Google Analytics: 20/20 vision
It’s kinda a big deal. One of the most widely used website analytics services has taken a giant leap forward in the evolution of its product. Google Analytics 4 or GA4 has been described as a major upgrade from its predecessor. With a new look and feel, robust features have been developed to extend beyond the needs of 2020.
GA4 is here to stay and soon will be the new norm in website tracking. Whilst you can still set up properties in Universal Analytics, GA4 is now the default option. Google have made it clear that their focus will be on enhancing Google Analytics 4 and likely retire Universal Analytics in the future.
A little history…
Over 15 years in the making
Google first acquired Urchin, a web statistics analysis program, back in 2005. Swiftly, they rebranded the enterprise level analytics tool to Google and released for free to the masses. Demand rocketed.
Between 2008 and 2011, Google released a number of minor and major updates to enhance the product including custom reporting, advanced segments, and event tracking.
2012 ushered in huge changes to the product in beta form. Universal Analytics became the latest version of Google Analytics. It offered cross-platform tracking, flexible tracking code enabling data collection from any device, and custom dimensions and metrics. In addition to this Google launched Google Tag Manager (GTM), a tag management system, with the ability to manage and deploy marketing tags.
In 2016, Google began to differentiate their offer with a premium software suite known as Google Analytics 360. It combined the main return on investment and additional marketing indicators. Products included: Analytics, Tag Manager, Optimize, Data Studio, Surveys, Attribution and Audience Center. The basic version remained free.
In October 2020, Google released the next version in the evolution, hailed to be the biggest update to the product since launch — Google Analytics 4 (or simply refer to as GA4 ).
What’s it packing?
Google Analytics 4 is looking to the future and its latest batch of features are a direct acknowledgement of the changing data landscape and development of users needs.
Cookie-less world
Google Analytics was structured around collecting user information via cookies. For each page visit or level of engagement, the data was packaged up through cookies and sent to Google Analytics for processing.
With data privacy a fast growing concern, and changes in the law regarding GDPR (EU)and CCPA (California), users are required to opt-in to have their engagement monitored. As a result, data gaps are likely to emerge.
Google Analytics 4 blends existing data with machine learning to bridge the gap in its reports and ultimately to continue to thrive as a useful analytics tool.
Alerts and predictions
Based on data already collected, GA4 has embedded AI capabilities to model what we know about how users engage and create predictions overtime. Users will be supported by AI-powered insights offering predictions and automatic alerts regarding trends and demand increases.
Data deletion
Advanced capabilities have been added to data deletion process, which for the first time, will enable us to delete certain aspects of data we collect. Plus, there is a preview mode for final checks, so we can understand the impact before we permanently remove data.
Integrations
GA4 has built-in integration with Google BigQuery, a scalable analysis tool with built-in machine learning capabilities. In addition to this, there is greater ad integration for marketeers to manage their users across web + app.
Reporting — Moved, merged, replaced, removed
One of the easiest changes to spot is that the look and feel of Google Analytics reporting is fundamentally different.
Whilst this may appear jarring at first, Google have redesigned Analytics with the customer journey at the heart. In other words, the out-of-the-box reporting has been consolidated in terms of the user’s lifecycle:
Acquisition > Engagement > Monetisation > Retention
For those desperately seeking their standard go-to reports, the ‘Explore’ section will allow users to generate their own custom reports. A definite step-up from the often limited ‘Customisation’ section in the previous Universal Analytics.
Other areas to note:
- Real-time remains fairly intact
- Audience reports are now in the User sub-section
- Events are managed in their own sub-section, but there’s a good reason for this, which I’ll get into more detail below.
It’s all an event
In a bid to adapt Google Analytics to provide meaningful data, there has been a pivotal shift in all items we consider as events. Page views and events have been consolidated to both exist as ‘events’. Strictly speaking, all ‘hits’ and ‘interactions’ are considered events.
Events are no longer limited to collect information in category, action, label and value. Google Analytics 4 enables us to send a list of parameters as data, with each user ‘hit’ or ‘interaction’. The benefit? Significantly more relevant data with each user action.
Take video engagement for example
Previously, we would use Google Tag Manager to define which variables we would want to gather information such as provider, url, title, duration etc.
We’d need to distinguish video events from others events, therefore the ‘Event category’ would need to say ‘Video’. We’d need to refer the video title we were collecting data on, ‘Event label’={{Video Title}}. Then this would leave ‘Event Action’ to capture a remaining piece of data — a little bit messy.
Google Analytics 4 has video engagement as an enhanced measurement (see more detail in next section). Automatically enabled to capture and begin data reporting on; video title, video duration, video provider, video url, Video visibility, etc. No need to dive into Google Tag Manager. Parameters are already defined and there are more than 4.
Types of events
Of course, no two events are the same, and in fact Google Analytics has 4 different categories for the types of events that one can set up.
- Automatically collected — these events are triggered by basic interactions with the site and are enabled by adding the GA4 tracking code to the website.
- Enhanced measurements — typically these were a set of events that required basic understanding of Google Tag Manager and were commonly added into most GA setups. Aware of the time burden in set up, Google responded to make this a built in feature with options to toggle on or off. Enhanced measurements include some of the basics in website tracking: page views, clicks, downloads, scroll, video engagement, external links
- Recommended events — Manual implementation required, but with Google’s predefined names/parameters. Google attempts to make these industry specific. They’ve released help information that links to developer notes, so as analysts and marketeers we can decide which additional events are required for our tracking set up
- Custom events — Events you name and implement yourself. If what you’re looking for doesn’t fit in the 3 categories above, you can still create custom events to boost your understand of your specific data need.
Goodbye Bounce Rate, hello engagement rate
That’s right! Google Analytics 4 does not calculate the bounce rate.
Bounce rate — the percentage of users who leave your site after viewing a single page, rather than continuing.
This has been a fairly standard indicator in website analysis, however Google has decided to replace bounce rate with a more meaningful metric called engagement rate.
Engagement rate is the percentage of users who meet the following criteria:
- Interacted with the website for 10 seconds or more. Note: in the foreground, being minimised or off-screen won’t count;
- A conversion event was fired, marking that a business objective was met on the page;
- The user records 2 or more page views, which is the current definition of “not bouncing”.
Key difference? Users who only visit one page but read the content are now considered engaged.
Engagement rate will likely be higher than just the inverse of bounce rate.
Converting goals into conversions
What’s the aim of the website? Can we measure it? Goals have been used in Google Analytics as a definition of key website objectives i.e. sign up to a newsletter, submit a form.
Each goal could be defined by either:
- destination — a page a user lands on
- duration — an amount of time the users spends on the website
- pages/screens per session — number of pages/screens visited whilst on the website
- event — the triggering of a specific event
This has now changed. Goals have now been renamed to Conversions.
Conversions are all marked by the triggering of an event — no destinations, duration, or pages/screens per session.
Any event in Google Analytics 4 can be toggled on as a Conversion to track meaningful objectives.
Any existing destination goal will need to be set up as an event in order to be tracked as a conversion.
Goal capabilities were limited in Universal Analytics. The move to Conversions will enable us to measure incredibly detailed and specific website objectives.
Data filtering
Not all traffic is good traffic. Here, I don’t mean uninterested users, but you, your team and company. These users are likely to skew the data on what your customers are trying to achieve and so we use filtering to remove them from the data collection process.
It’s a tricky game, get it wrong and you can lose data.
In Universal Analytics filtering was completed at the view level, allowing teams to generate multiple different views on the data with various filters. However in Google Analytics 4 there is no view level. Instead filters need to be applied at the property level — terrifying.
Two types of filters are available:
- internal traffic
- developer traffic
Filters are based on parameters added to events.
Thankfully, Google have responded by introducing a testing mode for users to apply the filters and see the effect before final implementation.
Path analysis
Already a feature in Google Analytics 360, the Analysis hub tool has now been rolled out as staple feature in Google Analytics 4.
An improvement to the navigation explorer and behaviour flow reports, path analysis will become fundamental to understand how users are navigating before reaching intended website goals. Combined with the update that all hits and interactions are events, we can identify exactly what users are doing page to page.
Alternatively, we can also work backwards with ‘Reverse pathing’ to identify the steps a user had taken to reach their end goal.
The benefits will allow us to target key landing pages, highlight drop off points and emphasise pages that should have call-to-actions.
Elapsed time
How long does it take our users to do ‘X’? Performance analyst are often tasked with trying to gauge how users are spending their time on websites. Measuring time has typically been limited to focusing on average session duration and average time on page.
Custom events could be set up to add timestamps to certain actions, however this hasn’t been an intuitive process and requires GTM knowledge.
GA4 have released an exciting, new feature to provide a better understanding on time. The ‘elapsed time’ feature will allow analysts to see on average how long it takes a user to complete a number of events, on any event. Combined with the enhanced measurement events, this will become a vital tool when trying to establish which sections are adding most value prior to completing a website goal.
What’s next?
- Product development. Google are continuing to expand on GA4 and hopefully soon more information will be available on the machine learning element.
- Migration plan. The general consensus from most analytics experts is that it’s best to set up a GA4 property to run in parallel with your current universal analytics property. Begin working on a migration plan to map over your existing website objectives into GA4.
- Try it out. It’s a new tool with lots of shiny features, however, a lot of the old favourite reports cease to exist. It’s going to take some time to get up to speed with how best to utilise this latest version. So get going…
Sources and additional reads:
- https://searchengineland.com/google-analytics-4-adds-new-integrations-with-ads-ai-powered-insights-and-predictions-342048
- https://raddinteractive.com/what-is-the-new-google-analytics-4/
- https://www.aillum.com/blog/5-things-you-should-know-about-google-analytics-4/
- https://blog.google/products/marketingplatform/analytics/new_google_analytics/
- https://www.analyticsmania.com/post/how-to-track-events-with-google-analytics-4-and-google-tag-manager/
