Year-on-Year comparisons in marketing analytics: Comparing UA to GA4

Building a robust Web Tracking Setup for Year-on-year Overviews

In the world of marketing analytics, having a robust web tracking setup is a monumental necessity as it is crucial to identify the sources that are acquiring traffic, new visitors, and conversions. Whether you are an e-Commerce website trying to optimize revenue, or a B2B company focusing on gatherling leads, assessing visitor behaviour is a key requirement to your success. There are many web analytics solutions available in the market, but the most adopted one worldwide is the Google stack solution, consisting of Universal Analytics (now deprecated) and its successor Google analytics 4.

UA was deprecated in July 2023, meaning that if companies had not moved at that time to GA4, they would have had to fully migrate to the new platform. As we are getting closer to the end of the year, comparing the data between 2022 and 2023 is not that straightforward, for the companies who decided to wait until July to move to the new platform. The reason being that the two platforms have completely different data models, different metrics calculations and different data ingestion methods. So eventually if you are using the Google stack and want to perform a year over year comparison, you have to be very thorough, as your one source of truth has changed.

Comparing Data between Google Platforms UA and GA4

Comparing data between Universal Analytics (UA) and Google Analytics 4 (GA4) can be challenging due to the differences in metrics and tracking methodologies between the two versions. However, it is possible to perform a year-on-year comparison with some considerations.

GA4 vs. UA: Data Quality

We believe that Data Quality is the most important consideration! Our approach is: “If I am comparing faulty data to begin with, my end assumption would be faulty as well”. When we mention data quality in that context, we refer to the set up of both UA and GA4 on your website. This is a key step into starting your YoY assessment.

If you want to analyze your data between two distinct platforms, you need to make sure that they are both recording data with a certain level of consistency between the two. You will never get a 100% match in data analytics. Rather, you can notice patterns and trends in how the data is being recorded. If you notice a discrepancy in traffic acquisition of 40% for the same time period between UA and GA we definitely don’t recommend proceeding with your comparison. In this case there’s definitely something wrong with the setup!

Here we also consider consistency. If there are suspicions of badly recorded data, assess where you might have a faulty setup. Always think of quality over quantity. More often than not, analysing faulty raw data will end up in taking erroneous “data-driven” decisions. If you switched early enough to GA4 (beginning of 2022) – Congratulations! You most likely have access to a much better and precise year-over-year comparison, as you will be comparing data recorded on the same platform

Event Tracking in UA vs. GA4

As previously mentioned, the two platforms follow very different models. While UA was focused on hits (events, social interactions, transactions, etc..) GA4 is based on events. Everything a visitor does on a website tracked via GA4 is an event, including a pageview. The way conversions or goals are set up is also different. Why is this relevant?

Events and conversions are tracked very differently so it is very difficult to conduct an objective comparison between them, especially if events and conversions were migrated automatically from UA to GA4. You can only compare events between the two platforms if you have a robust tracking framework and events in UA were replicated one to one in GA4.

For Ecommerce websites, the most important metrics to keep an eye on are the ecommerce events. These events include but are not limited to a “view item” event at the top of the funnel and a “purchase” event all the way down in your funnel. However UA and GA4 respond to different structures of the ecommerce object. If there are discrepancies between the structures we advise you to avoid comparing purchases. Any change or faulty setup in one of the two would lead to erroneous interpretation of the data.

Interpretation of UA and GA4 data

Speaking of interpretation, we arrive at our third topic to consider when conducting YoY comparisons.The approach you chose when analysing your data is critical. Numbers, even in a perfect world and a perfect setup, will never fully match. There will always be discrepancies. Throughout your analysis, try looking at your website’s performance holistically. That means: don’t focus on absolute numbers! Rather, put an emphasis on overall trends, seasonalities, tendencies, and patterns. Always assess your data by putting a context behind it, taking into account how this context can affect the volatility of your readings.

Comparable Metrics between UA and GA4 in a YoY overview

To help provide more guidance, we will now go over objectively comparable metrics between UA and GA4 so you can learn how to more accurately assess the behaviour of your visitors.

Acquisition metrics

Acquisition metrics include both user and traffic acquisition (sessions and new users). Sessions and new users can be objectively compared as these are heavily related to the setup of the platforms. Even though sessions are calculated differently in GA4 than in UA you should only see a low discrepancy between the metrics. If you notice a big difference of around 30% you need to reevaluate your migration to GA4.

Alongside acquisition metrics, user and traffic sources, mediums, and channel groupings like organic, direct, referral are important. Keep an eye on these metrics to ensure your marketing and non marketing efforts are being properly received and attributed in your GA4 reports.

Behavioural metrics

Let’s move on to behavioural metrics on your website, like pageviews and landing pages. Those are, in terms of behaviour, the most objectively comparable metrics between UA and GA4. What it comes down to is a comparison of the most viewed pages on your website. In case you see a big discrepancy there, you’re facing an issue with your GA4 setup.

Even though events and conversions fall under the behaviour umbrella their comparability depends on how robust and mirrored your GA4 property is in respect to your previous Universal Analytics setup.

Event tracking metrics

Conversion rates, purchases, event counts, revenue, conversions: all of these metrics are very delicate to compare. Be very wary of your setup and be diligent in deciding whether to proceed with the comparison or not. Ask yourself: Are you confident in your new GA4 setup? Are you sure that you mirrored your previous UA event configuration in GA4? If your answer is “Yes!” then you can add those metrics to your YoY comparison. More often than not, however, GA4 properties are considered “fresh and new” setups as they don’t follow a strict event structure and are very flexible in the way you can configure events.

Keep in mind…

Data quality is the most important and crucial factor when conducting a successful year-on-year comparison between UA and GA4. As previously mentioned, the two models are extremely different and as such, it would be very misleading to prepare a YoY comparison when large discrepancies are obvious in the raw metrics. We recommend you be extremely careful in the approach you decide to take.

Keep in mind: Numerical values will most likely never match exactly due to the different tracking methods UA and GA4 use. When conducting YoY comparisons, focus instead on insightful metrics, trends and patterns. Additionally, ensure that your tracking setups in both UA and GA4 are accurate and consistent to minimise discrepancies in the data.

If you ran into issues when preparing your year-on-year overview or you have doubts about your GA4 migration or property, don’t hesitate to contact us. We have successfully migrated several UA setups and created new GA4 properties from scratch. We’re also always looking forward to to exchanges with other Data Analysts about GA4, UA and all things Web Tracking and Analytics.