No one can deny that reacting to customers when they need it most is crucial to building great relationships. It is important to understand customer interactions across different channels and integrate them in real time so you can respond with most relevant offer quickly. The ideal offer will not only be time sensitive, but will reflect the what is known about the audience from attitudinal, demographic, and other behavioral data.
Marketing organizations do their best to respond to the customer in real time, but it’s typically based on outdated, historical knowledge, or they’ll respond late as they often lack the integration of true real-time insight. As a consumer, I have many examples where I interacted with brands across multiple devices and the brand was not able to keep up with current information. Recently I was in the market to purchase a car as I had totaled the car I was driving (long story!). I immediately started searching online inventories near me for the model and year I was looking for. I received offers on my next visit purely based on my search criteria, but not from dealers who are in my area. Since I needed a vehicle fast, I ended up reaching out to my close network and made a purchase with a week of starting my online research. Right after I made my purchase, I started receiving calls and emails from many nearby dealers with available inventory and best deals. Sadly, all of these offers were irrelevant as I was no longer in the market for a vehicle. Clearly, my online search activities were not actioned up fast enough by local dealers with relevant offers at the right time and in the right place. The other very annoying part was that they continued to harass me with offers for a long period of time after I had already made my purchase!
Marketers can avoid such situations by getting up-to-date information, making sense of that information, and responding with the most relevant message and offer at the most relevant time. While the need sounds simple, the necessary data integration is challenging and quite costly, and it is collected across so many channels and devices with so many different formats that it requires robust data hygiene and matching processes to build the consolidated view of the customer. And let’s face it, doing this in real time is no easy feat!I suggest an intermediate path to data integration that enables you to address the complexities and respond to your customers in a timely manner with thorough offline knowledge of the customer.
Key points for this approach:
- Use your existing CRM Mart data relationships.
- Maintain natural keys and surrogate key relationships for each source.
- Create a new set of real-time tables separate from your existing CRM Mart tables.
- Capture real-time information through web services or other key API’s and then update the real-time tables by leveraging existing relationships and natural keys
- Create best view and map applications to leverage real-time data at the logical block levels while other information still reflects data from CRM Mart.
To ensure proper Data Integrity, let the real-time data also process through regular batch cycles to combine the data from other sources and capture them in the CRM Mart. This process establishes the keys for each entity to create relationships between objects. While this work can be extensive up front, having it in place when an event occurs pays off significantly, enabling relevant responses to the customer. It shortens the normal integration process in three steps: leverage natural keys available as part of incoming real time data; fetch the related offline keys and information from the offline knowledge base; and combine with real-time data to populate into new set of tables. Finally, this enables marketers to create views with the best information between real-time tables and offline data tables and map to the offer generation end applications. This approach uses the existing data relationships and refreshes the relevant data from real time. When unknown individual information is received, you still don’t have to worry about creating formal keys, but assign temporary keys to establish relationships based on incoming data per existing data structures to enable the most relevant data views and see customers’ information as soon as possible.
