We audited the analytics implementations for a sample of UK businesses. Here’s what we found.

Here at Yard, we pride ourselves on the standards of our Adobe Analytics implementations. We have a huge array of qualifications amongst our teams and have earned the moniker “Specialized” at Adobe Analytics. Our founder is also a Subject Matter Expert in CJA, Adobe’s next generation analytics tool. We also know that sometimes implementing a specialist tool like Adobe Analytics can be challenging; often fiddly, often very technical, and always reliant on a well-formed plan. So we decided to take a look at the state of implementation in industry, hoping that this research will help guide businesses to improve their own data capture and reporting capabilities.

First, the approach. We did not have access to any of the Adobe accounts for the research subjects. Instead, we took a look at the tags being fired on each of the websites we analysed, interpreting the Adobe Analytics requests to understand what data was being passed up. This has some limitations; you can’t see post-processing rules, you can’t see transformations that are carried out server-side, and you can’t see any data being added by way of uploads or classifications. But of our sample, we are confident that we could get a good view of what was good and what was not so good, which is the basis for our results here.

Classifying the implementations

 We decided upon a classification scheme for the implementation quality, with definitions as below:

·       Good – a good standard of implementation, no obvious issues and a wide range of collected metrics

·       Average – a good implementation with no issues, but there are obvious things we would add to supplement the reporting that would be useful to the business

·       Fair – a relatively basic implementation, no obvious issues with the data being captured, but plenty of opportunity for capturing additional data points

·       Poor – the implementation has some obvious issues and data is not being collected correctly

We looked at a sample of 100 businesses from various sectors in order to classify the quality of their implementation, with the results per the graph below:

Our research shows that only 25% of implementations in our sample were Good. In 75% of implementations, there are obvious areas for improvement, and the biggest group was Poor, with 32% of implementations showing obvious issues with the data being collected.

Where were the issues?

So far, so vague. We’ve shown that there is room for improvement in the majority of implementations we reviewed, but what were the issues that need to be addressed?

We took copious notes when completing the analysis, which allowed us to classify the main issues that we found. Broadly categorising the main issue for each research subject then produced the following summary results:

This garnered some very interesting outcomes. Analysing these outputs allows us to identify a few themes which can be used to target improvements to an implementation, including:

·       Cookie consent is still potentially quite difficult to understand. The majority of issues here were on sites with cookie banners, where the Adobe cookie was set on page one, before accepting cookies and in breach of the rules. Worth adding that some sites classified Adobe Analytics as a functional cookie, so this was in some cases by design.

·       Ecommerce tracking can be hard to get right – and some implementations hadn’t even bothered trying. This is often the most important part of the site, where the business is winning revenue and sales from visitors, but the data being captured here was often basic or broken.

·       Implementations need to be managed over time. In a number of cases, the implementation was either broken or entirely missing from areas of the site. Maintaining an implementation is often as important as the first implementation. And when new content or journeys are added, don’t forget to include the tracking.

·       Variables & Parameters are generously offered in Adobe Analytics, but we often saw that very few were being used. Be bold with your requirements, and make the most of the flexibility of Adobe Analytics when planning your implementation.

Issue Themes

Further analysis of our notes and comments allowed us to pull out additional themes to supplement the above. Although we can’t guarantee the consistency of the written notes, some obvious word themes were seen, per the word cloud below:

These allowed us to quickly identify a few additional issues:

·       Cookies and cookie acceptance. As discussed above, it seems that a lot of implementations struggle with successfully handling the cookie acceptance.

·       Product tracking. Product being the biggest word seen, and supporting the point that product tracking can be hard.

·       Missing vs improved. A similar focus was seen on things being missing entirely, versus there being opportunities to improve.

Marketing Channel Rules

Our approach meant that we couldn’t access the Adobe configurations for the companies we audited. However, another common theme from conversations with clients that we work with is frustration with measuring marketing channel performance correctly. We typically find that this is due in-part to marketing channel rules not adhering to best practice and some channels not being measured at all. These clients do not have visibility on the granular level of data to truly know which channels are performing best.

What does this mean for me?

A well implemented Adobe Analytics solution can make a huge difference in understanding and refining your marketing spend and improving your site conversion rates. There are a number of issues in this article that you can focus on to test the quality of your implementation, and a number of ways to consistently improve the quality of your collected data.

Adobe Analytics often requires some love. Make sure you are getting what you need out of it and don’t be afraid to push the boundaries of what is possible in terms of data collection. Most often, this is achieved with our clients by understanding what reports would give them data they’d love to be able to understand to answer their burning business questions. And then work with your data agency to make sure all of this is delivered accurately and correctly.

Or alternatively, give us a shout and we can guide you from wherever you are today to a best-in-class implementation that is a perfect fit for your business.