The revealing truth of real-time personalization

A person using a smartphone with illustrations.

Real-time personalization is the ability to customize a message to an individual user that is relevant and reaches them at the optimal time on the optimal channel. You can think of it as an evolution of the “right customer/right time” goal of targeted marketing or as one-to-one marketing at the tactical level. Even though you may be doing real-time personalization right now, you’re probably not doing it as well as you think.

If you want to know what effective real-time personalization looks like, look to Google and Netflix. They know which users are engaged, what they’re engaged with and serve up content and product recommendations based on that data to keep their customers engaged and move each of them along a personal customer journey. What does bad real-time personalization look like? It looks like a personalized pop-up ad that feels janky or an email that starts off with “Hello Ash” and loses me in the next sentence with an irrelevant offer.

You say hello and I say goodbye

Real-time personalization is only as good as the data you have to support it. When you have the right data used in the right way, your marketing feels highly relevant. When you don’t, it can feel spammy. The right data—and, by extension, the right targeting—will tell you not only what messaging channel is best for the user but also what time for that message is best. If I’ve just visited a brand’s website, do I want to see an email from them six hours later? How soon is too soon to show me retargeting ads on my social media channels? If I get both emails and retargeting ads, will that boost my engagement or drive me away? Answering questions like these are key to getting real-time personalization right.

Marketers worry about over-messaging their customers and with good reason. If I get a pop-up ad that feels intrusive, I might leave the site. If I get too many messages from a brand, I may be prompted to unsubscribe from their marketing entirely. Real-time marketing efforts that fail have real consequences.

CDPs provide real personalization

Using a data management platform (DMP) to build audiences, compared to using a customer data platform (CDP), is like the difference between guessing and scoring. With a DMP, a marketer might use rule-based heuristics to create an audience of users who haven’t been on their website in the last seven days. They would then label this audience “ripe for re-engagement.” 

In a CDP like Lytics, however, marketers have access to data science driven scores such as customer’s engagement level based on cross-channel behavioral data and predictive analytics. For example, they might find that a customer with a momentum score between 70 and 90 (out of 100)  is the best audience for a given offer. They’re showing frequent and high interest in a given product or service, and therefore are likely to act in response to the offer. In our experience, audiences based on data science perform a lot better than audiences based on some arbitrary rules that may or may not be relevant.

Building better content recommendations

Another area where CDPs can have a big impact is with real-time content recommendations. This can be scary for marketers in the beginning, because it requires some faith. Marketers can’t see each recommendation being made to their customers, but they can preview those recommendations with a robust CDP like Lytics and get a good idea of whether their recommendation engine is really working. The difference between a relevant and irrelevant recommendation could be the difference between a site visitor staying engaged and bouncing.

Earning ROI with your CDP

Content recommendation tends to be a more mature use case for CDPs. Generally, we recommend starting off with something that can deliver ROI quickly, like thinning out your email lists to remove users who are either already highly engaged already or unlikely to engage by email. Given the fact that most email providers charge per email or per user, creating a leaner list based on behavioral and engagement scoring can net marketers the same results and save them a lot of money.

Once marketers see the value of using behavioral data plus data science to drive fundamental marketing decisions like email lists, they’re more confident to try things like content recommendations and creating customer journeys. And that’s where real-time personalization takes on real importance, as it becomes more  than a tool to improve channel-based marketing campaigns, but the foundation for great customer experiences across all channels.

Want to see how you can earn ROI with a CDP like Lytics? Try Lytics for 30 days!