By now, you're reading every Big Data article with dread. They all say the same things:
- It's important
- It's inevitable
- It's the key to competitive advantage
- You must understand it to master it
- You can't understand it without a Master’s in rocket surgery
I believe the first three and offer up just enough herein to disprove the last two.
If you run an airline company, you do not need to be able to fly a 747. But you will be expected to know how to manage the pilots and people who are responsible for fixing the Dreamliner's batteries.
If you run a massive retail chain, you do not need to build a Web site. But you will be expected to understand the limitations of e-commerce enough to value the director who prioritizes Twitter and Pinterest over Instagram.
At Least Know This
In marketing, Big Data refers to the advent of behavioral and attitudinal data that overwhelms the comparatively minimal transactional and survey information we had 20 years ago.
A few technical advances are central to the three Vs of Big Data (Volume, Variety, Velocity).
Disk drives became cheaper so we can store more data, and we can split up the processing across a multitude of processors that have never been so cheap. As a result, we can take that wide variety of data (unstructured social media blather, for example), capture it, store it, and preprocess it. That makes them fit fodder for the big, familiar analytics systems we've been using all along.
That preprocessing is the third technical advance. "MapReduce" is the beginning of Big Data technology: We're mapping a massive problem across countless processors and reducing the results back to the mother ship where statisticians can mine for business-informing trend and insights.
Why Big Data Rocks
"What if?" is a trigger for a predictive analytics model building process rather than an idle rumination. Unfortunately, it's easy to ask 1,0000 questions that launch 1,000 models--at a cost. Big Data lets you answer marketing imponderables.
The more jigsaw puzzle pieces you have, the more likely you are to find pieces that fit together. This kind of customer will respond to this sort of promotion in this general circumstance if he has previously been exposed to this type of collateral.
But what makes Big Data worth the Big Bucks is that it gives you answers to questions you didn't think to ask: the unknown unknowns.
The machines are being taught to find patterns humans didn't think to look for. When you discover that Web site traffic from the NASCAR site buys 45 percent more baby strollers than any other source, it does not matter why. All that's important is you can now monetize that traffic better and faster.
That enormous pile of unstructured (messy) data is rife with insight. But only a computer has the wherewithal to pull it apart, put it back together, and find the hidden picture.
What's A CMO To Do?
Now that you know enough to join the conversation, it doesn't matter that you can't write your own SQL query. Chances are superb that you can't assess the ink-dye sublimation process on a 4-up off, offset print job or light the sound stage for a 30-second video product spot taping. You don't need to.
You do need to understand how to manage the people who understand the details and technology. You do need to know what motivates them and how to help them be as productive as possible.
Here are some tips.
1. Help Them Automate
The real value of Big Data comes after the data has been collected, cleaned, normalized, moved, integrated, and reported out in a 100 dashboards and considerably more reports. Sadly, these things take a great deal of time and offer little value.
Analysis is the creative part--the rewarding part. Help your data team move through the mechanics as fast as possible so they can get to the fun stuff. Help them automate the sifting of the flour, the mixing of the batter, and the baking of the cake so they can get to the fun part of decorating it with icing, sprinkles, colored sugars, edible pearls, and dragees.
This analogy breaks because a cake with no frosting is still edible and tasty and will go to your waist. Data with no analysis is simply a waste.
2. Lovers Gonna Love
Technical people love technology, so be patient when they wax rhapsodic about open-source cluster management, capacity scheduling, multitenancy, and data discovery checkpoints. Then gently bring them back to the business problem.
3. Give Them A Proper Problem To Solve
Data-loving people have raw material (data) and tools (software and systems), but they need a real problem to solve. "Take a look at this and tell me what you see" works in medicine because the goals are to avoid disease and improve health.
Telling patients to eat less and exercise more is like telling data scientists that you want to increase revenue and lower costs. Yes, of course. But how would you like to do it today?
Invest some time explaining the art of marketing from brand attribution recall and direct response to sales and customer satisfaction. Get as specific as possible about what you are trying to accomplish at the moment, be it to raise awareness, improve opinion, influence influencers, inspire engagement, optimize for lifetime value, or drive endorsements.
4. Convince The Creatives To Create With Data
Big Data has another management downside: dealing with the naysayers and refuseniks.
- You can't replace great creative with numbers!
- Marketing is an art, not a science!
- Your Big Data can never come up with Big Ideas!
Last generation, we had to deal with Photoshop luddites. Using a computer to draw a picture was computing, not art. Now unaltered snapshots are their own category in photography competitions.
Explain that they can create tests. They can create what-if scenarios and work with the dynamic content management platform to invent new, creative ways to persuade customers.
Some will get it. Others will not.
Your job remains the same: using all available tools (from pencils, sandwich boards, fliers, and bumper stickers to retargeting, social meme surfing, personalized mobile apps, and multidimensional segmentation) using a team of diversely talented people. Give them specific goals. Tell them what you want them to accomplish and then let them figure out how.
And let us know how you're getting on. We can all benefit from each other's experience.