Let’s step into a time machine. Imagine you’re a decision maker in the year 1943. You lead the Statistical Research Group, a top-secret, data-informed think-tank in Manhattan. Your mission? Give aircrafts–bombers and fighter planes–their best chance of getting home. The team that you’ve assembled, in the words of your colleague (and the SRG’s Director), W. Allen Wallis, are “the most extraordinary group of statisticians ever organized.” You know they’re up to the task. So, where do they start?
Like many of today’s data-driven marketers, the SRG began with collecting data. They assembled charts and diagrams that detailed, down to the inch, every single bullet hole on every single plane that had made it home. Soon, they began to notice a few trends.
Many of the planes were riddled with bullet holes in the same spots: their wing tips, tail, and the center of their fuselage. The SRG’s mathematicians began to perform complex calculations, aiming to figure out how much armor to add to these apparently-vulnerable locations, balancing the plane’s weight, flight performance, and the cost of the armor in an attempt to find an optimal solution. They were pleased with their findings: the data suggested a handful of key, consistent improvements, and they felt that a significant improvement was in sight. Mission accomplished.
Until their findings landed on the desk of Abraham Wald. Wald grew up in what eventually became Romania, and was recognized as a gifted mathematician even in his teenage years. As Austria’s economy struggled in the 1930s, Wald emigrated to America, and soon found himself as a Professor at Columbia University, where his involvement with the SRG began. Though he was surrounded by mathematicians of equal computational talent, Wald had a special knack for the big picture: to him, the challenge typically wasn’t finding the answers. It was asking the right question in the first place.
As Wald reviewed the SRG’s recommendations, he was struck with a realization. The right question, it turns out, wasn’t how to reinforce the planes that had made it back. It was understanding what happened to the planes that didn’t.
Parsing through the data, Wald realized that it was the spots that didn’t appear to get hit that were truly the most vulnerable. The data’s absence of bullet holes in engines, the cockpit, or the forward empennage (the front part of the tail) wasn’t because those spots weren’t hit–instead, it was because planes that were hit there didn’t make it back.
Wald’s realization led the SRG to completely reconsider their approach. Instead of recommending reinforcements to the wingtips and tails, they focused their attention on adding armor to each aircraft’s engines, their true weak spots. Their recommendations were implemented, and countless lives were saved as a result.
In marketing, our theories seldom possess the significance of reinforcing fighter planes, but they suffer from the same challenges. Our analytics strategies often focus on developing a detailed account of what happens: who visits each page and what they do when they get there. Though this isn’t useless information, it’s trapped in the past. It doesn’t help us become smarter for the future.
Instead, we need to take things one step further. We need to ask why. Why is this landing page converting so effectively? Why do customers abandon their carts so frequently? Why do users who discover us via social media become much more valuable customers?
Asking “why?” is harder than asking “what?”. It forces us to leave the comfortable–and, often, misleading–confidence of analyzing numbers and drawing conclusions. It pushes us to hypothesize, experiment, and sometimes be wrong–but it’s the foundation on which novel, insightful, and sometimes crazy-sounding ideas are built.
Moving From “What” To “Why”
Believe it or not, asking “why?” begins with knowing “what”. In order to test our hypotheses, we need a way to assess and measure them effectively. Tools like Segment, Mixpanel, and Google Analytics empower us to assemble detailed, nuanced reports that provide us with insights that are far deeper than common, high-level key metrics. Without measurements and insights that are tailored closely to the unique facets of your business, we’ll never be able to ask “why?” effectively.
I shudder when I see data-driven marketing strategies that are entirely focused on generic variables: conversion rates, pages per session, and time-on-site. We need to take things deeper: developing our own in-house metrics aligned with our broader marketing strategies, our specific content initiatives, and each step in the conversion process. We can’t stand out from the crowd if we’re only measuring the same things as everyone else.
To be effective, our strategies need insight into everything we’re doing. Organic and paid social media marketing? Content redistribution through other channels, like Medium, YouTube, or Instagram? Our customer’s relationship with our brand begins long before they make their way to our website. If we don’t understand how that relationship develops from day one, we don’t have any hope of improving it.
Once we have a solid measurement strategy in place, we’re ready for the fun part–asking “why?”. This is where the real fun begins. We’re no longer in the world of quantifiable and verifiable data. “Why?” is asking questions and testing theories. Do social media leads convert better because we’re offering them better deals? Let’s test it and see. Does this lead page work better because it’s in a different tone of voice? There’s only one way to find out–try it out somewhere else. “Why?” empowers us to put our creative thinking caps on, and do what we love to do–be innovative and try crazy things.
If your data-driven marketing practice feels boring, or isn’t delivering results, it’s because you aren’t asking the right question. You’re asking “what?”–not “why?”