This is part 2 of 2 in a series on the power of email and personalization co-written by Boomtrain and PostUp. Read Part 1 here.

It can be a scary world out there for modern marketers. We’re constantly dealing with new influxes of data and competing in an increasingly noisy online space. Add a bevy of new tech tools and analytics to the mix, and it’s easy to see why we’re still wrapping our heads around how to use that data to take action.

Hence, two age-old marketing challenges persevere:

  • How can I better understand my users?
  • How can I better communicate with them to keep them engaged with my brand?

Enter personalization: a perhaps over-used word that represents various marketing tactics, from auto-populating first names in emails, to generating dynamic on-site experiences. For us modern-day, data-driven marketers, it’s more about predictive intelligence — a way to process large amounts of data to automatically send the right message to the right person at the right time, in real-time.

The Personalized Personalization Problem

Personalization algorithms come in various shapes, sizes, and segments. Even the most advanced solutions that take into account individual behaviors maintain one inherent problem: It’s almost impossible to optimize for your unique business needs and goals. Up until now, and perhaps ironically, personalization technologies have been one-size-fits-all. In other words, you’re not allowed to personalize your personalization.

Facebook is a great example of this conundrum. Along with Amazon, Netflix and more, the social network has made great strides in advancing machine learning for marketers. It’s a great channel for several marketing purposes, like retargeting segments and new user acquisition. In many cases however, Facebook is actually a Big Black Box (which should come as no surprise to most publishers). Users go in, engagement comes out. Great data like the User Interest Graph gets trapped, and it’s impossible to optimize an algorithm you don’t own.

Let’s take a look at what this can create:

  • Silo-ed data: User behavior and content data remains inaccessible to Page owners.
  • Lack of control: If Facebook changes their (pretty opaque) algorithms, you could be screwed.
  • Lack of ownership: Facebook owns all of the engagement data generated by your content and your followers.
  • Limited view of users: With Facebook, you’re effectively ignoring all other data resources (on-site, email engagement, and more).

Again, Facebook is fantastic for many engagement goals and we don’t foresee the network going anywhere anytime soon; but it’s not the full story. If your goals include re-engagement or retention, you just need to dig deeper into your data to fully understand your users over time, over channels, and many other data points.

Take Action on your Data with Predictive Personalization

It’s time to take back control! Cheesy, I know. But you can now automatically send the right message to the right person at the right time, in real time, and optimized for your unique business goals. Furthermore, you can deploy across multiple channels — email, on-site, in-app — for a true cross-channel retention strategy. It’s all possible thanks to your own first-party data and one simple but powerful predictive intelligence tool.

You may be feeling overwhelmed by this new revelation, and a little fearful of all the data involved — rightfully so! So let’s take a peek under the hood of Boomtrain’s predictive intelligence layer to understand what goes into delivering 1:1 personalized content to every individual:

  • Understanding your users: Not unlike Facebook, predictive personalization begins by developing an interest graph around each person to comprehensively understand users on an individual level. Behavior is the #1 indicator of intent, and to understand this, Boomtrain looks at the last article someone read, the time they clicked on an email, etc. Furthermore, your users behave differently across channels and under many different contexts. A user’s graph begins building from Day 1, even if they are still anonymous, and it links to your email communications once they subscribe. You get a 360-degree view of each user, and that data is accessible at any time. Neat, right?!
  • Understanding your content: Machines not only help us better understand humans en masse, but also content at scale. Layer in semantic context, along with collaborative filtering, and algorithms instantly help marketers make faster decisions and predict what a user might like to read next. If you’re using tags like Open Graph to connect your content to the broader user interest graph on social, you can use those same tags to connect your content to the audience journey across all your channels. In short, it’s all about fueling discovery.
  • Understanding your business rules: Herein lies taking back control of the user experience. Customize your own algorithm with your personal goals based on what brings value to you and your business. To clarify, this goes beyond simple rules-based or triggered communications; it’s a means of informing the algorithms in the beginning of implementation and building experiences that self-optimize over time. For example, are you a news site? Create a rule that refuses to surface content older than 24 hours in your email newsletter to keep users returning to site and boosting ad revenue. Is your content geo-centric? Create business rules that deliver relevant content to the right region – great for the travel industry, deal sites, or any publisher looking to keep people up to date on the latest happenings in their area.
  • Understanding 1:1 automation: When you understand what goes into 1:1 recommendations, the next step is using automation to take action on your data. To compete in today’s market, personalization should be the core of your strategy across channels. It’s the anti-Black Box strategy, bringing fantastic engagement via email marketing, on-site recommendations, mobile re-engagement and more.
  • BONUS: deep content insights: While ingesting all your user and content data, Boomtrain surfaces deep insights about the value of your content across users based on virality, consistency, and evergreen (long-term) value. For editors, this means you can build your content strategy, or even social activity, around what benefits readers most. Directly turn all those influxes of data into more return frequency to your site, more intense sessions (more page views per visit) and more viral sessions (more social shares).

See?! That wasn’t so bad, was it?

Main Take-away:

Predictive Intelligence means proactively putting your own data into action to the benefit of your business and your own users, just like the Facebooks and Netflixes of the world! The result? Hyper-engaging 1:1 personalized experiences and a tribe of loyal users constantly engaging and sharing your content.


Share This