Google Analytics Checklist 2020
Updated: Jan 4
Attention, Google Analytics users — there is a chance the data you're evaluating to assess the performance of your website isn't clean. What I mean by that is, the data you are using to make your most important digital marketing decisions is potentially inaccurate.
(Often, when we take a look under the hood of websites, data accuracy in Google Analytics is an issue.)
Of course, your Google Analytics may be fine. But if you're not sure, you need to make it a priority to either confirm your data in Google Analytics is fine, as-is, or resolve any issues you may uncover.
I mention this because the very last thing you want to do is make important business decisions about your website based on data that isn't painting a true picture of how many people are visiting your website, and what they're doing on your website once they get there.
Is this a big deal? Absolutely. Even if you have HubSpot, you need your Google Analytics in tip-top shape from a set-up perspective, if you want an accurate picture of your website's performance, and here's why...
Why is Google Analytics so important?
Going back to basics for a moment, Google Analytics gives you full throttle control over your website data, allowing you to create overhauled marketing funnels, audiences, reports, and highly customized views for everything everyone does on your site… ever.
It’s amazing what you can do from a tool that is 100% free for every website owner out there on the internet.
When Google Analytics is properly set up, you'll gain invaluable insights into the journey a user takes when interacting with your brand online. You can learn how your users are getting to your site, which pages are magnets for leads, where users are the most likely to bounce away from your site, why they decide (or decide not) to give you their business, and more.
Even though Google is notoriously tight-lipped about how many sites are using Google Analytics, the estimate is more than 50 million worldwide. On top of that, Google Analytics is widely regarded as the most accurate tool for site data analysis. (You’re getting information directly from the source after all.)
So, regardless of where your site is built or what other tools you are using to track your users’ behavior, Google Analytics is a must-have.
That's why this question of “What my Google Analytics data isn’t accurate?” is such a critical one to address.
If that is the case, then you could be making unbacked recommendations to the rest of your team as to what efforts are working and what you should be trying next. It would be like walking into your next marketing meeting and making decisions on a gut feeling.
How can data discrepancies in Google Analytics create issues?
Let's imagine for a moment that you're evaluating the top sources that drive your website traffic — where your users are coming from — and you find that Bing is your highest contributor to your site’s traffic.
As a result, you might move efforts into optimizing your presence there and reduce the work you are currently doing to optimize your performance in Google search results.
Imagine telling your higher ups: “I've got a feeling that we need to go all in on Bing.”
Now what if that isn't the case? And later on, you find out that Google does end up sending you most of your traffic, and Bing isn't nearly the driver of visitors you thought it was?
You might have just choked out your best performing channel.
“Wait a minute, Dan. Why do I need to worry about this if you just said Google Analytics is one of the most accurate sources of website data?”
Google Analytics is only as smart as you make it
Well, this is the heart of the problem I mentioned at the start of this.
Yes, Google Analytics is powerful and seemingly all-knowing. But that's only true if you set up your data sources properly because, out-of-the-box, Google Analytics is... well, a little dumb.
It’s not its fault though.
Much like a new, adorable puppy you need to potty train, Google Analytics needs to be taught what to do — and more importantly — what not to do.
You need to tell it what data you actually care about and what it should be removing from the data set it is showing you. Filters like employee IP addresses, locations, and self-referrals all need to be set up when you install Google Analytics, otherwise you won't see the true website visitor data you're looking for.
So, for all those site owners, webmasters, and wayward digital marketers out there, here is a checklist of items that you absolutely need to address in Google Analytics before you can even think about making decisions based off of your reporting:
Tag Setup: Is Google Analytics installed properly, and in a way that doesn’t conflict with your site’s performance or other tracking tools you might be using?
Security Protocols: Which version of your site are you tracking? Are you sending users to the most secure version of your site? You have to tell Google Analytics which one you care about.
Custom Goals: In a perfect world, what actions do you want your users to perform? Giving this information to Google Analytics enables you to view what steps led up to that action.
Engagement Audiences: Are you bucketing your users into different groups based on the last time they were on your site? Again, Google Analytics needs to know this, too.
Filters: What data do you care about? What do you need to make sure isn’t included in that data? Telling Google Analytics what to remove can significantly improve the accuracy of your numbers.
I know, that was a lot of stuff. So, let's go through each one.
Let's start with overall tag setup
The question we're going to address at this step is, simple:
Are your Google Analytics tags properly tracking all users across all pages across all of the domains on your site?
We need to make sure that the entire site is being tracked and controlled within your Google Analytics account. Leaving pages or subdomains untracked can lead to users going unnoticed, which can include vital steps of their buying journey.
I use Google’s own Tag Assistant to make these settings are where they need to be, because you sure as heck can't count on the Google robots to tell you what’s right and what’s wrong. (By the way, Google Tag Assistant is a free Chrome extension that records your site loading and scans for all Google tags installed on a given page.)
Google Tag Manager hosts all the tracking tags and scripts you have on your site in one “folder.” It lets you make changes and add more scripts as you need, while relieving your site from hosting all of that code. Meaning, instead of loading multiple scripts (tracking, custom plugins, etc.), all of that custom code renders on your site in a single lean script.
This, in turn, reduces page load time and improves your site speed, which we've talked at length previously about why that is such a big deal.
Honestly, I could talk all day about how Google Tag Manager and Google Analytics party together (and I probably will), but for now that’s really all you need to know for the context of this article.
Another thing to check out while in this setup phase is your remarketing tag.
This is a setting within your Analytics account that lets your site cookie users so you can advertise to them later.
If you've ever abandoned your shopping cart on the Old Navy or GAP websites, you have been the target of remarketing. Because whatever was in your cart will follow you around online (Google, Facebook, and so on) until you finally cave and buy them.
Within your Property Settings under Tracking Code, simply turn on remarketing.
Even if you don’t do any kind of remarketing in Google Ads, you should still have this on just in case. This way if you do want to get into some retargeting of users, you don’t have to wait. You're already ready to stalk your website visitors across the internet.
While you’re in there make sure your Google Search Console and Google Ads accounts are linked. This lets you view Google Search Console data within Google Analytics — all your numbers will be under one roof — and also lets you make use of that remarketing tag we were just talking about within Google Ads.
With all of that done, congratulations!
You just made a giant step into making sure your data is clean. Addressing these items should give you a significant accuracy improvement.
But we're not done just yet...
HTTPS security protocols are a critical SEO variable
A few of you might be saying at this point, “I already have that. My whole site redirects from http:// to https:// automatically.” (If not, this article about SSL certificates and SEO is for you.)
That’s all fine and dandy, but there is a chance that’s not what you’re tracking in Google Analytics.
So, in the View Settings panel, you have control over your site’s security protocols.
If this setting is pointing to http and not https, you are only gathering data from people clicking links or manually entering http into the address bar.
⚠️ Again, this is a bad thing. ⚠️
If anyone on the internet is linking to your site via a http URL, only the redirect is being counted. Which means all of the organic traffic you are getting from Google isn’t being tracked accurately.
You only care about the pages users are going to with the correct protocol. You’ll find that by changing this setting, your overall traffic numbers will change.
This can be for the better or worse — fewer or more sessions — but regardless of what happens you will now have a better representation of the real sessions you are getting on your site.
Goals, what you want your users to do
Now that our settings are in tip-top shape, it's time to "potty train" Google Analytics by teaching it what we care about. We do that by setting goals.
If you’re using HubSpot or another CRM to capture lead and contact data, you know which goals (conversion actions) are valuable to your brand. This could be form fills like a contact us or request a demo — or, for you e-commerce folks, it’s purchases made on your site.
There could be a lot more things you want to track as well, such as an “Add to Cart” goal or a video view. The point is, you know what actions you want users to take in order to give you their business.
Google Analytics, however, has no idea what actions you want visitors to take... that is, until you tell it.
By heading into your View settings under “Goals” in Analytics, you can set these up.
This goal will fire when someone views the URL www.mywebsite.com/thank-you
By telling Google Analytics what these goals are, you can get a bigger picture of your visitor and lead activity, outside of what your CRM will tell you. We can get information on how they got to this page (organic search, paid search, social media, etc.), and what pages they viewed before they converted.
You can then analyze that data to see where you can reduce friction and increase the desired user activity on your site.
You’ll also notice some more options under “destination” — equal to, containing, or regular expression.
When it comes to destination goals like a thank you page, “Begins With” is often the best option because it won’t exclude users visiting the page with a tracking URL from an ad or wherever else tracking URLs are used.
Quick Note: There are a lot more complex goals you can set up — button clicks, video clicks, etc. These require a bit more coding knowledge as well as a Google Tag Manager container, so we will revisit those down the road.
Hot (or cold) audiences
What we are talking about here is “Engagement Audiences.”
Engagement audiences bucket your users based on when they were last on your site. Every user gets cookie'd when landing on a page, and as time goes on they move into their respective buckets.
"Why do I care about these numbers?"
These audiences are best applied when you run Google Ads. You can target your “hot” users (those who have been to your site recently) differently than you would someone that hasn’t been there in over two weeks.
But even if you aren’t using Google Ads, these engagement audiences can give you a lot of insight into how long your sales cycle actually is.
Additionally, these buckets can look differently based on your business goals.
The example above is probably best suited for quick purchases, like for an e-commerce company, where staying top-of-mind is the key to purchasing. A B2B company with a longer sales cycle might choose to bucket their engaged users into 30, 60, and 90 days instead of shorter seven, 14, and 30 days.
The great part is, there isn’t much work to do to get these set up. These audiences are already included in Google Analytics, they just need to be turned on.
Good job, Google.
Employee traffic to your website is something you need to filter out of your Google Analytics data.
Internal employee traffic and referrals only artificially inflate your numbers and skew audience location demographics — sometimes significantly. If you do care about internal traffic, you should probably create a new view within Google Analytics that excludes them so you can see the difference between the internal and external audience data sets.
If you don’t know if your traffic is being filtered, head to your view settings again and click filters. This is what you don’t want to see:
If this is what your filter menu looks like, don't stress out — getting them set up isn’t too intense. Gather the IP addresses for all the employees, offices, or vendors using your site and hit “ADD FILTER.”
Now it's time to brainstorm
Woohoo! Now that your data stream is cleaned up and your numbers are all spick and span. That means it's time to pat yourself on the back. You’ve just unlocked the full potential that Google Analytics has to offer.
Sit down with your team and talk about the data you really want to see. With Google. Analytics you are given up to 20 views for free. Each of these views can have different sets of filters, audiences, and goals.
Want to look at what people in California are doing on your site? Set up a filter to only include traffic from California, and presto!
Interested in how your blog subdomain is performing compared to your main site? Easy enough — simply track traffic from the domain or pages you are interested in. It can be a lengthy conversation, but regardless of what you end up deciding you can be confident that your data stream is clean, accurate, and ready to be analyzed.