The new Google Analytics
This week, Google announced that new analytics properties would by default use “Google Analytics 4” (formerly, App + Web). What does that mean for The University of Edinburgh?
In March, I demonstrated ‘App + Web’, which was then being touted as the future of Google Analytics. This new model, based on interactions (‘events’) that promised to enable a better understanding of our users over the complete lifecycle of their website usage.
March Web Publishers’ Community (University login required)
At the time, I thought it might be two years before this became the reality of analytics. If you too want to look silly, make predictions about the future.
What are the advantages of Google Analytics 4?
Google’s new model promises to deliver a better understanding of how users interact with our websites and applications.
Data will no longer be broken into reports that can act to shield site usage truth; instead, by focussing more on what people do, we can understand user interactions across all their visits to our web estate.
Google assure that in future our users will have more control of their analytics privacy: cookie/identifier free data collection, analytics modelling and machine learning can tell us about how our website is being used while respecting our user wishes.
That sounds great. I’ve been sceptical of machine learning ‘insights’ in Google Analytics, as the varied purpose of our large and diverse web estate never seems to translate. But this could be because our analytics set up is not compatible with how Google thinks we should be working.
Google’s announcement can act as an opportunity for us to rethink how analytics is delivered in our estate.
Introduction to Google Analytics 4
Predictably, Krista Seiden has provided the best introductory coverage for this new model.
Access to the new analytics
We enabled a test property using the new collection and reporting model on the central site (www.ed.ac.uk) last year.
If you are a Univerity user and want to experiment with existing site data, please mail firstname.lastname@example.org with your Google-enabled email, to gain access.
Google Analytics access at the University of Edinburgh (University login required)
University analytics in future
In the short-term, at the very least, we need to update all of our support and training (existing materials refer to a product that soon won’t exist).
We’ll also need to review existing analytics implementations to ensure they are collecting valid data.
But beyond this, we need to consider whether our current analytics is actually working for us. We should review whether this new model works for our web estate. Our current Web Strategy reflects the need to do this; this week’s announcement only makes it more imperative.
We need to focus our data collection efforts on what is of value to us, in a way that respects how our users want to use our sites.
If we conclude the new Google model doesn’t suit our measurement needs, we need to look at alternatives.
But using a different platform isn’t cheap or easy.
Alternative solutions may bring benefits (in insight, or privacy) but alternative solutions (even a nominally “free” open source solution) will mean additional needs in:
- training and support, as few will enter the University with knowledge about how our analytics works
- web development, to ensure data collection is functioning effectively
- database storage and security, as data will need to be safely stored
We need to understand our full web estate to establish our needs, then plan an analytics future that works for us. Google’s announcement could prove just the impetus we need to begin that process.