics are important and why, to facilitate better decisions, management, and storytelling to attract donors, audiences, and supporters. And with it, the ecosystem is bringing, among other things, economic activity, education, and of course, diverse art and cultural programming. Creative meets analytical. Art meets science. Data meets story.
Conclusion
Entrepreneur and venture capitalist Vinod Khosla has said that the hardest and most lucrative problems to solve are non-technical. I agree. Data is the easiest and most straightforward part. The hardest problems will always require judgment, context, and acknowledgment about the choices we make and the narratives we tell ourselves. It means leaving simply “data-based” thinking behind (pun intended), and building self-aware stories into our management.
