CT No.61: The brand new Google Analytics 4 and what it means for content analysts
What's one more major change in a year where norms are shifting?
To all of you who responded that you’re interested in education: I’m still making meaning of your responses but you are next on my to-do list, I swear.
Intro to our house of horrors
Why learning analytics is good for content strategists and creators
Changes coming to the new Google Analytics 4 setup
What you should do right now about GA4
A special offer for Content Technologist subscribers
Links, per usual
No review this week because I learned GA4 yesterday, so I’m gonna let that count!
You can’t get rid of the Googadook
I enjoy a good scary movie. I steer clear of gore for gore’s sake, but I will happily watch a solidly thinky, too-pretentious horror flick with clever storytelling and stunning visuals.
The one type of horror film I hate, though, is an it’s inside the house movie. I was a kid who was scared of the dark, so things that go bump in the night can happen anywhere but in my closets, thanks. I don’t enjoy being frightened in general, and calling into question my one safe place is not my idea of fun. Paranormal Activity will always be a hard no. I couldn’t finish The Babadook because I was too frightened, although friends had told me “The Babadook is as scary as Rosemary’s Baby.”
No way, The Babadook is wayyy scarier than Rosemary’s Baby. For me, anyway.
I want my house to feel safe, especially this year.
This week I was planning on writing an essay about Google Analytics setup and analysis errors that feel inside the house, that make our data foundations feel unstable. I had a whole kooky setup that involved inflated bounce rate monsters and tentacles made from out-of-context line graphs.
And then Big Goog literally dissolved the floor beneath me when it announced its transition to a new analytics interface, Google Analytics 4. Goodbye, comfortable professional house. Hello, evil twin.
I tried to rewrite my original house-themed horror post with the new Google Analytics 4 context in mind, but it was really dorky. When I found myself near tears, it was like: woman, there’s no need to commit to some goofy spoofy essay about bounce rate creatures when Google is likely nixing bounce rate. I’ve seen far too many other monsters in 2020 to be felled by one of my own imagination.
Admittedly, I wasn’t jazzed about the sudden new Google Analytics changes. The framework I’ve relied on for so many years — at least I will always have GA for content analytics, and it’s free! — shifted drastically.
Of the big tech companies, I’ve always favored Google because of its analytics and the data it gives away for free. But at the end of the day, my reliance on Google for my income often feels like a trap, like some Babadook that also shops for my groceries and serves me dinner while it slowly takes over my home.
I did not finish watching The Babadook, but I read the plot summary on Wikipedia.
High level: Google Analytics 4 is certainly different, and it’s clear Google wants you to use it to purchase more ads. It’s going hard on the predictive analytics, real-time trends and automated insights that are very sexy to describe but don’t always stand up when you start asking hard questions about sustainability.
But the high-quality good content analytics are still there in GA4, and they’re souped up with audience analysis, out-of-the-box event tracking and better engagement metrics.
My tentpole landing pages report will disappear, but different reports will grow in its place. Privacy protections are much better, and on-site behavioral analytics will be easier to prioritize.
Watching your house shift overnight and your old software buddies turn to skeletons is part of digital life. Like all aspects of marketing and the entire concept of capitalist markets, digital measurement is full of ghosts and sprits, inaccuracies and faux trends, shady attempts at attribution models, trackers that we don’t understand, different sets of numbers from one tool to the next.
The best content marketers and publishers can do is understand the core software we’re working with — and the methods and theory behind it —and use it to make better decisions about how we create and distribute content.
C’est la guerre algorithmique!
If you’re looking for non-scary experimental horror films about houses, I highly recommend the 70s Japanese cult classic House (aka Hausu), which is on Criterion and HBO Max rn. It’s bonkers in all the best ways.
Now I’ll leave the weird horror metaphors behind.
The new Google Analytics 4, explained for content operators
The new Google Analytics 4 will eventually replace the Universal Analytics setup that was introduced in 2012. I’m sure it will be at least two years until most companies fully transition to Analytics 4, but eventually the old UI most likely be deprecated and permanently replaced.
Google’s been testing the interface for about a year, in a feature called Web + App, but I’ve previously ignored because I only use Web and not App. The new setup is still technically in Beta.
You can transition to the new GA4 today, while maintaining your old Universal Analytics profile. Run them both concurrently, especially since GA4 will only start populating data from the day you install it. Almost no one’s going to have GA4 data from before October 14, 2020.
You probably don’t have brain space to learn a new analytics tool in 2020. I don’t blame you. But one day, I advise getting to know content analytics standards so you can read and react to the new platform.
Or you can ignore it and let someone else make the decisions for you. That works when you have an analytics partner you trust. But that’s not the case in many organizations.
Why do content strategists even need to learn analytics?
Content has never been just words. I encourage digital content strategists — whether you’re in content marketing, media production, or analytics — to become certified in Google Analytics. It’s not only a career-advancer, it’s also a job protector.
I’m not a fan of the “everyone needs to know and do everything” school of thought. Writers don’t need to learn code (kinda wish more coders would learn to write tho). Graphic designers don’t need to learn the intricacies of grammar. Reporters don’t need to learn deep content strategy, and content strategists don’t need to know how to file a breaking news story. We all have specialized skills.
For me, though, knowing content analytics is part of fact-checking: an essential digital content skill. You don’t need to know statistics or forecasting; you just need to understand what’s happening with your content.
Understanding at least the basics of Google Analytics, aka content analytics, is a crucial dimension for knowing how the content we create affects business performance as a whole. Why?
Content creators who understand analytics are able to build data stories and create more nuanced, strategic insights that drive results faster and more predictably than any AI-generated analytics tool.
Why? AI-generated analytics “insights” are more like highlights. They spot patterns and abnormalities, an unusual increase here or a higher-than-average segment there. When many analytics n00bs read those auto-generated insights, they tend to think the recommendation is “do more/less of that one tactic” without understanding the circumstances of the full story.
Storytellers and content creators are deft at putting those pattern anomalies in context, adept at seeing the full picture. They can also identify healthy and sustainable growth trends, which are far more important to business growth than random spikes and barely impactful rando segments.
Similarly, content folks make data scientists’ lives easier. Even if your data scientist is pulling out gobs of worthwhile data stories, a content partner who speaks analytics can translate those stories to editorial actions more clearly. Writers and data folk working together make beautiful collaborations. Add in a visual designer and *chef’s kiss*!
Understanding analytics is a fabulous bullshit detector. When your boss says something like, “Let’s make more of the clickbait stuff,” analytics knowledge helps you demonstrate why that’s a bad idea. When Kendall Roy eliminates your digital media property with Your bounce rates are low, you can combat him with better data and the knowledge that low bounce rates have never been an indicator of failure (knowing full well it doesn’t matter because his dad told him to but at least you tried).
There is a scene before this one where Kendall talks about bounce rates and metrics and he clearly doesn’t know his ass from his elbow and it drove me nuts because it was so realistic.
And when your advertising sales team says something like, “We need to keep the Luxury content! It’s getting so many pageviews and Richie Rich loves it,” a skilled content analyst can counter, “But the Luxury content actively drives regular readers away, while the LowFi content is attracting more faithful audiences.” (A bit of an ideal-world hypothesis, but really: very few people look at the luxury brand content.)
Once you understand analytics, the smoke and mirrors around digital marketing and, really, the entire advertising industry will start looking like gas and glass.
Most of us highly verbal folk dropped math courses from our schedules as soon as we could in favor of more creative classes. Knowing analytics helps exercise your math muscles and generally become more rounded. You’ll get better at reading all data — like COVID-19 graphs and polling trends. And GA is not even that much math, it’s mostly just playing with pretty graphs.
Massage your own ego with the knowledge of how well your content performs.
Come to terms with your own limitations with the knowledge that your content doesn’t always perform as well as you would like it to and maybe you need to spend some time on your website.
With that said: here’s the skinny on changes from the old Universal Analytics and the new Google Analytics 4.
Hold up! Are you new here? You can
10 big changes in Google Analytics 4 that publishers and content analysts should know
More focus on users and sessions, less on traffic like pageviews. Daily, monthly and weekly average users are now calculated for you, so there’s not as much custom setup involved.
More info on customized audiences and lifetime value. Audiences have been available in Universal Analytics for a few years, but the GA4 interface provides more information for comparing audiences and finding the true value of loyalty.
The brand new engaged user metric highlights users who have either been on a page for 10 seconds, visited more than two pages, or converted. Sure, 10 seconds is a low threshold for anything at all, but it’s better than the nothing from before! With this metric, GA4 is absolutely trying to compete with Parsely and other engaged-time analytics tools.
The engaged time metric isolates only the time users are actively reading the website, which is more accurate than straight session duration. The new metric eliminates tab dormancy time, a metric I just invented that describes how much time websites spend sitting open in tabs, waiting to be read or closed out. (Anticipation time? Wishful tabbing duration? I would actually love a breakdown of how much time users spend actively reading versus letting tabs gather dust, but Google’s not going to invest in that human interest story.)
Content-specific events like scroll depth tracking are now automatically tagged, so you don’t need to set up custom event tags in Google Tag Manager.
The new setup also automatically tracks clicks on links off-site, which will be massively helpful for understanding user path and experience. Before this event metric was difficult to set up and use properly, so I’m stoked about this feature.
We’ve got consent mode! Or at least, we will eventually have consent mode. I haven’t seen the best way to implement setup mode, but this interview with a Google rep in SearchEngineLand says that it’s coming. Easily remove user data when requested and comply with the law, or just be a good web citizen in general. Consent is fantastic, and if a marketer or advertiser thinks customer consent is a roadblock to success, please don’t give them your business.
More focus on real time metrics than in Universal Analytics. GA4 moves real tie metrics to the forefront. In my experience, real time recommendations distract from long-term vision and distract from solid content strategies. Finding good data in real time is similar to refreshing your Twitter feed every 30 seconds and expecting to see something genuinely life-changing.
Also real time analytics are part of a surveillance mindset and contribute to bad content practices like trendhopping or newsjacking that make the web a dumpster fire of halfbaked misinformation and thin content. Websites shouldn’t be 24-hour tv news networks! What is “real time” anyway?
Gripes aside, I’ll take my historical weekly and monthly content data, thank you, but GA4 has more real time metrics if you want to drive yourself and your staff crazy.
Deprecation of metrics like bounce rate! The elimination of the organic keyword report, which has been useless for 7 years! Lots of consolidation of old reports into a more organized, meaningful fashion.
All website health metrics have moved over to Search Console for good. Site speed is no longer available in GA4, probably for everyone’s benefit. You’ll still be able to see device- and browser-specific dropoff points in GA4, which is helpful for diagnostics, but for the most part: Search Console will be the technical website performance go-to in the future.
What are the long-term implications of the GA4 interface?
Google Analytics is the standard for web analytics, and we’ll all eventually follow the slight shift in terminology and reporting because the Googadook lives in our house.
Because Google Analytics is free — and that’s really where they getcha — it’s the single best benchmark for comparative performance on the web. I trust that Google Analytics numbers are collected the same from website to website. They’re more reliable than ComScore numbers as far as I’m concerned.
Although it’s in the midst of an antitrust investigation, GA4 is clearly trying to ape Parsely’s and Chartbeat’s focus on actively engaged users and real-time content analytics. That’s not to say GA4 will replace either of those publishing-focused tools, but when you’re looking to cut costs, if you have two sets of software that measure almost exactly the same metrics, and one is free, well…
The new interface puts equivalent weight on real-time, content lifecycle and audience analysis, which will likely be a boon to most devoted marketing and analysts in the long-run. The cookieless setup and privacy compliance are also excellent developments for the web in the long run.
As a publisher and consultant: the most immediate advantage is automated events tracking. No more hiring a custom analytics consultant, creating a detailed taxonomy for outside link clicks, or deciphering what qualifies as a non-interaction event for basic tracking tasks: the new GA tags your site right off the bat.
But after setup, the out-of-the-box advantages are not as apparent. Custom reports need to be constructed. You still need to know a fair number of specific definitions and customized terms to make GA4 make sense.
Straight up: I don’t think the new interface will be easier to read for analytics beginners and small business owners. Training on how to use the new interface and setting up custom reports to hone in on content performance will be necessary.
You should start collecting data for Google Analytics 4 ASAP.
Set up Google Analytics 4 now so you can start collecting data. The new interface doesn’t analyze retroactively, so it will only collect data beginning on the day you set it up. You can return to your custom reports and actually using GA4 later, but just get the data in there as soon as you can.
For a long time — probably two years, I’m guessing — you’re going to be toggling between reports in Universal Analytics and Google Analytics 4 as your teams get up to speed on the new setup. I don’t think you need to learn the intricacies of GA4 just yet, but again: start filling it with data and it’ll be there for you when you’re ready to switch.
In the meantime, if you don’t understand Google Analytics already, now’s your time to start anew.
Some important concepts that are carrying over include metrics, dimensions, events, sessions, channel, source, medium, and scroll depth. Spend your time learning what those mean in the context of a website, and you’ll be in a good spot to get to know GA4 once data populates in your new property.
A special offer for Content Technologist subscribers
If you are a brand or publisher and receive thousands of visitors per month, I’m offering a limited number of Google Analytics 4 setup packages at an immensely reduced price.
What’s in it for me? I learn the tool and get more practice.
What’s in it for you? You get deeply discounted fresh new content analytics.
This offer’s limited to 5 applicants from corporate brand or publisher side. Ideally you regularly publish web content, are interested in growing or maintaining an audience, and attract at least 5k visitors on a monthly basis. (More than that is completely welcome.) Sorry, agencies and freelancers, it’s not for you, but watch this space for more in the near future.
Content tech links of the week
Ernest Wilkins’ Office Hours is a delightful marketing-oriented newsletter that’s the inbox highlight of my Monday. This week Wilkins dives into the Share of Culture metric, which I can absolutely get behind (it’s paid, but worth it).
Developing for the semantic web, via Smashing Magazine. Semantic content will likely be the future of web content, as long as we can get our poop in a group.
How do we fix journalism’s trust crisis? Two very different frameworks are out there, in this study summary in Nieman Lab. (Guess which one I advocate! You’re right, it’s engagement journalism!)
The W3C is working on new standards for Content Aggregation Technology that don’t involve Google, Facebook, or Apple. Cool! Do I understand them? No! But I’m trying to dive into this report.
You should know: Google is providing personalized keyword search data to law enforcement, which is not ok on so many levels. Do I know what to do about it? Not yet. But maybe use Duck Duck Go if you’re going to do research for crimes.
As always, please
if you find this post useful.
See you next week for some zombieeeeessssssssss