A data layer
is a variable -- or set of properties and values -- that define a piece or grouping of information to be accessed and used for the purpose of tracking data and/or interactions. At its best, a data layer creates a "one-stop shop" for referencing any data that needs to be tracked.
Client controlled means that it is created and maintained by the client, whether it is done via the IT department, marketing department, or within a CDN. But, what are the core reasons why you should have a client controlled data layer? And how can you take the steps to get there? Read on to learn more.
Page Scraping (Client-side):
With this methodology, the only values available to track are those that exist within the content of the page(s). This also means that there are certain scenarios that have inherent flaws in their implementation
of tracking. These are just a few to consider.
Sometimes, the content of your page (headings, titles, etc.) is in such language that only a core group of people may understand (for example, doctors). Medical
ailments, treatments, and their descriptions may make sense to the ones looking at the site, but not to those doing the analysis of the data
. So, the data that is pulled directly from the page may just be passed into the analytics tools "as-is", which can lead to confusion when trying to analyze the data. If any conversion takes place within the browser (client-side), then that is just more setup that needs to be updated and maintained, and can generate excess load on the page to load the extra conversion data.
Elements and page templates change over time. As content is added, removed, and/or updated on pages, the following steps may need to be taken:
New elements may have to be added to the logic in order to "scrape" the content from the page
Removed elements should be deleted from the logic for clarity
Updated elements would need to have their logic updated as well, even if the values did not change
Depending on a site's content or purpose, there is often a need to obfuscate data to hide the value behind codes and algorithms so as to not let the true values be seen in the data stores. In the client-side methodology, this process is less secure due to the need for these algorithms to be made available locally. This means that with these algorithms available for anyone to access, someone with malicious intent could reverse engineer the obfuscation to make any future calls visible to them.
Data Layer (Server-side): Within this methodology, you have complete control over what values are to be utilized for tracking. The same scenarios that have flaws in the client-side methodology are ultimately solutions in this one.
When setting the values within the data layer, IT has the ability to make the values human readable, which will only aid in performing analysis. This can allow for grouping of content more easily, classifying pages, and more. No longer does it mean having to decipher Medical terminology by those without medical degrees. It also means little to no code has to be sent to the page in order to make these conversions. In turn, this means less load on the page.
As elements and page templates are changed over time, no updates to the data layer may be necessary as content changes do not necessarily dictate value changes. This means that the same data structure will still fire exactly the same tracking calls regardless of the content of the page. However, if values do need updating or adding, IT can make the necessary changes to the data values and/or structure, and the tracking code would only need to be updated to accommodate new properties within the structure.
For data obfuscation, the algorithms can be stored and process the values on the server before making the data layer available on the page for tracking. This allows for more secure algorithms since they won't be visible to anyone who visits the page. A malicious visitor would have to engineer their own methods for reversing the obfuscation.
You! If you have a site that needs to obfuscate data, such as healthcare, financial
, or any profile data, you need to take the right steps. In addition, if you need to make your data human readable in order to make sense for reporting
, or if you have a site that changes content often (but not the needs for tracking), then this all applies to you!
Your site is where a controlled data layer occurs. This can be implemented across your entire site, or set of sites, in order to create a unified approach to data collection.
There’s no better time than the present! The sooner you implement this approach, the sooner you can see the benefits.
to learn more! Our team can personalize the approach to your site and your vertical, from healthcare to finance. Plus, we will also help you create the best solution for your needs, allowing you to reap the benefits of a controlled data layer.
A client controlled data layer is an important part of the tracking process. Understanding who, what, when, where, and how you achieve results will create a better experience for your team, and ultimately your audience.