Imagine you’ve just walked into the corner store to find something quick for dinner. The shopkeeper asks “How may I help you?” and you explain you are looking for an easy but somewhat healthy dinner. The shopkeeper points out that she just started carrying both Lean Cuisines and Amy’s Kitchen meals in the freezer section. Bingo. She assessed your needs in the moment and served up exactly what you were looking for.
After buying Lean Cuisines three nights in a row, you stop in again. The shopkeeper asks, “Dinner again? How about a salad?” This time you respond that you have a few friends coming over to watch the game, and you actually need some snack food. She points out her favorite dip in the refrigerator section and then directs you to the chip aisle.
"Customers will continue to demand relevant, contextual communications, and markets are already beginning to embrace AI and machine learning"
Marketers often have trouble making this real-time adjustment. Once we have bucketed someone into a “persona,” we rarely consider that there is an actual person on the other side of our digital interactions, at varying stages of their journey to making a purchase, with lots of varied motivations for doing so. Rather than asking “how may I help you?” and reacting contextually, in the moment (like the shopkeeper), many of us profile our prospects based on static data and assume to know what they want or need. We then push content to these individuals through marketing automation and other channels that we believe will meet these assumed needs. Sometimes we get it right, sometimes we don’t. When we don’t, we risk alienating our buyers.
Thankfully, technology enables us to build conversations with buyers that are more like our dynamic exchange with the shopkeeper. We can achieve these types of interactions with a customer data platform (CDP)—a system that employs a unified, persistent, single platform for all customer behavioral, profile and other data, from any internal or external source. A CDP uses this data to generate signals—units of intelligence— that help us to understand our buyers’ context, across multiple interactions, over a span of time. Signals also help us predict, in the moment, how we may best help each and every individual we interact with and then executes on those actions.
When we use this CDP technology to drive our marketing automation, we can converse more humanly at scale by constantly watching and reacting to our buyers’ cues, just like we would if we were interacting face to face.
The following are three specific ways in which a CDP helps us to have these more human conversations.
1. A CDP Facilitates a Non-Linear Buyer’s Journey
Marketing automation platforms (MAPs) enable us to create trigger-based journeys aligned with the traditional notion that buyers move in a straightforward, consistent progression from one stage to the next—there is no skipping around. New research on buyer behavior across multiple disciplines (e.g., economics, psychology) and sponsors (e.g., Google, Forrester) tells us three important things:
• Buyer movement through their journeys is often sporadic.
• Buyers are opportunistic during their journeys. They are often highly self-directed and are going to shop their way and don’t care whether suppliers envision a logical sequence of steps for them to follow.
• Buyers are often emotional rather than linear and methodical.
In other words, we simply cannot sell to buyers according to our own notion of how the journey should go; we can only facilitate the buyer through a self-directed journey. We need to be able meet the buyer at the correct stage of this sporadic journey, at any given interaction, with the right conversation. To do so, we must be able to look at the buyer’s behavior more holistically, across channels and through time. This cannot be achieved by housing the customer data in the MAP, which only houses a fraction of our total customer data. To successfully navigate a buyer’s non-linear journey, we must collect and analyze the whole of our customer data in a CDP. Centralizing data enables the ability to leverage it like an asset that feeds intelligence into all your enterprise applications, not just MAP alone.
2. Artificial Intelligence Uncovers Patterns, Improves Conversations
Speaking of those traditional, linear buyer’s journeys, they’re often built based on our own assumptions of what content and messaging should come next in the buyer’s sequence—and we need to manually build those journeys in the MAP. If someone abandons a cart, we send them an email to come back and buy. If they come back and buy, we send them another email with related products. If they don’t come back and buy, maybe we send them a discount offer.
But are these really the correct next steps in the conversation for any given buyer? As humans, there is just no way for us to uncover all of the patterns in our data that tell us our next best marketing action based on a buyer’s behavior. And even if we could know what our next best action is in every given situation, manually building all these journeys in the MAP would be unwieldy. Humans alone cannot manage this complexity. However, a CDP with integrated machine learning and artificial intelligence (AI) leverages signals to reveal patterns and make accurate predictions of future activity. In other words, it takes the place of a human when marketing complexity outgrows human capacity.
Rather than traditional rules-based marketing automation, AI applies behavioral data to tailor messaging and orchestrate contextual communication across channels without manual configuration. For example, AI may detect that I purchase clothing only two to three times per year. I am particularly susceptible to an offer for special help from a sales consultant for a custom fitting event for private clients. The platform can then send me that offer.
These AI-driven next best actions are what should ultimately inform our marketing automation for contextual, human conversations with our buyers. And more contextual conversations will increase our marketing performance.
3. Machine Learning Elevates and Extends Model Training
AI within the CDP tell us what is likely to work next for any given customer, but it will never be 100 percent accurate, and it can always be improved. That’s where machine learning comes in. Perhaps I received that offer for a custom fitting, and I signed up. The machine tests its theory, sees that it works, and then uses that learning to optimize the message the next time it sees someone like me. Conversely, if I didn’t sign up for the custom fitting, the platform learns from that as well, and continues to evolve its messaging for me—and others like me. In other words, the machine trains its own models and algorithms to become more precise over time, improving our results.
With a CDP watching every set of conversations, that works and doesn’t work in those conversations, applying the most appropriate content, and continually optimizing its own models, we can get as close as possible to a personalized dialog with each individual.
With a customer data platform in place, AI driving next best actions and machine learning constantly improving communications, we can all be more like the shopkeeper. We can digitally ask “how may I help you?” and have more human conversations with each and every individual.
This is the future of marketing automation. Customers will continue to demand relevant, contextual communications, and markets are already beginning to embrace AI and machine learning. We all need to embrace these trends and begin designing our roadmaps that outline how we will centralize data within a CDP for use in marketing—and across the organization. It won’t happen overnight—it will be something we build then continue to build on—but the time to start is now.