The Data of Contextual Intelligence
In the 1960s, computers were introduced to marketing departments. Essentially, their purpose was to pull in all sorts of data, “normalize” it in relational databases and make it available for query. This enabled marketers to begin to understand the science behind segmentation. Over time, whole companies like Experian, Axciom and Nielsen emerged becoming data aggregators. Today they provide this data to brands as a service, essentially appending the data they already collect about people, and making it available for business purposes.
With the proliferation of the Internet era, data became available to everyone. All the vast public databases from the census bureau, Internal Revenue Service and a dizzying array of federal, state and local databases became available to any company that could upload the data. Marketers now had access to much of the data that previously was only available via the Experians of the world.
It wasn’t just marketers and database pros who looked for data. Everyone with a browser and an internet connection became data junkies themselves. With the advent of Google, online search became a verb, not just software. As everyone started searching for everything, Google figured out they could track those searches and understand what people found relevant. As the dominant search engine with more than 2 Trillion searches Google became one of the largest sources of data anywhere. Google sells that data as a part of its AdWords search marketing business, generating almost $50 Billion a year in data enabled revenue. All to help businesses figure out what people actually want.
It took less than a decade for the next era to appear, the social network. With the appearance of Facebook, Twitter and Instagram, billions of people worldwide began voluntarily sharing some of their most personal data on a daily basis. Facebook uses 98 personal data points to target the ads it sells to virtually every major brand on the planet. Facebook not only captures data like age, gender, location…, but it also captures the language, the images and even the jokes you tell. If your mother’s birthday is coming up, you will see ads from local florists. If you joked about 4-wheeling in a jeep, expect to get something from Chrysler, promoting local Jeep dealers. The problem is, the experience does not always produce a state of flow.
The context of Contextual Intelligence
With the advent of the age of experience and its inherent “in the moment” zeitgeist, customer experiences must be highly personalized, targeting and address ever more and more complex segments, approaching if not yet achieving segments of one. But even this isn’t good enough. A brand may know its customers, they may know about past behavior, but unless they understand and can react to the context of the moment, it will not be relevant.
Even Amazon can get the context wrong. How many of us bought books to read to our toddlers from Amazon? And as the years advanced, we bought doll houses and nerf guns, video games and concert tickets, but we still got ads for books for toddlers. Amazon had all our purchase histories, but missed out on the fact our children got older every year, they ignored context.
Are you the same person when you buy morning coffee and a breakfast sandwich on the way into the office, as you are on vacation at the beach or in the mountains? Do you order the same thing when you’re with a group of business colleagues as you do with your spouse? Will you respond to an offer for something you don’t like, just because you are a member of a loyalty program? The answer to all these questions is probably no. Does that make you a less loyal customer or reduce the opportunity to increase your lifetime customer value score? Also no. The key is for the brand to understand the context of engagement and respond appropriately in a seamless and perfectly natural way.
The mechanics of Contextual Intelligence
Compiling customer data is nothing new, but the where, the how and the what is quickly evolving. Over one hundred years ago, efficiency experts were hired to do time and motion studies on factory floors to understand how workers worked. They introduced new processes and in some cases, new technologies to speed up production. It wasn’t always successful, but they laid the foundation for collecting real-time data about what was actually going on. There was no theory to it, it was simply empirical data. Or as we used to call them, facts.
Fast forward to today. With the advent of visual recognition technology, it is possible to observe and learn from the behavior of people without violating their privacy, stealing their identities or in any other way compromising their anonymity. It is possible to capture how many people are in a store, restaurant or any venue and then compare their physical and emotive behavior to the POS data containing sales, tickets and products within the same time frame. Multiply by the number of locations, mash up all the data into massive database and you have the makings for an in-store profile which can predict individual behavior, even offline.
These insights can be used to make an offer more relevant and improve operational efficiency. Stores can spot a long line at the counter and promote a mobile solution to skip the line. Retailers can respond to tracking the volume of sales during a particular time slot and day part an offer by offering a product “Every Tuesday morning from 10:00 am to 10:30a” to level the load. When a coffee shop senses people sitting in the store with their laptop open for certain length of time in the middle of the afternoon, they can automatically generate an in-store promo for a refill and then add a food item as an attach.