8 minutes reading time (1627 words)

As Googles Universal Analytics comes to a close, what alternatives if any do you need?

Sept-Analytics

Google's Universal Analytics (UA) has been about since 2006 but with it retiring next year it is a good time to look at alternatives and what might best suit your needs.

It was back in April 2005 that Google acquired Urchin Software Corporation. Since then it's gone through many evolutions, including asynchronous tracking code, placing it in the footer, body and head and whether it should be added directly to your code or added via Google's tag manager.

A giant in the analytics world

Google's dominance of the tracking and analytics market has been a colossus amongst its competitors. Often the alternatives shelter in niche markets to allow them to cling on to some market share.

But in July 2023 there is a changing of the guard, the old Universal Analytics is retired and the new incarnation, Google Analytics 4 will take its place, or will it?

Could this be the point that others gain a foothold in the market?

Changing landscape

One reason for such a question is how analytics and the personal data collected is viewed by those its collected from.

Times were that such information was taken without any sign of collection or permission to collect the data and keep it asked. But over the years as companies have become rich using people's data the value of the data to these companies has become more apparent. We woke up one day and realised that what they had been collecting all this time, often without our consent was… us. Our lifestyle, our choices, our viewing habits, and such companies were making their money selling our habits, our choices to advertisers and others to help them spend less in order to sell more of what they thought we liked. And that margin of error between what they thought we would like and what we actually wanted was getting smaller and smaller.

Is it lawful?

Good question, over the last few years we have had different laws coming into place in different locations with different reach and intent. 

Cookies

The cookie consent was probably the first most people knew about the lawmakers attempting to highlight and explain what the advertisers and analytics people had been doing all this time. But it's far from Universal, with some small sites abiding by the rules laid out in their jurisdiction and other world-leading sites seeming to disregard or claim immunity from the obligation to disclose their cookies.

General Data Protection Regulation

Next up is our old friend GDPR and its muddled use and rollout.

Each country enforces it differently and some do not at all.

It's there to help me the user of a site understand the scope of the data that is collected by that site about me. It gives me the right to check, change and ultimately deny the information held against my name and my activities.

What's this got to do with Google Analytics? Well tracking you, collating and sharing your data is at its heart. And then its allegedly sent back to a specific organisation in one particular country in the world

And here we have a clash of authorities with some saying that it is not compliant.

To meet German requirements, Google Analytics should anonymise the IP address.

In 2022, Austria, France, and Italy prohibited the service which lets webmaster tools track and analyse their site traffic. Information collected via cookies is transmitted to the United States and it was alleged could potentially be seen by third parties and the government as a  breach of GDPR 

Google Analytics 4 is meant to deal with the problems by anonymising the data at the source, but there have been ideas such as cohorts. Federated Learning of Cohorts (FLoC) is an example of this type of web tracking, collecting similar people viewing similar data and grouping them together to still allow a deeper understanding of their likes and needs.

And this is the opening many need to claim their stake in the analytics territory. They come at the approach and idea with different tools and philosophies. Some are letting you self-host others claiming that they give far greater analytics as they do not sample the clicks but log them all.

But do you really need any analytics?

So the story goes that in order to get more sales you need to remove anything of bloat and get the website shop loading as quickly as possible. This speed increase can be helped by switching off large download packets such as analytics code, it can also help to switch off anything that might block your site from loading while it downloads.

So you switch all the analytics off and wow, you get more sales. Wouldn't it be great to add some analytics to the page to see who, why what produced those sales? Yes but then you will get fewer sales and the site will load slower!

It's a conundrum. But don't you already have an analytics system of sorts? Most site owners can get access to their access logs and these logs can give a wealth of information about the user base and the user journey. It can give trends, the best days to send newsletters and much more and does not involve adding a single line of code to your website.

Does it matter?

So perhaps the first port of call is to sit back and see if you really need those analytics.

One of my go-to mantras is Marvin in the hitchhiker's guide to the galaxy. No, not his brain being the size of a planet but his little chat to himself at the bottom of a very deep dark hole.

“Does it matter? Even if it does matter, does it matter that it matters?

That second check of the question has helped me on many an occasion. We get swept up thinking we should, must, have to. But that's often just peer pressure, concern that we have missed something and that doing something will cover our backs, but it could be the other way round. Doing something could cost money, it has to be set up; customers, it will often slow the page load; and time analysing.

If you are someone who has made great gains using analytics or you have helped your clients to unlock sales and customers then this bit is not for you. But if you are someone who has made analytics reports, paid for analytics reports and then NOT changed your website significantly then what was the point? 

I have customers who like to see the up and down of the analytics but one thing is for sure, if they are not getting analytics, the numbers would not be lower but could be higher!

So maybe now is the moment to remove Universal Analytics, remove google tag manager and look to SEO reports and the bottom line to determine what needs to be done to improve your site. Fix the SEO issues and regularly check your customer journey yourself, or use the server stats.

But I need analytics

Ok, but which analytics do you need?

There are many alternatives to Google  Analytics 4. 

In future issues, I will be doing a deep dive into the setup and use of GA4 and if there is feedback to support it, alternatives to GA4. For now, let me end the Google part of this with the differences between what is and what is to come as far as Google is concerned.

What's the difference between UA and GA4?

Reporting Interface

The new version looks different with features being renamed or removed. While you can build customized reports in GA4, it’s a complicated process.

Data model 

Universal Analytics captures user behaviour as ‘sessions’ and ‘pageviews’, whilst GA4 is based on an event and user data model; all interactions are classed as ‘events’.

Introducing new metrics

Some metrics have been removed, and the way they are reported is different. Bounce rate, for example, has been replaced with metrics such as engagement time and engaged sessions.

‘Goals’ are replaced by ‘Conversion Events’

Goals were typically used to measure events, sales, page views, or funnel performance. Now, you need to tag specific events as goals for reporting purposes.

A greater focus on privacy and compliance

GA4 includes a number of features designed to facilitate compliance with GDPR. These include an improved data deletion mechanism, data retention settings, and updated data processing terms.

Data-driven attribution

GA4 includes ‘data-driven attribution’ by default, distributing credit for conversions based on data for each conversion event. You’re also able to retroactively apply attribution models in order to analyze historical data.

Alternatives to GA4

Some alternatives to Google Analytics include Plausible, Fathom, Objective, Jentis, Matomo,  Piwik Pro and Abralytics to name just a few

Let us have a quick look at the last and one of the newest which claims to be a very simple Analytics.

Abralytics 

Formed in 2021 its a fully bootstrapped interface

It is added with a very lightweight piece of code some 10x smaller than Googles (when GA4 is installed via GTM)

Its mission statement is “to protect end-user privacy while simplifying analytics.”

Unlike GA which samples data, Abralytics show every single event.

The features include automated email reports, configurable on a weekly or monthly basis.

They are based in Dublin, Ireland and all data is hosted in the EU.

Often seeing a demo is the best way to see if something is for you so here is a link to their demo.

Demo link - https://app.abralytics.com/#/demo

Depending on the analytics of this article and my in-depth analysis and report, I will probably form a working group to see if we should drive forward the implementation of another article to do with analytics… or perhaps I will just do it anyway!

1
Joomla! Out Of The Box - What Can You Do?
Leadership Interview: Nadja Lamisch, Treasurer
 

Comments 1

Already Registered? Login Here
Alan N on Monday, 03 October 2022 11:38
Is it now just a matter of choice?

Very interesting read, thank you Philip.

It seems that since cookie control came into being the benefits of Google Analytics and any others have been some what lost. If cookie control tools are implemented and work properly, you cannot gauge the full picture of what's happening on a website.

I get asked to provide stats but have to add that they are only based on visitors that allowed cookies. Some take the view that OK, it at least gives us a good idea - but does it? So I also provide stats from the server logs - which usually look much better, but are harder to make sense of when compared to the user interfaces of the analytics tools.

Whilst there are commercial, regulatory and ethical considerations, maybe using analytics is a matter of choice, but importantly, as mentioned in the article, do the insights gained from analytics really matter? If the phone keeps ringing and the contact forms keep coming is that really all that matters? Are you at a disadvantage if you don't use tracking analytics? Would your visitors appreciate the fact that you don't use tracking?

That said, you then have the analytics tools that provide heatmaps and record where clicks take place which are super useful when planning changes to a site so you can identify bottle necks, mobile device issues and user frustrations. We are working on a replacement for a Joomla 3 site and have used these tools to help us - arguably this type of data is far more important than conventional tracking data. But, then you come back to the performance over information conundrum!

An industry has been built around website analytics and it will be interesting to see where it heads in the coming years - do you pay for analytics solutions that might not paint the full picture? My experience of adopting opt in cookie control was a dramatic (even scary, how am I going to explain this?) drop in the number of visitors tracked, posing the question as to how useful the information actually is.

As a final thought, there is a question that sometimes arises around this subject - "if I remove Google Analytics or use an alternative, will Google penalise me in their search engine rankings?" My research (on Google!) suggests not but maybe someone knows different? Bounce rates especially seem to make some people anxious as to how they might affect ranking.

Reading this article I'm seriously considering suggesting that we drop GA. Save some time on scrutinising stats every month. It also seems a shame to impact upon the much improved performance of Joomla 4.

Then, using Joomla 4 built in functionality, do the best I can to make the content relevant, accessible, and set up for SEO. Then I can just refer to the server stats when required. If I have to use analytics, it would probably be Matamo based on previous experience with using it.

1
Very interesting read, thank you Philip. It seems that since cookie control came into being the benefits of Google Analytics and any others have been some what lost. If cookie control tools are implemented and work properly, you cannot gauge the full picture of what's happening on a website. I get asked to provide stats but have to add that they are only based on visitors that allowed cookies. Some take the view that OK, it at least gives us a good idea - but does it? So I also provide stats from the server logs - which usually look much better, but are harder to make sense of when compared to the user interfaces of the analytics tools. Whilst there are commercial, regulatory and ethical considerations, maybe using analytics is a matter of choice, but importantly, as mentioned in the article, do the insights gained from analytics really matter? If the phone keeps ringing and the contact forms keep coming is that really all that matters? Are you at a disadvantage if you don't use tracking analytics? Would your visitors appreciate the fact that you don't use tracking? That said, you then have the analytics tools that provide heatmaps and record where clicks take place which are super useful when planning changes to a site so you can identify bottle necks, mobile device issues and user frustrations. We are working on a replacement for a Joomla 3 site and have used these tools to help us - arguably this type of data is far more important than conventional tracking data. But, then you come back to the performance over information conundrum! An industry has been built around website analytics and it will be interesting to see where it heads in the coming years - do you pay for analytics solutions that might not paint the full picture? My experience of adopting opt in cookie control was a dramatic (even scary, how am I going to explain this?) drop in the number of visitors tracked, posing the question as to how useful the information actually is. As a final thought, there is a question that sometimes arises around this subject - "if I remove Google Analytics or use an alternative, will Google penalise me in their search engine rankings?" My research (on Google!) suggests not but maybe someone knows different? Bounce rates especially seem to make some people anxious as to how they might affect ranking. Reading this article I'm seriously considering suggesting that we drop GA. Save some time on scrutinising stats every month. It also seems a shame to impact upon the much improved performance of Joomla 4. Then, using Joomla 4 built in functionality, do the best I can to make the content relevant, accessible, and set up for SEO. Then I can just refer to the server stats when required. If I have to use analytics, it would probably be Matamo based on previous experience with using it.

By accepting you will be accessing a service provided by a third-party external to https://magazine.joomla.org/