We talk a lot about Customer DNA at NGDATA; it’s a critical part of what we do here. When we say Customer DNA, we’re referring to a collection of thousands of predefined metrics based on individual customers’ profiles, behaviors and preferences. These are calculated in real-time and organized for each individual customer. The metrics are the key component to improve customer service and experience, and even increase up-sell opportunities, which increases customer lifetime value.
Customer DNA, on the surface, appears to be a simple, common-sense idea. But there’s a lot that goes on behind the scenes that organizations really need to focus on if they are going to become a customer-centric organization with the help of Customer DNA.
We’re able to collect customer data from more channels than ever before, which in theory would make an organization smarter about its customers. However, slow feedback loops between the different customer touch points and the customer data warehouse prevents organizations from building an always-up-to-date customer profile. For example, a customer who runs a car loan simulation at 9:30 am and later that day runs a second car loan simulation for a slightly larger amount might still be contacted that evening by the call center with an up-sell credit card offer based on a monthly updated credit card upsell score. Since the company’s web interaction data upload is only scheduled for once a day at midnight, and the company only scores the customers once a month, the customer’s upsell propensity moves slowly too. Result: the company reaches out with a less relevant credit card upsell offer while the customer clearly signals an interest in a car loan. By the time the customer’s propensity to get a car loan is updated, the customer’s interest might have declined or he might have secured a car loan from a competitor.
Customer DNA done right means continuously updating the customer’s profile in real-time at a granular level and in context. This implies four major shifts:
- A shift from collecting data, integrating data and updating model-based scores at a low frequency to a much higher customer-triggered update frequency: as the interaction happens, the customer’s profile is updated; as the customer’s predictor’s value change, the customer’s model-based score is updated.
- From a slow-moving customer profile based on slow-moving data only, such as socio-demo and longer-term behavioral summaries (e.g. number of car simulations last month), to a fast moving customer profile enriched with fast-moving metrics (e.g. number of car simulations last hour). Lily excels at summarizing the customer’s multi-channel behavior at different granularity in real-time using both logical (per minute, hour, day, week, month, etc.) and rolling time windows (per 60 seconds, 60 minutes, 24 hours, etc.).
- From generic interaction logging to enriching a customer’s interaction with immediate situational context including temporal context (time, part of day, time of year, season, etc.), location (absolute and relative location, work/home), channel and device as well as customer-relevant context (e.g. the number of products in portfolio at the time of the new product purchase interaction). Lily summarizes these augmented interactions in real-time at varying levels of granularity.
- Fourthly, a shift from logging the interaction from an operational point of view (e.g. customer completes online information request – how) to logging the interaction from a customer’s point of view (customer requests product information – why) by adding the customer’s goal of the interaction.
Customer DNA done right also means closely monitoring the customer’s evolution. A fast (acceleration) increase (trend) in the number of product X web page consultations signals a fast increasing customer’s interest in product X leaving little time to respond. Understanding how the customer evolved to the current customer DNA through the metrics’ trend and acceleration allows making the right offer at the right time.
Finally, customer DNA done right also means learning the customer’s true DNA from customer feedback. Lily learns about the customer’s true channel and marketing preferences through a continuous real-time marketing action feedback loop allowing making the right offer with the right format and design elements through the right channel.
Empowered with customer DNA done right (real-time, continuously updated DNA at a granular level in addition to context), making the right offer to the right customer at the right time through the right channel becomes easy.
What could you do with Customer DNA in your organization?
Anita further discusses this very topic over at PreInvented Wheel’s expert roundup on Customer Facing Data.