Personalization boosts customer experiences II

Addressing customers personally requires data. In the first part, we saw how important this information and its analysis are. However, the success of implementation also depends on a high-performance CMS that integrates seamlessly into existing systems.

Measurable goals and KPIs are the prerequisite for personalization. It is important to analyze user behavior on a platform in as much detail and as productively as possible. This includes data such as the bounce rate, duration of the session, click behavior, movement on individual pages of the domain, and generation of leads to other platforms. It is this data that makes comparisons possible: Did the user receive information that was relevant to them? Did they enter the site via the predefined click path and then make the desired conversions? The constant optimization of marketing automation is only possible if the focus is not just on relevance but on the measurability of the data.

When deciding which customer information to use for personalization (tracking data or your own CRM data), it essentially comes down to the size of the operation and the business impact. A combination of both sources is ideal for gaining interesting new insights – for example through a comparison that becomes the basis for an informative and practically applicable scoring model. An example:

As a result, the customer shows an affinity of 0.7 for the beauty segment. That means that it is very likely that she will see content from the cosmetic, beauty, and wellness segments when she visits the site – or from segments in which she also shows an interest based on the available data and her online behavior.

For perfect marketing automation, personalization needs to be made as simple, efficient, and elegant as possible. Performant access to the content requires a dynamic exchange of content. Technically, this can be realized with two alternatives: 

Personalized content on

JavaScript or React solutions are ideal for the second method. The client-side solution is much faster, for one thing, and takes the pressure off the server. The Magnolia Content Management System implements both variants, which means that standard and headless rendering can be easily combined. In practice, standard content is provided first, then the content is replaced by personalized elements.

In the ideal scenario, the user doesn’t notice this process in the background. To them, everything feels consistent, fluid, and easy. The trick here is a combination of skillful placement and seamless transitions.

Setting up product suites is becoming more and more complex. There are a variety of tools for customizing and often only experts know how to use them. In the long term, it is cheaper for a company to store segments and use them throughout the entire ecosystem (websites, apps, etc.). This creates clarity and makes nuanced use on different channels easier.

A dedicated institution manages the business logic behind this. This logic determines the preferences of the content for the relevant user. Making this logic cleanly available for third-party systems requires precise encapsulation, which can be achieved via an API. Personalized content from the content hub is displayed either headlessly via REST API (example: Android App) or with a streamlined React app (for content on different websites).

Although an individual solution saves on annual licence fees, the initial costs are greater than for a solution based on a finished product. Often, the most effective option is a mix of existing tools and a self-build.

Content is created on the basis of customer segments and stored centrally. An effective CMS for gathering, managing, and distributing the data is Magnolia. It is unbeatable in terms of the setup work involved and does away with manual entry of the content in the individual teaser in the CMS. Automation also has an impact on the editorial side of things. That’s why manual adjustments to the logic should only be made with extreme care in order to strengthen faith in the personalization engine in the long term. At the beginning especially, it is important to ensure that the results are meaningful.

It’s best to get started in personalization by consciously limiting yourself to the essentials. A simple MVP (Minimum Viable Product) is suitable for this. It allows penetration across all systems involved. Also minimize the complexity of existing systems and pay particular attention to the measurability of the results. It will quickly become apparent where personalization is proving especially successful. These successes can then be transferred step by step to the different touch points of the customer journey.