You've got a site, and you've made sure Google Analytics (GA) is installed.
Now, you look at various metrics and reports in Google Analytics...once in a while. It's interesting. Maybe even useful at times. But not everything is as easy or clear as you'd like it to be.
Let's face it: you've got some questions when it comes to tracking metrics in GA. And some stay unanswered.
- "Why do I have multiple pages with the same URL showing in GA?"
- "What's this "fbclid=123456789" part that seems to be different for every page? How can I get rid of it?"
- "What are those facebook.com, m.facebook.com and lm.facebook.com in my referrals?"
- "Is the Sessions metric the most useful and reliable one I should look at?"
- What is the answer to the Ultimate Question of Life, the Universe, and Everything?
The last one is easy: it's 42.
But what about the other questions? Those are all symptoms of real problems. To provide some clarity for those, let's go through each of them one at a time.
Does your traffic look like this?
Yes, they're all the same page, but with a fbclid parameter added to it. But you didn't add this part, and it's not telling you anything helpful.
So what is it?
This is a parameter added to links shared on Facebook, but it's not useful for your Google Analytics at all. In fact, quite the opposite because it dilutes the metrics of a page into multiple "pages."
This is, to some extent, breaking your data.
What's the solution?
It's actually pretty simple. There is a configuration in Google Analytics to remove query parameters like this one.
First of all, I highly suggest that you have a test view on which you can add this configuration first, and then, later on apply it on your master view (and please, always keep a raw view, with no filters). The details as to why and how to set this up Test, Master, and Raw views are a bit out of scope of this articles, though.
Here are the steps to fix the fbclid problem:
- Click on the "Admin" link at the bottom left of your Google Analytics screen:
- Click on the "View Settings" of the third column:
- Add fbclid to the box under "Exclude URL Query Parameters":
- Save that, and you're done!
If you've applied this to your test view (recommended), come back in a few days and make sure that your page views doesn't have fbclid in their URL. Note that this will be in effect only from when you do the change. It is not retroactive. If you were to do that and on the 3rd of the month, it means data starting on the 4th should not have fbclid.
2. What are l.facebook.com, m.facebook.com and lm.facebook.com in my referrals, and what do they mean?
Their meaning, just like the Ultimate Question, is not THAT important. Instead, this is a symptom of a bigger issue.
But let's go through it because I love to help people understand these kinds of things. I also think it's a good idea to develop your Google Analytics symptom analysis skills here.
There was a time where traffic from Facebook came only from facebook.com and m.facebook.com. Back then, it was easy to explain what those two referrals were: one was for desktop traffic, and one was for mobile traffic.
There was still the big question, though: which post/ad is it coming from? More on that in a minute.
Nowadays, you'll likely see something like this:
We know what facebook.com and m.facebook.com mean, but what about l.facebook.com and lm.facebook.com?
These are subdomains used by Facebook to protect their users. All links shared on Facebook go through a redirection system that also detects phishing or other types of fraudulent links.
Have you ever seen a warning when clicking on a link saying that Facebook thinks it might be a fraudulent link or fake news? This is the redirection service, that they also call Link Shim, doing its job.
Often times, it's doing everything silently, so you don't ever see anything.
So when you see those two domains in your referral, it just means that the person clicked on a link that went through the redirect service.
Now, let's take a step back. Why are those classified as referrals and not social or advertising traffic?
This is the big problem.
Because Google Analytics can't assume how to track this traffic for you, you need to explicitly tell GA how to deal with it. It has no way of knowing that it is post X or a specific ad or campaign.
The solution to that is easy with the use of UTM parameters added to your URLs. UTM parameters tell Google Analytics which source ("facebook"), which medium ("social" or "cpc"), and which campaign the page view should be classified with.
It's a lot simpler than it might sound at first. If you're not too familiar with UTM parameters and how to use them in this context, I highly suggest that you go and read our article about UTMs. This post tells you what they are and how to use them the right way.
Once you start using UTM parameters for everything on Facebook, you'll be able to know which traffic is coming from organic posts, advertising, and which post or ad. That means you'll be able to calculate your return on investment (ROI) for all your social and advertising efforts.
3. I am only looking at Sessions as a performance indicator...Can I do better and track signups or sales?
Yes, and you should.
How many brick & mortar stores do you know that track how well their business is doing by counting the number of people coming in their store?
Spoiler alert: none.
They measure it with revenue and profit. Why wouldn't you do the same thing?
For that reason, I find Sessions very misleading for most businesses. It's an interesting metric when you've got nothing better...but in the end, it's just a vanity metric.
And that's why we published an article called What KPI Metrics Make a Valuable Report for Small Business. That post covers this topic (and tells you what you really should be tracking).
Why is it so important?
Because without a proper KPI, you won't be able to calculate the ROI of your advertising campaigns or any marketing efforts. You'll operate completely blind. Or worse, you'll assume that more visits mean more sales.
That means you might put more money into advertising campaigns that bring more people, sure, but not more buyers.
Let me help you identify what you should be tracking as your main KPI:
Why did you pay thousands or tens of thousands of dollars on a website?
The answer to this question will give you your main KPI. It's that simple.
It's unlikely to be "just for fun", "because I needed" it or "because my competitors have one". If so, why? Why does your competitor have a website?
In your case, just like the vast majority of people, it's very likely to sell something or find leads for whatever you sell. Let's dive into that thought in more detail but broken down by category.
If you're an ecommerce store with multiple products, you should make sure you have Enhanced Ecommerce setup. This will give you your sales, products sold, and all sorts of useful metrics, in addition to being able to track sales as your KPI.
If you use an online platform like Shopify or Snipcart, you should already be covered.
If you have your own site with Wordpress or another CMS, it likely has a plugin that is simple to install and adds this functionality for free. For example, WooCommerce has both a free and a paid option with additional features.
If you have a fully custom platform, your developer or agency should be able to add it for you, but it will likely take them some time (and cost you some money).
Software as a Service
You should track your signups: both trials and paid.
And then you can configure a goal. Or you can do that with a destination/URL if you have some sort of "thank you" page that your new subscribers only see once.
Consultancy or high priced industrial equipment
You probably have a lead magnet, a for to give price quotes, or something similar. The submission of this form should probably be your KPI.
It's even better if people can buy your time or equipment online. But if they can't, your lead form will be perfectly fine.
Similar to Software as a Service businesses above, you'll be able to configure a goal based on a destination URL (a thank you page) or with an event. If you have a "thank you" page, use that to configure your goal.
If you don't fall in the "I sell or get leads online", you sadly fall in the "it depends™️" camp. Maybe you sell advertising space (like news sites), or that your mission is to inform the public (like many government and public entities). Maybe Sessions are the right KPI; maybe it's the time on page or a complex mix of scroll percentages and actual time on page.
It really depends and is out of scope for this article.
That said, if you need help with that, you can contact us because we offer consultancy services.
Once you have Enhanced Ecommerce or goals setup, you can now use that as your main metric in Google Analytics.
You'll now be able to see your conversion rate (percentage of visits turning into sales), sales per campaign, or per social media, for example.
And this is how you will get the most of Google Analytics.
The short version is: you'll understand what affects your bottom line the most and what has the highest ROI.
When you have a fever, the doctor doesn't examine you only to say, "Welp, you've got a fever."
The fever is just a symptom that helps the doctor find the solution.
You need to do the same with Google Analytics. Finding an issue is the first step.
Before choosing a solution you need to identify the root cause, and the right solution.
In the case of the Facebook referrers, for example, the root cause is their link redirection. But Google Analytics isn't able to infer anything from those referrers, so you've got to tell it how to classify this traffic explicitly.
The goal behind the solution to all those symptoms here is to be able to reliably track what's working (or not) in your marketing efforts, which you can do with the right KPI, UTM parameters, and by fixing erroneous data.
The final step is to use that data to get clear KPIs and move your company forward.
The easiest way to do that is with Metrics Watch to receive email-based, mobile-first marketing reports with all your important data from Google Analytics (and other platforms). We offer a free 14-day trial, no credit card required.