Universal Google Analytics (GA), is Google’s legacy, wildly popular and free web analytics platform. It’s the number one web analytics platform and is utilized by approximately 50% of websites. Google Analytics 4 (GA4) is Google’s newest version of their web analytics platform. GA4 has been available for several years, but on July 1, 2023, Google forced GA clients to migrate to GA4. Clients using their paid service (GA 360) have until the end of the year to migrate. 

 

Many of the enhancements in GA4 are features GA customers have been asking to get for a long time. The core of the changes surround event-based tracking, machine learning/predictive modeling, and improved cross-device tracking. Specifically, these enhancements include: 

• Universal GA utilized pageviews and sessions as its primary metric base. It tracks user interactions using pageview-based tracking, where each pageview is counted as a separate interaction. Events (button clicks, email sign ups, form submissions) in GA are additional interactions. Accordingly, reporting in GA is often segmented by dimensions and metrics, but it can be limited in helping analyze the full user journey.
• GA4 is derived from event-based tracking. It collects data based on a user’s interactions, or events, across the entire user journey on a website or app. A pageview in GA4 is just another type of event vs the basis for most of the data collection. This sort of tracking enables a user-centric approach and allows you to track a user’s journey across multiple sessions and devices.

 

There are some great things about GA4 that website analytics consumers have been wanting for years. For example, event-based data collection will make it considerably easier to track micro-conversions on websites and apps. That said, it’s a brand new tool and Google’s well-established lack of documentation on Google Analytics is even worse with GA4. 

 

Bottomline: It’s going to take time for GA4 users to really understand what they are getting and how it’s different from their previous tracking. 

 

Why Should Marketers Care? 

Marketers should care about the impact of moving from GA to GA4 for several reasons: 

1. Did you set budgets based on GA and now can only track in GA4? The different tracking methodology and default setting of using Data Driven Attribution (DDA) in GA4 advertising reporting could throw off your actual results compared to forecasts. And this means you’re going to waste time researching whether the differences are caused by model changes or material channel performance changes. 
2. The GA4 DDA model has changed a lot from GA to GA4. Unsurprisingly, the changes they made to the model favor Google’s search and display channels. 
3. The inclusion of cross-device tracking also impacts the basic foundation of a user journey. Specifically, there are fewer ‘journeys’ when you enable the ability to track across devices. This isn’t necessarily a bad thing, but it will certainly impact the touchpoints in a path, which then impact the underlying model results. 

 

We believe the changes Google made to their attribution model, while not malicious, weren’t necessarily unbiased. Because of the data they have on their own products they can count things like video engagements or display impressions more easily as a touchpoint. In the affiliate channel, we have all sorts of earned media value in impressions that don’t lead to direct click-through conversions but expose consumers to brands (i.e., provide an impression). This is particularly true for content partners who are typically situated at the top of the funnel. Because publishers don’t provide this data to affiliate platforms or GA, we can’t quantify the value of these impressions, but Google can in their attribution model. 

 

Ultimately, this means that many marketers are seeing a drop in their affiliate channel revenue in GA4 relative to GA. There have always been and will always be differences in the tracking between a web analytics platform like GA/GA4 and a marketing technology platform like Partnerize. While these differences can sometimes matter, as long as the differences are consistent over time, you can account for them in your forecasting and budget allocation. 

 

What Can Marketers Do About It?

• Get a handle on the differences. If you’re lucky, your analytics team turned on GA4 tracking last year and you’ve got a year of data under your belt. Compare channel performance GA vs GA4 vs martech platform. What are the default settings for lookback periods, channel groupings, models used? With some clients, we’ve seen the gaps normalize over time as they have adjusted some settings.  
• Remember, GA4 is in the very, very early stages. We suspect that the DDA models will change over time as more data is collected and fed into them. Press Google for more documentation on what goes into the model and why their channels are favored more often. Admittedly, this is a bit easier to do if you pay for GA 360 and have a human resource at Google you can speak to. 

 

Introduce an Agnostic Marketing Measurement Tool

Ultimately, as a data analysis veteran, I think you should forget about GA4 for marketing measurement. It’s great for website and app tracking, but you’ll never get an unbiased measurement from the company you spend most of your marketing budget on. This isn’t really a new perspective for us at Partnerize though. Unlike some of our competitors, we’ve always believed that your marketing technology platforms should not also be your marketing measurement platforms. 

 

What Marketers Should Take Away

• Reporting in platforms like Partnerize should be utilized to optimize within channel spend given this is where you’re actually making payments to partners. You will always have more data points in the platform to optimize on compared to your web analytics tool.
• Cross-channel measurement and channel level budget allocation should be done with an agnostic third-party measurement solution. There’s many out there depending on your budget, global region, and vertical. I particularly like the folks over at Rockerbox. They are affordable for mid and large size brands and have been focused on innovative features that drive insights. Other attribution companies worth mentioning are LeadsRx, C3 Metrics, and Measured. 
• If you are a small brand and investing in an agnostic attribution tool isn’t in your budget, consider using more than one GA4 attribution model to measure channel performance. Review last click, first click, and DDA. Determine which one is closest to the results you had in GA. Utilize this one through the end of the year. Then, beg for budget to get an independent measurement tool. Rockerbox even has a free starter plan you can try to prove out the value the tool can bring.  

 

As a marketer, it’s your job to educate your company on the impact GA4 can have on budget allocation, forecasting, and channel-level measurement. 

 

Make the case for an agnostic marketing measurement tool that will enable unbiased decision making for cross channel budget allocation and your organization will be better for it.