“Arguably, the number of real visits it fails to record is now so great as to render the tool's hit-counting prowess worse than an educated guess.”
One decade ago, any suggestion that something as eminently sophisticated as Google Analytics could be deemed unfit for purpose, would have categorised the suggestion-maker as something of a fruit loop. Offered free of charge, by a true leader in the tech genre, Google Analytics had become the gold standard in web traffic monitoring, and had found its way onto the majority of websites you or I would be likely to use.
Of course, GA does not only deliver data to the admins and owners of the websites who use it. It also delivers data to Google. And based on the “knowledge is power” principle, that phenomenal scale of insight into public behaviour has given Google an incredible advantage as a force in online advertising.
Google is thus desperate to maintain its standing in the analytics genre, but of late, things haven't been going quite as smoothly as the marketing world has been led to believe. In this post, I'm going to make the one-time sacrilegious assertion that Google Analytics may no longer be fit for its original purpose, and I'm going to explain why…
If you use a blogging platform with its own, built in analytics, and also add Google Analytics for a more in-depth study of visitor stats, you'll almost inevitably see a discrepancy in the two hit counts.
Back in the mists of time, the discrepancy was not that great. The blogging platform itself - let's say Blogger - would normally display higher traffic totals than Google Analytics, but perhaps only by a margin of up to 10%. The difference was attributed to hits from bots, referrer spammers and the like, which Google Analytics had the ability to filter out, but Blogger did not.
However, today, you may well be seeing a much larger discrepancy, with Blogger, in some cases, reporting 30%+ more traffic than Universal Google Analytics. So have bots gone wild across cyberspace, or is the increasing discrepancy down to something else?
I recently explored this whilst researching a recent article on businesses losing sales due to the analytics blind spot. And the widening gap is not the doing of bots. It's actually due to the increasing public awareness of privacy tools, the use of browser incognito modes, the installation of ad-blockers, etc. Incognito mode blocks cookies, and Universal Google Analytics (UGA) requires cookies to run. So if someone visits your site in incognito mode, and you're using UGA to monitor stats, that visitor won't show up. Likewise if a visitor's ad-blocker zaps out Google Analytics, their visit won't be reported.
The use of these privacy tools and measures is hugely more prevalent today than it was ten years ago, so here in 2020 the discrepancy between Blogger's native visitor count, and your UGA visitor count, can be enormous. But unlike in the old days, when Google Analytics was considered the wiser reference, Blogger is now providing the more accurate total.
You can test this yourself, by accessing your Blogger blog via different browsers, with different settings. You'll quickly see that while Blogger records your visits, UGA very often fails.
Google has made an attempt to address this problem with the roll-out of Google Analytics 4. GA4 is a completely new version of the popular free metrics suite, and it doesn't require cookies to fire. It also offers site admins and owners deeper analysis of visitor behaviour. Much deeper analysis, up to the level of recording page-scrolls, link click events, media engagements, etc. So that's the problem solved, right?
Unfortunately not. What the entire data-desperate tech business consistently fails to grasp is that the precise reason the public are adopting privacy tools in far greater number, is that they are SICK OF SURVEILLANCE. If you are trying to address a public backlash against surveillance by building in MORE surveillance, the eventual result is going to be that stats counters can't record any visits at all.
Big tech, and the smaller marketers who follow its lead, have to realise the difference between monitoring the success of a piece of content, and recording every last vibration of each visitor's fingertip. And they have to understand that the latter is not necessary in order to achieve the former.
Go to Twitter. Find a stalker. Is he popular? No, he is not effing popular. He's only ever had one follower, and that was the Police. Stalking is deemed unacceptable right across society. So why would you think that you can set up a website to stalk people, and they'll tolerate it?
I get that these increasingly aggressive tools exploit a combination of public ignorance and convenience to evade counter-measures from web users. But you can't keep people in the dark forever, and we've already seen that people will act to protect themselves from online dystopia once they become aware of it and the preventative tools themselves embrace convenience.
So we should not see Google Analytics 4 as a solution to the increasing inaccuracy of website hit counts. Even though it doesn't require cookies to operate, it's still highly vulnerable to in-browser block listing, and that vulnerability will inevitably upscale as more and more sites use the new, aggressive spy functions irresponsibly and word gets round. There are already stock browsers, such as Brave, that default to blocking privacy invasions. It would only take one product like that to go mainstream, and Google Analytics' hit stats could be rendered completely worthless. And GA4 is making that more likely, not less.
With Google Analytics 4, we're seeing extremely powerful spying tools being novice-packaged and placed into the hands of any random stalker with a Google Account. That is not going to end well.
So returning to the question in the post title; is the old, Universal Google Analytics still fit for purpose? Not if you see it as a means to establish an accurate visitor hit count, no. Arguably, the number of real visits it fails to record is now so great as to render the tool's hit-counting prowess worse than an educated guess. And Google Analytics 4? It's more accurate than UGA because it records incognito browser hits. But it's not accurate, because it doesn't record visits from people who use the ever-increasing number of privacy tools that block it. That situation can only get worse.
The purpose of Google Analytics has been refocused away from counting hits, and onto peeping at individual visitors' behaviour. I suppose moving the goalposts is one way to preserve the claim that GA does its job. And whilst there are visitors who don't block it, it will remain fit for that modified purpose. But big tech's obsession with “data points”, and extracting evermore dystopian detail on individual users' behaviour, has already killed off GA's ability to deliver even a moderately accurate hit count.
One would think Google might learn a lesson from that. But on the evidence of GA4, one would be wrong.
It's sad to think that the de-motivational drops in hit counts rendered by Google Analytics have probably resulted in some really interesting bloggers giving up, while tiresome marketers with gruellingly boring content persist in driving forward the data point economy that caused the problem. If you blog, do not accept Google Analytics hit counts at face value, and always assure yourself that you get a lot more traffic than GA records.
My own tests have also shown major voids in genuine referral accounting - particularly from social media. For example, hits known to be from Twitter being recorded as direct visits. So be positive, people. Keep writing. Keep publishing. And if your blog has a native stats counter, definitely place more faith in that than you do in Google Analytics. Unless you have a fully enclosed login system, the age of being able to accurately measure page hits is pretty much over. And for that you can thank the advertising companies who didn't, and still don't, know where to stop.