Blog Entry

B2C Personalization Blog Series Part 3: Knowing Your Customers So Well You Can Predict Their Needs

In part 1 of this series, we looked at the importance of responsiveness and reacting to customer behaviors. In part 2, we looked at the rules for relationship marketing and how to further level up your communications so that your customers know you know (and love) them.

So you may be asking: What’s next? The next and final step is creating a full 360-degree customer profile that uses data from across your organization and then using it to super-charge your interactions with your customers and exceed their ever-growing expectations. This is the gold standard most organizations are striving to achieve. Pursuing this goal will take you to the cutting edge of relationship marketing. Those who can master it see big returns and have customers who love them and are oftentimes big advocates for their brand. 

Let’s look at some key things this level of personalization can do for you and how it uplevels 1:1 relationships to almost something out of science fiction:

Real-Time Connected Data Is King

To create a frictionless experience with a consumer, your brand should be able to connect all touchpoints you may have with a consumer and use them to build a robust customer profile. This data can come from lots of sources like your ESP, web traffic, call center, chat interactions, app activity, in-store purchase data, etc. This data can also be combined with things like purchased demographic and psychographic data to more fully understand who that customer is as an individual. 

Once this data is connected, the systems that are capturing it and using it need to be able to sync in real-time. This can be achieved through the use of APIs. Once you’ve harnessed all of this data, you can create robust personas that can be used to provide customer interactions that show you really know the customer on a 1:1 basis. An example of brands that typically do this well are airlines—they know you booked a ticket on their website, so it shows up in-app and tailors your in-app experience based on your booking, and when calling the call center agents have access to all the same data the consumer has seen across channels. This creates a frictionless experience for the passenger, who can travel at ease knowing that no matter where they look or who they ask that their gate information is up to date, their bag has been loaded on the plane, and their seat assignment hasn’t changed. Once you’ve got rich interconnected data, there are a lot of advanced things you can start to do.

Continually Evolve Your Decision-Making with AI

With a dynamic and intelligent 360-degree customer profile in place, you can start using machine learning and artificial intelligence to look for patterns and trends. As an example, machine learning may identify that a person who opens their email daily, has called the call center, and visited a store within the last five days is a good target for an up-sale promotion. 

Computers are able to analyze millions and millions of possible relationships and identify patterns that a human may not ever be able to detect. As these patterns and trends are identified, you can begin determining which are important and then take the important ones and start scoring consumers to determine logical groupings. The groupings can then be used to help prioritize your marketing. 

Think of this as a type of RFM (recency-frequency-monetary) model on steroids—it is a much more advanced predictive model that can take thousands of inputs. Machine learning makes it possible for these models to continually optimize themselves in real time based on the results that are being fed back into them.

This sort of modeling can help identify consumer needs as quickly as they come up—or even before, if you’re using predictive modeling. Consumers love brands that can pull this off well since they make their life effortless and always seem to know exactly what they want or when they may need a bit more TLC.

Predicting What, How, and When to Send

Having predictive models built into your marketing programs allows you to uplevel how you interact with a customer. What might that look like? For example, a brand could create a marketing program designed to help close sales—using data and model scores, the program could then determine things like:

  • What to send. What is the right product message and mix of content based on what the brand knows about the consumer? The model may find a correlation between people who buy product X in a specific category and people who buy product Y in a totally different category. Including both those products in the same email might help close a sale on two products instead of just one.

  • How to send. Each consumer has their own unique marketing consumption habits. Using what you know about a consumer, you can determine what channel makes the most sense. One consumer might like to read emails from your brand, while another prefers SMS or push messages, while a third might like a call from a sales associate. Using data, you can route the message to the right channel for every customer.

  • When to send. By collecting interaction data, we begin to learn what day of the week and time of day an individual consumer typically likes to engage with a brand. Send time optimization makes it possible to use that data and send an email, SMS, or app push notification at the time that will likely have the highest engagement.

Start Small, Test to Learn, Then Grow

Having a massive amount of interconnected data, then using it for high levels of personalization, can be quite complex, so it is important to start small and test, then grow. See what is working—and more importantly what isn’t working—then take that data to adapt the program and evolve it as you move onto the next growth phase. Taking this phased approach is key in ensuring continuous learning via testing is taking place.

 It is also critically important to be looking at results at a macro and micro level both within and across channels. Rolled up macro-level results help give you an idea of how to steer your big picture strategy. For example, you may notice that there is a drop off in engagement after a defined timeframe that is much sooner than you initially anticipated, which would mean you should tighten up the number of touchpoints. 

You’ll also see what is working within each individual touchpoint and channel and will want to fine tune on those more micro-level trends, too. For example, a specific segment may not be clicking a call to action where another segment may be very engaged with the same CTA, indicating you need to tweak your messaging or CTA for the under-performing segment.  

Regularly reviewing performance at both the macro and micro level will ensure your program is as optimized and fine-tuned as possible. This also helps build a culture of continuous testing, where results are being used to inform future tests on a regular basis, which allows for maximum effectiveness and faster optimization.


Read Part 1 of this series to see How to Take the First Step Toward One-to-One Marketing.

Read Part 2 of this series to see How Creating a Single Brand Experience Can Grow Relationships