As a member of the digital marketing industry, you’ll likely be familiar with the term data layer, but perhaps lack confidence when it comes to understanding exactly what it means. We sat down with Incubeta DQ&A’s Analytics Manager, Monika Mesnage to answer just that and provide some clarity, and understanding, on data layers.
There are many definitions of Data Layer, with my favourite being from Analytics Developer, Simo Ahava’s piece The Data Layer:
- The description of business requirements and goals, aligned in a format that is readily transferable to technical specifications
- The concept of a discrete layer of semantic information, stored in a digital context (dq&A note: digital context expanding the definition from a ‘website’)
While I consider this a great definition, it didn’t quite resonate with those of my colleagues who hadn’t worked with a data layer before. Rather, it is a great definition for those who already understood what a data layer is.
After a while, I came to produce my own definition of said data layers. Choosing to describe it as the information surfaced on the website about the current user and their journey, in an easily readable format. This information can be passed to an analytics solution, and be used for personalization and marketing.
An example of a data layer with page information looks like this:
Using a Data Layer to foster greater Collaboration across Data, Marketing & Development teams
Please note: Data Layer is not the only way to achieve the below, and doesn’t have to be the best for your organization and circumstances. For more information, please reach out to Incubeta DQ&A to discuss how we can help facilitate your maturity journey.
There are multiple benefits to implementing the data layer. It is important to say, however, these benefits are not exclusive to the data layer – they can often be achieved in other ways. This is why it is crucial to explore the wider company data strategy, marketing and technology capabilities before recommending this solution. However, if it is chosen, this is how we’ve seen it work.
1: Planning and strategy review:
Data layer needs to be designed to answer business questions, which triggers a collaborative organizational process defining the main business questions which the information contained in the data layer needs to help answer . As the requirements are defined and translated to business questions, departments often plan which marketing actions can be taken as a result of having more information.
In theory, this process does not need a data layer project at all and can be done completely independently! From my experience however, these tasks are often de-prioritised as business BAU – logistics of multi-department collaboration require persistence and organisation beyond what is achievable on a regular basis for many businesses, and implementing the data layer is motivational for the stakeholders to get involved and have their requirements considered.
2: First party data audit – and how it can be utilized
According to an article by Boston Consulting Group, ‘first-party data is a key requirement for more advanced uses such as cross-channel orchestration and personalization, which enable companies to serve customers better’. That being said, less than a third of marketers are consistently integrating this data into their marketing efforts.
While the barriers to using this kind of data vary greatly by vertical, when talking about B2C businesses, it is often the siloed nature of the data. Data about customers, their transactions and preferences is often collected in internal data warehouses, and far less frequently used to personalize customer experience – or marketing messages – based on the customer purchase lifecycle.
Much like the planning and strategy review, this review can be done independently of a data layer project. However, when a data layer is being designed, one of the requirements is often the enrichment of user-related information, which results in developing ways to import back-end data into the data layer, which in turn allows for higher levels of personalization of marketing and website experience.
3: Implementation and QA
Although far from an attractive process, the actual implementation of the data layer variables and the QA of the values which are passed through can be extremely valuable. With developers being one of the most scarce team resource, a data layer – which is understandable by every marketer within the business – offers an opportunity for shared responsibility for QA and maintenance of the data.
This benefits the business by decreasing the time investment from IT, while enabling marketing stakeholders to drive the requirements and utilize the data available in something very easily auditable, using one of the many user-friendly extensions.
Partnering with Incubeta DQ&A
As a Google Sales Partner, businesses often contact us when they reach the limits of their free analytics tool. While some analytics solutions are set up very well, and only need scaling, it’s more commonly the case that the analytics solution was designed many years (and employees) ago – no one understands the current implementation, and the data is not trusted.
If that’s the case, we often recommend implementing, or re-implementing, a data layer. The process itself forces the stakeholders from different departments to work together to create a list of business questions which the analytics implementation needs to facilitate.
If you’ve read this far, it’s likely you are a professional marketer, and you want to know how to achieve your business goals – and not how to design a data layer. If that’s true, we recommend:
- If you have a very limited internal expertise, reach out to someone who has plenty – either a contractor or a consultancy (we recommend ourselves here naturally) – this is because when you plan how to measure something for the first time you’re not able to foresee all the consequences of setting up a variable one way or another – while an experienced professional would be able to do so
- If you have some internal expertise, you can prepare the measurement plan on your side – but we do recommend at least consulting with a seasoned professional to ensure that what you are implementing won’t have unexpected negative consequences
- If you have people with this expertise internally, great! Ensure all business stakeholders bought into the measurement plan and are aware what the workload required (and by which teams) is, what the development and implementation timelines are and that the priorities are clear throughout the business