Introduction

Understand why we created the Generic Digital Data Layer (GDDL), an event-driven data layer approach that can be applied generically, regardless of the vendors and platforms being used.

Today's digital analytics challenges

The GDDL framework was developed by the analytics practitioners of Stitchd, a digital analytics boutique active in Belgium and the Netherlands. The initiative was sparked by the challenges around data quality that we encountered at almost every enterprise client we were working with. Think large corporations that maintain multiple platforms and with diffuse responsibilities when it comes to tracking implementations. The framework is developed with as main goal to improve data quality and keep analytics implementations manageable.

We saw the same digital analytics challenges surface in many organizations. Challenges that expressed the need to move towards a new way of working with, and approach of, digital data:

1. A website doesn’t exist of static pages anymore. More and more single page applications are being used, plus modular development frameworks are on the rise (e.g. Angular). This is a problem for traditional web analytics implementations that rely completely on page loads.

2. Dynamic content is used to provide a personalized experience. This implicates different content is being shown to certain groups of users and context becomes much more important. The main challenge is how to add this context in our data analysis and reporting? Again, traditional page-based tracking falls short, as you no longer know by the URL of a page what version of the content was seen by the user.

3. Personalization also brings another challenge: the need for a 360° customer view in real time. Connecting digital data with other business data sources (such as CRM) and sharing it over all touch points is a huge challenge. A challenge that many marketing cloud vendors claim to tackle, but experience learns that it takes more than buying some licenses to make this work. A sound data foundation is key.

4. Another challenge is content visibility. As multi-device behavior has been in play for a while, we need to know how users interact with the content visible on screen. Depending on the screen size this will bring different insights. In order to define the next best action, we need actionable data. We’re repeating ourselves, but also in this case traditional page-based tracking won’t tell you if a certain element of the page was actually seen by the user or not.

5. Digital activities are often not limited to one platform but entail a digital eco-system containing multiple websites and/or applications, with different underlying technologies and tracking methodologies. Which makes it complex to stitch the collected data around a user and recognize a returning visitor from another platform. Main question is how data collection and reporting can be aligned across all those platforms without too much complexity for both the analyst and the developer?

6. Privacy legislation already has a big impact on behavior tracking and this will only increase in the coming years, as the public awareness raises. Customers expect ethical behavior of the companies they interact with and understand that their personal data has value.

The common theme in all these challenges is data quality and governance. To be able to overcome these challenges, corporates need a clear vision of how tracking should be handled. They need to enforce this vision on all stakeholders involved (internal and external). And most importantly, they should take ownership of their digital data.

The flaws of page-based tracking

Taking today’s digital analytics challenges into account, it’s clear that traditional page-based tracking no longer provides sufficient insight. As a solution, many analytics practitioners have turned to event-driven analytics implementations that rely heavily on custom link clicks (Adobe) or event calls (Google). Such implementations involve the risk of customization: values and logic that only make sense to the person who implemented them.

You end up with an analytics implementation that is tied to the knowledge of a specific person, and once that person leaves the company, you realize that you have a bunch of inconceivable data. Specially, when you are working in an environment where multiple platforms have to be maintained and a large number of stakeholders is working on that data.

All too often, this results in the undesirable situation that a company is not capable of evolving beyond the implementation phase. With the war for talent going on in our industry, the churn rate of employees working in digital is relatively high. Every time a new analyst enters an organization, he or she will initiate a complete overhaul of the tracking implementation, bring in new tools, etc. But without a well documented and broadly aligned vision, the same historic issues will resurface after a couple of months. And again, focus will be on audits, new implementations, etc. Contributing to the fact that digital analytics is all too often perceived as a cost center, instead of a profit center.

The Generic Digital Data Layer (GDDL) applies an event-driven approach. As this is necessary to address the challenges mentioned before. However, the GDDL methodology embodies a clear vision that is generically applicable because of its standardized, abstract and vendor-agnostic approach. This prevents the risk of getting stuck with an implementation tied to a specific person or agency.

In the following chapters, we will explore how this vision is reflected in the implementation, tag manager configuration and the roles and responsibilities within an organization.