The time has come. Google Analytics 4 (GA4) is right here and prepared for motion. Actually, GA4 is right here and prepared for takeover.
As you’ll have heard, the present Google Analytics platform, Universal Analytics (UA), will now not acquire knowledge as of July 2023. At that point, your solely possibility can be to make the change to GA4.
But why? Why is that this needed? And how will it have an effect on your enterprise? Read on to study the historical past of Google Analytics, the variations between UA and GA4, plus loads of professionals and cons, ideas and tips that will help you navigate this variation.
Table of contents
History of Google Analytics
For the search engine optimisation nerds on the market, right here’s the timeline from Urchin in 2005 to Google Analytics 4 in 2020.
2005: Google acquires a platform known as Urchin. Urchin was used to “beautify” log recordsdata and make them extra palatable at a time when log recordsdata have been our solely supply of knowledge.
2006: Google redesigns the platform and types it as Google Analytics.
2012: Google Analytics is discontinued and replaced by Universal Analytics. Universal Analytics permits customers to trace cross-platform, device-specific monitoring, and create customized dimensions and metrics. This was needed as apps and cellular searching have been changing into more and more fashionable and types wanted a approach to optimize for it.
2018: Europe rolls out knowledge privateness and safety legal guidelines known as General Data Protection Regulation (GDPR). These legal guidelines defend residents of Europe not solely in Europe but in addition on any website they browse whatever the nation the location is hosted in.
(I want to make a disclaimer that I’m not a privateness skilled and that every one companies ought to seek the advice of a privateness skilled and lawyer to make sure they’re in compliance with relevant legal guidelines)
The knowledge requirements set in GDPR really result in Google Analytics being deemed illegal in France and put tens of millions of companies liable to being fined for utilizing the platform. Not lengthy after, New Zealand, Japan, Brazil, California, and plenty of different international locations/territories roll out their very own privateness legal guidelines. With extra laws passing, Google has to make a change in the way it collects, shops, and strikes its knowledge.
October 2020: Due to the rising issues round privateness and safety, Google rolls out Google Analytics 4. GA4 has many baked-in options and instruments to permit extra management over your knowledge. Google has gone so far as to say that GA4 was built for privacy.
March 2022: Google announces that as of July 1, 2023 Universal Analytics will cease gathering new knowledge—that means you’ll have to change to GA4.
NOTE: Even although this change isn’t occurring for one more 12 months, it’s vital to notice that your knowledge won’t migrate over from UA to GA4. So if you wish to have a 12 months’s value of historic knowledge, your finest guess is to change to GA4 now
To study extra, examine our blog post covering the announcement.
UA vs GA4: Data assortment & privateness
First and foremost, listed here are the variations between UA and GA4 with regard to person knowledge and knowledge sources.
- No IP addresses. The most notable change in GA4 is that IP addresses are no longer collected or stored. In Universal Analytics, you have been in a position to see IP addresses until you bodily went in and anonymized them. This is the primary of many adjustments made to guard person knowledge.
- Easier data deletion which permits admins to delete knowledge on the request of any person(s).
- Data retention options permit you to select how lengthy GA4 can retailer user-specific knowledge. By default, the time interval is 2 months, nonetheless, you prolong it to 14 months if desired. (It needs to be famous that this doesn’t imply you’ll solely have entry to knowledge for 14 months, it simply means you lose user-specific knowledge after 14 months.)
- Options to show off location-specific knowledge and advert personalization knowledge by nation.
Data Streams for the brand new omnichannel panorama
The knowledge assortment variations aren’t solely helpful for knowledge privateness and safety causes, however they’re additionally useful for higher understanding the complete person journey.
Universal was rolled out in 2012. At this time 45% of individuals had smartphones, there have been 540,000 apps within the App Store, and there was $225.5 billion spent on e-commerce. We have been additionally nonetheless instructing a standard advertising and marketing funnel of consciousness, curiosity, need, and motion. This funnel targeted solely on the acquisition of recent prospects and didn’t actually deal with retention in any respect.
In comparability, firstly of 2021, 85% of individuals had smartphones, there have been 4.5 million apps within the App Store and $578.50 billion was spent on e-commerce in 2019. Oh, and there are only a few “mobile websites” left as Google had already rolled out mobile-first indexing.
With all of those adjustments, we noticed the introduction of the marketing flywheel. This flywheel places extra emphasis on retention and turning prospects into promoters (extra on this in our post on growth marketing).
One of the primary methods GA4 addresses the brand new person journey is thru using Data Streams.
Data Streams are the supply that gathers data and sends it to your Google Analytics property. In GA4, you may have a number of knowledge streams per property. This means you probably have two web sites, or you probably have a web site and an app, you may see how your customers work together together with your properties multi function place.
UA vs GA4: Data & reviews
When you examine the knowledge obtainable in Universal Analytics and Google Analytics 4, not an excessive amount of has actually modified. It might appear like GA4 has extra data on customers, however UA has the identical data, it’s simply tucked beneath the “Audience” tab.
The greatest distinction right here is in language round “engagement” and “monetization” in addition to the addition of “retention.”
- Engagement has changed conduct, however covers most of the identical issues with added reviews on conversions (beforehand often known as “goals”).
- Monetization, then again, is a brand new beast that we’ll deal with afterward on this article.
Now, like we touched on earlier, there was a rise in curiosity to maintain prospects and switch them into advocates during the last 10 years. Google has put retention reviews blatantly within the Life Cycle report back to replicate its significance. Though this part of GA4 is at the moment sparse, I can think about we are going to see updates within the coming years.
Users vs site visitors
As it’s possible you’ll know, a person is an individual that involves your website. If they arrive by means of their telephone one time and their pc 3 times, they’re nonetheless the identical person. Traffic, then again, refers to periods.
In GA4, person acquisition and site visitors acquisition has been cut up into two completely different reviews.
This has resulted in separate acquisition dimensions for each customers and periods, so be certain to pay attention to which one you choose when creating reviews.
Behavior vs engagement metrics
In Universal Analytics, conduct metrics have been included in nearly each report. These metrics included bounce charge, pages/session, and common session length.
Google Analytics 4 changed these metrics with new “engagement” metrics. These new metrics are based mostly on “engaged sessions,” which Google has so kindly outlined as: Sessions that lasted longer than 10 seconds, or had a conversion occasion, or had two or extra display screen or pageviews.
The metrics embrace:
- Engaged periods
- Engagement charge
- Engaged periods per person
- Average engagement time
That’s proper… No extra bounce charge!
Monetization reviews have taken over the previous residence of “conversions” reviews. The huge factor to notice right here is that monetization in 2021 regarded far completely different than monetization in 2012.
Traditionally talking, monetization was restricted to ecommerce in a technique or one other. Today although, monetization might embrace advert income, in-app gross sales, or ecommerce. In GA4, monetization reviews take all streams of income into consideration.
On the draw back, when you don’t observe something of direct financial worth in your website, this complete part of GA4 is ineffective.
This is a bummer as a result of in UA, conversion reviews weren’t solely targeted on cash. They additionally thought-about the non-monetary conversions and the attribution behind these conversions.
Session-based knowledge vs event-based knowledge
Universal Analytics collected session-based knowledge. Each session was made up of various “hit types.” These hit sorts included web page hits, ecommerce hits, social interplay hits, and so forth. In distinction, GA4 is an event-based platform, that means every little thing is taken into account an occasion. First go to? Event. Pageview? Event. The record goes on.
Now this variation is nice as a result of it offers us extra data on how folks use our properties. The draw back is that it additionally signifies that the 2 datasets (UA & GA4) don’t play nicely collectively…
And that’s why you begin from zero when switching from UA to GA4.
Event & conversion creation
Not solely are occasions handled in a different way, however they’re additionally handed by means of to GA in a different way as nicely. As you may see within the diagram above, gone are the times of motion, class, and label. In GA4, we’ve got parameters, enhanced measurement, and in-platform occasion creation.
When you arrange your Data Stream you should have the choice to activate enhanced measurement which collects web page views, downloads, outbound clicks, video engagement, and scrolls. If you retain enhanced measurement on, all of those occasions can be collected mechanically! No further work needed.
If there may be an occasion that you just’d like collected however doesn’t fall beneath enhanced measurement, you might be able to set it up proper within the GA4 platform utilizing situations and parameters. Parameters are items of knowledge despatched alongside occasions that present context on the person or person motion.
Once you’re gathering an occasion, you’ve gotten the choice to mark it as a conversion by merely clicking a toggle button—no extra “Goal” setup!
Increased performance in Google Analytics 4
During your preliminary setup, you’ll want to make sure you deal with all the choices obtainable to you within the “property” settings. We’ve already coated a lot of them, however I extremely advocate checking all of them out.
One of the final choices beneath property settings is Product Linking. It’s doubtless you’ve used Product Linking with Google merchandise earlier than, possibly by means of Search Console or Google Ads. Basically all it does is hyperlink knowledge between completely different Google Products, there isn’t any actual distinction with Product Linking in GA4 aside from the platforms obtainable.
In GA4 you now have the choice to hyperlink to Google Merchant Center, Google Optimize, and BigQuery natively and at no cost!
Perhaps essentially the most underrated characteristic of Google Analytics 4 is the flexibility to customise the platform to satisfy your wants.
Not solely can you create customized reviews in GA4 like you may in UA, however you may really add these reviews proper to your navigation. That’s proper, no extra having to navigate to a separate a part of the platform! In GA4, if you wish to create a brand new report, you may put it anyplace within the Reports navigation.
Other customization choices embrace altering the identify of collections and subjects within the secondary navigation utilizing the Library.
You can create dashboard-esque reviews utilizing the Explorations in GA4. I wish to say that Exporations is the love-child of customized dashboards from Universal and person interface of Data Studio.
Machine studying capabilities
We began to see the roll out of machine studying instruments with Universal Analytics. Insights have been rolled out in 2018 to assist folks discover traits and outliers of their knowledge. As time went on, the machine studying received extra subtle and allowed customers to “ask questions” within the search bar as a means of shortly discovering data.
When GA4 was launched, insights have been baked into the platform greater than ever. Not solely are insights obtainable and highlighted all through the platform, however new machine studying instruments have been additionally launched!
Anomaly detection, which was beforehand reserved for insights, is now obtainable on most line charts. This software is extraordinarily useful for figuring out statistical variations in your knowledge at a look. In order for anomaly detection to work, Google should have ample historic knowledge to construct predictive fashions. Which results in the following new software—predictive analytics!
Previously, if somebody wished to entry predictive analytics, they must pay for one more software or create them themselves. But, no extra! GA4 launched predictive analytics for LTV, churn chance, doubtless churners, doubtless prime spenders, doubtless purchases, buy chance and extra.
For the previous few years, Google has been engaged on its instruments. With the rollout of Universal Analytics, it rolled out its multi-channel funnel instruments. Then, in December of 2019, it rolled out its Attribution Project software which allowed customers to entry a gaggle of instruments surrounding attribution and conversion.
Now in GA4, we see a bit within the major navigation known as “Advertising,” Upon clicking into this part of GA4, we are going to discover all the identical instruments from the Attribution Projects. This consists of conversion paths, time to conversion, the mannequin comparability software, and touchpoints to conversion.
But the instruments aren’t the one attributions you’ll see in GA4. In the previous few months, Google introduced that every one customers have been gaining access to Google’s data-driven attribution model. Once that rolled out, all GA4 properties have been modified to utilizing that mannequin as their default conversion mannequin, whereas UA nonetheless depends on last-click.
Lots to study
Here is the factor: We are simply beginning to actually perceive GA4. And very like each different Google product, it adjustments day by day! There continues to be a lot to study and uncover in GA4, and this course of will look completely different for everybody.
I extremely encourage you to get on the market and mess around in GA4. It could also be uncomfortable, however I guarantee you it’s far more like UA than you suppose it’s, only a bit extra subtle. And who could possibly be mad about that?