A few weeks ago, I got a question from a client that I haven’t heard in a while: “Can you measure a website?” It made me realize just how much I take our understanding of digital metrics for granted. As people who are living and breathing this stuff every day, it’s easy for us to forget others out there don’t know the intricacies of Google Tag Manager and Zapier connections.
So, we’ve decided it’s time for a data refresher. How should you be measuring website efforts in the era of AI?
Website Metrics
How can you tell if a website is performing well? Great question. For this, we’re going to rely on Google Analytics 4. First thing’s first. If you don’t have that installed on your site, do it now. Seriously. Stop reading. Go add it.
Done? Good. We can move on.
Obviously, every site is as different as the business it’s built for. That means what matters to you will be slightly different than the metrics for someone else. Still, here are some common ones to keep an eye on:
Key Events
Weird name, right? These are the important things people complete on your site, like completing a purchase or signing up for a newsletter. They’re the actions that move the needle. Google automatically includes purchase events, but you might need to get a little techy to set up the others. Reach out to your team if you don’t have these set up properly. They’re the lifeblood of your site. After all, you want to know how many people are engaging with you in meaningful ways, and even more importantly, when they aren’t.
Acquisition Primary Channel Group
These are the ways people find your content. Google breaks them down into a few standard sources:
- Direct: Users type in your URL or visit your site through a bookmark.
- Organic Search: Users find you on Google, Bing, or another search engine.
- Paid Search: Users come to your site by clicking on Google Search Ads you’re running.
- Organic Social: Users find you on social and click on a link to view your site.
- Paid Social: Users come to your site through a paid social campaign, like Meta Ads.
- Display: Users click on a Google Display Ad.
- Cross-Network: Users click on a Google Performance Max ad.
- Referral: Someone visits another website, which includes a link to yours. They click on that link.
- Email: Users click a link in an email or newsletter.
- Unassigned: Google has no idea.
There are a few other sources, but let’s keep it simple with the above. These help you determine what your traffic drivers are, where you should focus your efforts, and even potential partners. If you look at your referral sources, you can see who’s already linking to you and determine how to collaborate in the future.
Let’s talk about AI. Platforms like ChatGPT are great at serving up links to sites. However, Google isn’t consistent about how it classifies that traffic. We also know Google isn’t properly measuring the traffic coming from those sites, which has been driving the digital community wild for a while now.
While writing this, I went to a random client’s Google Analytics 4 instance and looked up all traffic from ChatGPT in the last month. Google split up the 118 sessions between Unassigned, Referral, and Organic Search. To be clear, all of those sessions are from ChatGPT…Google just sort of arbitrarily shoved them into different categories. Because of this, when you’re looking at your Primary Channel Groups, make sure to also compare them to the session source/medium. That’ll include the group and the URL the traffic came from, so you can see if Google decided your AI traffic is social for some reason.
Sessions, Engaged Sessions, and Engagement Rate
These are the main dimensions you’ll see in reports. A session is any time a user visits your site. If the user visits the site and then just goes idle or clicks away, the session will eventually time out. Then, the next time they interact, it’ll count as a new session. For example, if I open our site and click on a blog, that’s one user completing one session. If I come back an hour later and visit our careers page, that’s one user with two sessions.
Engaged sessions are a little different. These are people who come to your site and then do something. A session starts as soon as the site loads, so if someone clicks on a link to your site, looks at the page, and leaves, that doesn’t show any engagement. If they click on another page within your site, it shows they’re interested. They’re engaged.
Google uses the total of engaged sessions compared to sessions overall to determine your engagement rate. In other news, if you have a 50% engagement rate, it means every other person who visits the site interacts with it. That’s not too shabby!
Sidenote: if you spent any time talking about metrics a few years back, you probably remember discussing bounce rate. That was how many sessions come to the site and immediately click away. While that was helpful, it was measuring the exact opposite of what you wanted—the number of people who cared. So Google got rid of bounce rate and replaced it with engagement rate. Think positively!
Transactions
If you have an e-commerce site, this is music to your ears. Google measures the purchases coming through your site automatically. However, the totals can be a little off compared to your POS, depending on how your Google Analytics has been set up.
Still, even if the dollar amounts aren’t 100% accurate, you can still recognize trends and revenue drivers through Google’s metrics. This can help identify slow seasons or days with record sales and then tie them back to the traffic source. Did that viral social media post actually result in sales? Now you know.
Helpful Tools
We love data, but we love clean, accurate, easy-to-access data even more. Here are some tools that can help even the least tech-savvy of you get a handle on how your site’s performing.
Analytics Advisor
If you’re like most people, you log into Google Analytics, and your eyes glaze over at all the charts. It’s a difficult platform to navigate. In fact, there are legitimate certification courses just on how to use it effectively. No business owner has time for that.
Enter the Analytics Advisor. When you’re logged in, you’ll notice some sparkles in the top right of your dashboard. If you click on this, it’ll open the Analytics Advisor. This functions like any other AI chat you’ve used before. You can ask it questions about your data, and it’ll give you the answer right then and there.
Looker Studio
Google Looker Studio is a platform that visualizes the data you want to see all in one place. This one takes a little bit of experimentation to learn, but once you do, you’re set. Spend an hour building a report today, and it’ll update with live metrics whenever you need it. Plus, Looker Studio contains a lot of templates you can use to make your report instead of starting from scratch. Just connect your data source, and you’re up and running.
Google Tag Manager
If you have a tech-savvy person on your team, this is a must-have. Google Tag Manager allows you to set up your account to track additional actions instead of just purchases or if someone visits a specific page. Sure, it can be a beast to set up, but just like the rest of the tools we’ve talked about, once it’s properly built out, you’re set up for success.If you have a website but aren’t actively looking at how it’s performing, you’re leaving money and opportunities on the table. We know small business owners are told to keep an eye on everything out there, but if you can promise to set aside a half hour every month to check in on your web metrics, you’ll find a wealth of information in data you already had…you just never looked at.
