2017 - Reaching Full Learning Potential With the Right Mix of Data, Machines and People

By: Kevin Roberts, Director, Data Scientist, Valassis
Published Friday, Dec 2, 2016



2017 - Reaching Full Learning Potential With the Right Mix of Data, Machines and People

As 2016 comes to a close and we look forward to the new year, it is expected that the following trends will continue their revolutionary impact on our world:

     - Mobile and other “smart” device presence and usage will continue to expand rapidly, and
     - Data, both produced and collected, will continue to grow at exponential rates.

With a rapid increase in available data, how do we integrate the right mix of consumer data and use technology, to process data in an efficient way so valuable insights are quickly found? 

An explosion of data leads to a more individual consumer view 

Technological advances and inserting computers into just about every device (watches, refrigerators, etc.) have introduced countless new opportunities, but also many new challenges.  While the sheer volume of information being constantly created can make data difficult to manage, it also opens the door to more possibilities for building a personal, 360 degree view of your customers.  Mobile device usage and location tracking has changed the time and place consumers research products, compare prices and look for offers. This, in turn, has opened a direct avenue of reachability at a moment of increased receptiveness. 

The availability of data and the view it provides into the minds and actions of consumers, makes it even more imperative for marketers to understand what the data reveals. We must allow it to guide us to a deeper understanding of the consumer and, ultimately, enable us to reach that consumer with the right message at opportune times.  In other words, it is important that we listen to the key signals that individual consumers are sending out on a daily basis so we can measure the data to learn more about each consumer and their response to products and marketing campaigns. The integration and combined power of online activity, real-world physical behavior and demographic attributes (purchases, coupon usage, location, age, income, etc.), coupled with marketing strategy (message, offer, frequency) is key to unlocking a true understanding of consumer preferences both today and into the future.   

Speed and agility are achieved with the cooperation of marketing and machine

The turn of the calendar brings yet another reminder of how quickly time passes and how fast things change. This includes the challenge we face when trying to slice through data and noise efficiently enough to uncover truly actionable, predictive insights while there is still time to respond. Today’s personalization requires increasing consumer segmentation to an almost individual level, and continually reshaping that process with the latest data. This helps ensure the right mix of dollars is spent keeping your best customers loyal while identifying the highest value prospects before they are swayed by a competitor. 

Although big data and machine learning have matured, I would still categorize them as teenagers for most industries and use cases. We are not yet at the point where we can just give them the reins to make all of the decisions on their own, but it is obvious that when it comes to data processing and analysis, computers have speed and youth on their side. There are many processes computers can easily handle with minimal “adult supervision," but the guidance of experienced marketers and technology staff are still important to increase trust in what computers produce. 

Experiments, results, and the associated data need to be provided so computer learning can occur and the learn-do-loop can operate quickly. With the right sets of integrated data and consumer recognition across devices, computers can highlight underlying factors and make predictions at the speed required to keep pace with the market. Marketers can then use that information to set up their next campaign to either provide the best return or the best opportunity for the next round of learning.

Tips to maximize learning across data, machines and people in 2017
  • Tip 1: Invest in data and/or partnerships that allow for detailed, consumer-level behavior analysis, whether for campaign measurement, meeting the needs of existing customers, or finding the best new prospects.
  • Tip 2: Use marketing experience to guide computer learning by posing questions or testing hypotheses that can be evaluated quickly and yield valuable answers (even if the answer is that we need a better question).  
  • Tip 3: Use maturing, self-service visualization and statistical tools to expose data and findings to a large audience and reap the benefits of collective, cross-functional insights. It takes a village to understand the consumer and intelligently plan a campaign.

The increase in integrated, timely data along with the cooperation of experienced people and machines will make 2017 a year full of increased data trust and usage, resulting in more intelligent and timely marketing decisions.