For those of you who had a quick flashback to high school sports complete with cheerleaders, sorry about that. But, in the age of POS systems, big data, analytics and visualizations, it’s hard to believe we’re still asking this question.
Most of our businesses, even small businesses, are awash in data – transactional data from our ERP systems, customer sentiment from our marketing management systems and financial data from our accounting systems. Long gone are the days when we polled subsets of customers to predict the behavior and preferences of the population at large. We can easy pull together and analyze the actions of every one of our customers and the financial impacts of those actions in our organization. We know what they bought, when they bought it, what they paid for it and how they liked it after the fact.
So why do we still struggle to make data-driven decisions? The short answer is cognitive biases – a mistake in reasoning, evaluating, remembering or other cognitive process, often occurring as a result of holding onto one’s preferences and beliefs regardless of contrary information (Chegg). As Anais Nin said, “We do not see things as they are, we see them as we are.” In my work, I’ve observed 4 specific obstacles to data-driven decision making. I want to offer some suggestions on how we can deconstruct them and replace them with something better. As futurist and philosopher Alvin Toffler said, “The illiterate of the 21st century will not be those who cannot read and write, but those who cannot learn, unlearn, and relearn.”
We believe all smart people think the way we think. It’s not surprising that we interact with people who think differently than we do. The variety in our nature, nurture, experience and education guarantees that no two of us are exactly alike. However, the remarkable thing is, we think those people who do think differently from us are not nearly as smart as we are. We believe, that presented with the same set of facts, all smart people will draw the same conclusion, make the same selection or opt for the same methodology. That’s not the case. Different isn’t dumber. Different is just different. When we examine data and discover findings that don’t square with us intellectually, I see a couple of choices – make a decision that aligns with the data and entertain the option that the data’s disagreement with your own opinion might not be an indictment of your intellect or, if you just can’t get over it, dig deeper and find the why behind the findings. Sometimes the data presents customer choices that have a root that you’ve yet to discover. Whichever option you choose, you’d probably best judge your intellectual horsepower with this quote from F. Scott Fitzgerald, “The test of a first-rate intelligence is the ability to hold two opposed ideas in mind at the same time and still retain the ability to function.”
We believe that the way things worked in the past will be the way they will continue to work in the future. We all seem to have an affinity for systems and processes that we know and have experienced. The line at the grocery store is more comfortable than the Blue Apron box delivered to our home. However, when we see data moving towards a new sales channel or towards an emerging product and away from an existing product, we must muster the courage to follow the data. If we, personally, are an early adopter, it might seem easy. But, if we’re part of the late majority in adopting a new product or service, it borders on the painful. Battling this bias requires more academic rigor than the others. I’d encourage you to examine the histories of formerly successful companies who assumed that the business model they had ridden for, in many cases, decades would continue to return stellar revenues for them going forward. Think about Digital Equipment Corporation (DEC), Kodak, Blockbuster and Toys-R-Us. When well-vetted data says it’s time to make a change, it’s time to follow the data.
We tie our own personal worth and identity to our tastes or work product. Maybe I should leave this point to Dr. Phil, but I’ll take a shot at it. Over the course of a long career, we will all come up with some great ideas. We’ll also come up with others that could use some work. Unfortunately, we’ll probably love both types just the same. When the data shows that the widget we designed isn’t gaining traction in the widget-buying community, we take it personally. Sometimes even more painful is watching customers lose interest in a product or service in which we have an intense personal investment. Maybe it’s been the staple of the organization for a very long time. The customer’s decision to purchase something else feels like personal attack on us. Make no mistake – you are not what you do. Your worth is not the number of times your product is rung up at the register, sold online or positively reviewed on Google or Facebook. Living that way will drive you crazy. Data is just data. It reflects the collective sentiment of the population who provided it regarding a single item or interaction. It is not a measurement of the worth of the person who created it. Get your worth from something that cannot be taken away. I personally find it in my Christian faith.
Make no mistake – you are not what you do. Your worth is not the number of times your product is rung up at the register, sold online or positively reviewed on Google or Facebook. #datadriven
When the data creates this situation – and it inevitably will – separate your personal tastes and most-prized creations from your personal worth. To paraphrase Rudyard Kipling, “treat the two impostors of customer love and customer rejection just the same.” Make decisions consistent with the data and move on.
We believe that experimenting with something new is expensive and risky. As we examine the data and the tide seems to be turning to new products or delivery methods, we assemble, in our heads, an entirely new manufacturing facility, a complicated new delivery infrastructure and sophisticated, new customer service capabilities. Each of these carries an excessively high price tag. Before we know it, in our heads, we’ve retreated to the comfort of the status quo before we even start. If the data indicates movement towards a new product or service, it’s a good time to employ a methodology from Jim Collins’ book, Great by Choice – Fire bullets, then cannonballs. Before creating an expensive, new infrastructure for a new product or service, construct a low-risk, low-cost, low-distraction experiment to prove the new direction indicated by the data. The ability to calibrate the offering by taking small, measured shots (bullets) and evaluating their appeal and effectiveness can be followed by crafting full-blown products (cannonballs) with the benefit of the empirical evidence you’ve gathered during the test. Some concrete ways to implement bullets then cannonballs – create a 3-D printed version of a new product instead of a full-featured version from an assembly line, roll out a service to a small test segment of your customer base, outsource the support of a test item to a third-party who could rapidly ramp up the support function and quick shutter it when the test is over.
Making the move to data-driven decision making isn’t easy. It often flies in the face of our “gut” and it often has a higher emotional price tag. But, when it’s all said and done, it’s the right thing to do for the organization. The findings from data analysis force us to have discussions we need to have. Implementing data-driven decisions reduces unnecessary risk and position us for success. The new decisions will create more data that we can examine and use to further refine our work.