Personalization in Episerver Part 1: The current state

On every stop along the way of the Episerver Ascend 2015 conference tour, personalization was one of the key themes. 

We personalize by using what we know about our consumers to engage with them on their terms. Getting rid of the generic boilerplates, in favor of relevant, tailored content and interfaces. 

With today’s available technology, it’s easy to collect a bunch of data about your consumers. Just around the corner, predictive algorithms - artificial intelligence - will determine the most efficient targeting for any given consumer in any given context and touchpoint.

The challenge is putting all that intel to use in a way that puts their interests first - which in turn serves yours. Getting to know your consumers so you can serve them better and more efficiently. 

Mixing some empathy with your automated algorithms. Striving for heartificial intelligence, so to speak.

Why personalize?

Even though your site might be filled with top quality content or products, not all of it is relevant to all your consumers. In fact, some of it might even get in the way of what the consumer really wants to find.

There’s an abundance of providers out there, which means consumers are holding all the cards. Their attention is their currency, and we’re all fighting to get a piece of it – and to retain it. 

Just take a look at these numbers:

• "86% of consumers say that personalization influences what they purchase."
  (Vanson Bourne US survey, 2013)

• "74% of consumers are frustrated by content that has nothing to do with their interests."
  (Harris Interactive US survey, 2013)

Personalization is not only for commerce sites, either. As marketing is shifting more towards customer experiences (CX), all site owners will eventually need to provide engaging, relevant content if they want to sustain a level of brand loyalty.

If you’re just starting out with personalization, expect to have to experiment to get it right. As with any transformation, the key is to plan big, but start small – and thenscale quickly when you find a formula that resonates with your audience.

The current state of personalization

Visitor Groups

Episerver CMS has long had a user-friendly, powerful, built-in personalization engine called Visitor Groups. Based on a set of pre-defined (or custom-built) criteria, we can track visitors as they navigate and interact with our sites, and segment them into a number of different personas. We can then personalize content for them throughout their visit.

This is an example of real-time, rules-based personalization. It takes into account a number of specific factors to determine which segment each visitor falls into. However, there is no actual machine learning involved – visitor segmentation is temporary, and changes as the visitor moves around the site. 

This means they might be bombarded with hit-or-miss content because the rules system is rigid and doesn’t evaluate behavior over a long enough time span to accurately predict intent.

Since the current incarnation of Visitor Groups is session-based*, it’s pretty much limited to personalizing based on what the visitor is doing right now. There’s no session history to pre-segment them from, and no data to predict what they’ll be interested in next. 

(*Note that I am talking about Visitor Groups out-of-the-box. Episerver is tremendously extensible, and there are several options to add cross-session data persistence, build a custom data store, or integrate any of the leading marketing automation platforms for advanced visitor profiling.)

Marketing Automation connectors

The various marketing automation connectors for Episerver are integrated with the Visitor Groups feature, so that extended Visitor Groups criteria become available to marketers and merchandisers from within Episerver’s user interface. 

The connector capability is refered to as the Digital Experience Hub (DXH). Connectors can be used to build lead capture forms that automatically puts customer data in the correct segmentation groups in the marketing automation provider’s database. This data can be retrieved and used to personalize content, forms and email templates from within Episerver. 

Because this data is persisted within the marketing automation provider’s database, it allows for a degree of learning over time – the consumer’s behavior and interactions will affect her lead scoring, which in turn affects her segmentation group, which ultimately affects how we’re personalizing for her. For site owners, it’s a huge benefit to be able to control all this from one location (Episerver), instead of having to login to several systems.

Tech gets you started, empathy gets you ahead

With the built-in and add-on tools currently available in Episerver, getting started with simple rules-based personalization requires little effort. 

But the thing with rules is that they require a personal touch to avoid feeling forced. This means that you have to constantly monitor how your efforts are performing, listen to your consumers, and tweak your strategy accordingly. Personalization is never fire-and-forget. 

The importance of putting the consumer at the center of your strategy was neatly summarized by Forrester analyst Ted Schadler at Ascend 2015:

• "The companies that succeed with personalization are those who love their customers."

In Part 2, we’ll take a look at some upcoming new features, and some concepts for the future of personalization in Episerver - especially how machine learning and behaviour analytics might help site owners.