Ever since the first browser appeared, we believed that the internet would deliver a digital experience smart enough to know our interests and wants.
We are only seeing the beginnings of personalization today.
Mobile and social have provided a huge spark. And the very individual nature of our mobile devices has unlocked an even richer path to understanding how to deliver a personal experience across commerce, news, entertainment and communications.
But the category of solutions generally referred to as “personalization” is confusing and crowded. The latest Gartner report listed 36 competing vendors in this category.
When we dig below the surface, we find that what is broadly called “personalization” actually falls into three distinct categories of solutions, each with specific use cases:
Optimization: Used for Decisions that Affect the Whole Target Audience
Optimization, otherwise known as A/B testing, determines the best possible presentation of something to a broad set of users. Typically using a challenger/incumbent approach, optimization rotates a series of candidates in front of an audience over a period of time until a winner is declared.
An example would be trying to determine which copy pulls the most clicks for an advertisement. Marketers can typically run these tests over and over with little complex coding or back-end work required.
The key is that optimization doesn’t seek the best presentation for each individual user, but rather, for the audience as a whole. Optimization shines when making decisions about website or app layout and design, shopping cart conversion workflow and pathing, and advertising design and layout.
Recommendation: Used for Decisions Based on Whole Audience Behaviors
A close cousin to optimization, recommendation solutions apply the same approach to individual pieces of content. Most commonly built around a technique known as collaborative filtering, most recommendation engines compare similar sets of audiences in terms of popular or most likely to be clicked content. E-commerce is the most common application.
Recommendations algorithms build profiles for each item in terms of the most closely related or often-purchased items when compared to other items. For example, the algorithms may find that a large percentage of shoppers who buy a certain set of shoes also frequently buy a particular handbag and therefore will recommend those items together as part of checkout workflow.
In the publishing world, content networks will present a series of story links that drive the most clicks given the topic or specific article being read.
As with optimization solutions, recommendation solutions form suggestions based on behaviors across a large group as opposed to tailoring results for the individual user.
Personalization: Used for Tailoring Content to Individuals
The big leap from recommendation solutions to personalization solutions is the collection of data from the browsing behavior of each individual user.
This is where we at Cxense focus our efforts: collecting data and making it actionable for real-time 1:1 personalization.
Generally collected through the use of tracking scripts, personalization solutions build a comprehensive profile of each user over a period of time, and in some cases, create detailed profiles of all the content a publisher is producing, as well.
These content profiles serve as additional inputs in user profiles. For instance, understanding the “aboutness” of a set of content can inform the profile of a particular user who likes to consume content about George Clooney, gold prices or Syria. The profile also includes information about the time of day, browser, geography and device of an individual user. Certain approaches can also unify the browsing profile of an individual across multiple devices.
All of these become additional “signals” that marketers can use to create a personal “conversation” with the user, which might include content, advertisements or offers.
Marketers can marry this extensive data profile to customer relationship management (CRM), email and advertising platforms to further enhance other use cases. Personalization requires a tremendous level of sophistication on the backend to support tracking, assembling and maintaining individualized profiles in real time.
These solutions get increasingly powerful and complex depending on the approach marketers take. However, the table-stakes expectation is that digital experiences get smarter over time and deliver ever more relevant content, ads and commerce.
The key to meeting and exceeding those expectations is to start with a specific goal in mind and apply the right approach every step of the way.
This article was originally published on CMSWire.com in May 2016. Everything in it is still very true, so we thought we'd share it with you here.