Select Page

layering within your tag manager

Adobe Launch is an advanced tag management system and like many Adobe tools is extremely powerful and flexible enough to suit almost any technical landscape for any sized business.

One common challenge with a product that has such flexibility is that it can be difficult to work out how to set it up in the first place. You would not be alone in thinking, ‘How do I architect all my digital marketing tools and tags when presented with a powerful and flexible yet clean slate platform?’

This is but one of many tasks you need to tackle when implementing an effective digital marketing technology architecture. Specifically,

  • Having a consistent, accurate and flexible data layer is a must,
  • designing and configuring your specific marketing platforms is another key component,
  • but equally important is the architecture of your tag management system.

Getting your technical architecture right from the start can save many headaches down the track and also ensure that your chosen tag management solution is setup for maximum efficiency and ease of maintenance by a variety of resources in your organisation.

Adobe Launch has a relatively under-utilised feature called Direct Call rules, which in their early life were most commonly used by some companies to actually embed into their site for the direct calling of specific rules in Adobe DTM. Now we certainly aren’t recommending this approach for a few reasons, most notably that it means your choice of tag manager is automatically embedded into your application code.

However Direct Call rules can also be called from within Adobe Launch itself, and most critically, you can also now use them to pass an object of data along to whatever subsequent rule is listening out for that Direct call rule.

There is a great post outlining this capability here.

What this feature also enables is a concept of layering within Adobe Launch.

By layering key phases of your tag management processing, you can more easily manage and segregate the different stages of digital marketing tag processing including:

  • Event listening
  • Data processing and transformation
  • Execution of your digital marketing tools

For example say you have 4 key marketing tools you want to implement across your site with some basic requirements as outlined below.

Scenario Adobe Analytics Google remarketing Ad server Social Media pixels
Every page

Requires a common set of variables

Requires an account id that is different per region

Requires an account id that is different per region

One account id for the entire site

Product search

Requires multiple detailed product attributes

Requires a subset of product attributes in different format

Requires a subset of product attributes in different format

Requires a custom event identifier

Modal window for registration confirmation

Requires a custom event

Requires a specific conversion label

Not tracked

Not tracked

Product purchase completion

Requires multiple detailed product attributes in tool specific format

Requires a revenue figure plus a basic category attribute

Requires a revenue figure only

Requires a conversion event only

 

There are many architecture options to implement these tools in Adobe Launch including:

  • Create 1 rule for each tool and scenario combination – e.g. ‘Adobe Analytics product search’
  • Create 1 rule for each scenario and have multiple tools sit within the actions for that rule e.g. ‘All tags – Registration Confirmation’
  • Create 1 rule for each tool and try to handle the different scenarios within custom code e.g. ‘Ad Server tags’

Each of these would likely achieve the desired outcome, but there are many other factors to consider such as:

  • How many rules will you end up with in your interface
  • How will you handle things such as new events coming along that need integrating in to the existing architecture?
  • How will you deal with fixing incorrect data being passed through from your application or data layer?
  • What happens if a new attribute is required for one of these scenarios?
  • Who will maintain your tags and what sort of skills will they possess?

Our layering approach to tag management architecture is designed to detach each phase of tag processing from one another, and our philosophy is that the mapping and sending of data for a specific tool should be independent of other installed tools.

Phase 1 – Event listening

The first phase is the Event Listening phase, whose sole purpose is to ‘listen’ out for events which are occurring on your site. These could include things such as:

  • A simple page load
  • A single page app screen change
  • A specific button click or user action
  • A data layer event firing

Whatever event you wish to detect on your site, should be defined in this phase, and subsequently this phase will trigger an internal Adobe Launch Direct Call such as ‘event-detected’ with an object of data including the event type detected and any additional data that may have come along with it such as the data layer event object.

Phase 2 – Data processing and transformation

It is pretty much a given that something about your data layer will be slightly incorrect or not 100% exactly as you wanted it to be when processing a particular event that has just occurred. It could be that an attribute is missing, or its in the wrong format, or it simply needs to be transformed in some way.

Aside from working with your development teams to resolve the issue at source, typically you would need to handle this situation within a data element which would be referenced by specific tools that require that piece of data.

By having a data processing and transformation phase, you can centralise this function in a specific Adobe Launch Rule completely independent of any tool that will later use the data.

Once your data has been processed and/or transformed (or left unchanged), once again, an internal Adobe Launch Direct Call such as ‘data-processed’ with an updated object of data can be triggered to move along to the next phase of processing.

Phase 3 – Execute your digital marketing tools

As the name implies this is the phase where you configure the specific logic for each of your required marketing tools. You may still end up with multiple rules per marketing tool for different scenarios however they will all be working with the same set of data and from a consistent set of events and defined event types that have been fully processed.

Another benefit of layering within Adobe Launch is that it makes testing and debugging easier. By detaching the different phases of processing from each other, it is much easier to determine where issues are occurring.

Furthermore, given that the underlying architecture of this layering approach utilises Direct Call rules, these can be manually triggered from within a browser console, so it is relatively straightforward to simulate and even automate different event types for ongoing testing and validation.

Final note, it’s worth mentioning that an Adobe Launch direct call rule is just like a generic browser custom event, so the concepts explained here could just as easily be applied to other tag managers including Google Tag Manager, Ensighten or Tealium, although some of these tools have their own similar concept also.