"Data-Driven Thinking" is a column written by members of the media community and containing fresh ideas on the digital revolution in media.
Today's column is written by Matthew Keylock, senior vice president and global head of data at dunnhumby.
Maybe a consumer in the pet category isn’t a person, but for most products, it is real people who buy stuff and consume or use it. Yet even with all the tools and data out there today, we don’t always seem to recognize this.
Marketers’ approach to segmenting consumers seems to have taken hold in the way we have sub-divided our own discipline. Since the emergence of geodemographic segmentation in the middle of the last century, marketing also seems to have been sliced and diced in various ways.
We have added market research capabilities and aggregated sales data and panels since the ’70s and ’80s. In the late ‘80s and ’90s, we added CRM and loyalty marketing, which seemed to me to be a major split of our marketing world into two parallel universes: a customer part and a brand part, which have co-existed separately since then.
In the late ’90s and the ’00s, we then added a whole new level of division, into shoppers and consumers, by different consumer groups such as ethnic marketing teams and by channel with the emergence of digital marketing teams. All of these have evolved their own sub-industries around them with specialist operators and agencies singularly focused. In the last couple of years, we have since added the new “data-driven” capability of programmatic to our approaches.
Perhaps some of the best proponents of data-driven marketing are an exception here, but generally we are still dealing in proxies. We don’t know the consumer and do not have a clear picture of the actual person we’re dealing with. We still really just know what age or income bracket they are more likely to represent, and which media they might have a slightly greater chance of seeing.
Programmatic may use online behavioral data and also provide a more instantaneous interaction but for the most part uses the online behavioral slice just for retargeting of individual items or abandoned trips. A massive number of the use cases for all our sophisticated marketing tools still use very unsophisticated data. Roughly 56% of content marketing is still driven by demographics, according to a OneSpot survey.
So, while we hear so much about “personalization,” underlying so much of our actions is the same demographics data of the mid-1900s. This is an approach that was designed to allow us to group people together, not distinguish and recognize them as unique individuals.
Move Beyond Segmentation
Marketers need to become students of “the person.” Our approaches need to include learning about the individual, not finding ways to group them into blended averages and proxy descriptors. Segmentation remains a very important part of the tool set, but it should be based on grouping individuals up into segments based on known needs and behaviors.
We also need to bridge the fragmented islands of capability we have created. Part of this is a data exercise. The convergence and alignment of the data is not a simple one, but with the right leadership, it can be done without high expense. There could even be expense saving, and if used correctly, there will certainly be significant new value unlocked.
The harder part is an organizational one. Just creating a consistent data view of people at a point in time, including consumers, shoppers and customers, doesn’t mean the business will use it in the right way. Marketing leaders need to simplify and connect their teams and efforts rather than divide them, and they need to lead the whole business on this – not just their direct teams and suppliers.
This means their leaders of data and insight must provide the same breadth of leadership across the organization, wherever customer knowledge and activation are required. If the teams remain structured in a fragmented way, they will quickly begin to create their own new islands of data again, either in-house or via their own selected agencies and suppliers.
It’s also necessary to expand our view of what is possible with data. There are many limitations in data approaches that we can overcome if we choose to and if we are ready to. We need to open our thinking on data and our approaches to managing it.
For instance, mandating that all data needs to be managed in-house places unnecessary and harmful restrictions on what we can do. IT organizations are rarely brilliant at liberating value from data. While they are generally good at doing things right, their whole requirements-gathering approach and expertise are not always ideal for doing the right things in the first place. There is data from outside the business that may not be possible to manage in-house.
We need to think about a much broader data ecosystem. This includes all in-house data, data from third parties and data from partnerships. We also need to start harvesting data from programmatic activation to gather the online behavioral view and to build a comprehensive view of the marketing consumers are exposed to and when and how they respond.
Understand The Consumer
One as yet underleveraged possibility is consumers actively contributing to the data. We have gotten a lot smarter at understanding people with behavioral data and use of real-time context, such as location or weather. We are even detecting emotions for a subset of consumers. These together might not get to a perfect answer because we do not yet have all the pieces reconciled and cannot yet know someone’s precise feelings and thoughts while they are choosing or buying. Maybe we never will. We can be much better, though. Being much better is not only about doing a better job with the data that is already in-house or out there, it is also about changing the engagement with the person themselves.
We can extend our implicit or explicit contract with consumers to encourage much greater two-way sharing of data. We must then commit to sharing in the value that is created by this new and best view of consumers – one that we can only build together. There are some big companies and some startups already moving in this direction, and with political, consumer and market pressure adding to this, the momentum is gaining. While some suggest that privacy is becoming obsolete, the notions of trust and value are not. People will make choices and choose brands that deliver on both.
Businesses that come to know consumers in richer, more complete ways will have a significant competitive advantage and find better and more numerous opportunities to provide the bridge between people and the products and services they need. This is nothing less than the next phase of the retail model unfolding.
Are you ready for it?