[Ultimate] Guide To Google Analytics For Shopify
You probably already have your ecommerce store Google Analytics account setup – because that’s what you’re suppose to do. But I’ll bet that you’re using it very ‘passively’; furthermore, I’ll bet that you’re completely ignoring or misreading what you see and not getting the most out of it. That’s a bold statement but it’s most likely true. But this post will change all that… Over the next 15-20 minutes that it takes you to read this guide, you’ll understand how to set up Google Anlaytics better than 90% of other Shopify stores and how to use it to actually grow your business.
You probably already have your Google Analytics account setup – because that’s what you’re suppose to do.
But I’ll bet that you’re using it very ‘passively’; furthermore, I’ll bet that you’re completely ignoring or misreading what you see and not getting the most out of it.
That’s a bold statement but it’s most likely true. But this post will change all that…
Over the next 15-20 minutes that it takes you to read this guide, you’ll understand how to set up Google Anlaytics better than 90% of other Shopify stores and how to use it to actually grow your business.
More importantly, you’ll see that it can give you valuable insights into where exactly your site is underperforming.
While working with a client recently, I noticed looking at their GA data that there was something odd: a big loop back between their cart and product page. In other words, a lot of visitors were automatically sent to the cart after they clicked “Add to Cart” but the majority of them went right back. This indicated that, perhaps, they were not ready to start their checkout yet.
We tested this, stopped the forced redirection to the cart, and it improved their average order value by +13.53%.
There aren’t many Shopify store owners reading this who couldn’t use a little of that.
Because Shopify and Google Analytics can work well together.
You may not be able to make such huge leaps immediately – every store has unique challenges after all – but even small improvements can make big differences to your revenue. And Google Analytics can help you make those improvements – if you know how to use it to its full potential.
How to make Google Analytics work for your Shopify store
Knowing how to use Google Analytics properly requires special training, which is either expensive or time consuming, or both.
Searching for Google Analytics advice is overwhelming, an endless series of tips on how to track this and that. No wonder why many store owners simply give up.
Almost every website in the world has a Google Analytics account but very few actually use them on a regular basis for this very reason.
And even those who use it relatively often usually have broken setups, bad data, and inaccurate reports. I know this because I’m yet to work with a client with a perfect Analytics setup.
Fortunately, Shopify makes it relatively easy to integrate Google Analytics and a lot of the complicated stuff related to ecommerce is already done for you. There’s no need to hire developers to setup proper ecommerce tracking; you can do it yourself in a few simple steps.
I won’t include the setup guide here because Shopify has an official manual that tells you exactly how to do it, step by step. If you have not setup your Google Analytics account, simply go here for instructions that require no special technical knowledge:
This guide acts as an extension to that manual. It will help you to clarify that you are getting valid data, help you decide what you really need to track, and define the most important and relevant reports and metrics for your Shopify store.
We’ll cover all this stuff broken down into the following three sections:
- Section 1: The fundamentals of Analytics and how to apply it to your business
- Section 2: Setting up your account to track all the important stuff your Shopify store needs
- Section 3: The reports that provide the most valuable and actionable insights to increase revenues
NOTE: I have purposely left some of the more advanced stuff out. Advanced Analytics is very complicated and requires hours of training. I’ve saved you the trouble by including only the biggest issues that will allow you to start using the data you get from GA and start making smarter business decisions immediately.
The fundamentals (how to actually read your data)
Before we get stuck in, I’d like to give more context on why we are setting up Analytics in this way and how to actually read data.
Most business owners are bad at this. They either settle for very basic and useless data which leads nowhere or, worse, misinterpret what they see.
This creates the risk of making ill-informed decisions or missing growth opportunities that are right in front of their noses.
Focus on metrics that matter
Not all data is helpful. Some can even mislead you, making you think that you’re onto something when, in reality, you’re not.
Such data has a name – vanity metrics – and, in Google Analytics, it’s the vast majority.
Your traffic has gone up? It must be a good thing, right? No, that alone doesn’t actually tell you anything.
How do you know when data is helpful? When you can act on it, use it to make a decision, and when it can answer a specific question you had in your mind.
This also applies to data that is supposedly important. For an ecommerce store, metrics like the following can give you valuable insights into how your website is performing:
- Bounce rates
- Exit rates
- And more …
But the data is only valuable if you have a clear intent for using it. If you don’t know what to do with it, then you don’t need it yet.
Averages only show the overall picture but very little insight
It’s not enough to just look at total average numbers.
In reality, there’s no such thing as an average experience. For the visitor, the experience can be different based on which device they are using to browse your store, where they are coming from, and whether they came with an intent to buy or just browse. It can also be different based on who they are, their age, gender, and location.
This means that they will all have different reasons for not becoming your customers. If you only look at the averages and the total numbers, you won’t see the whole picture.
Here’s an average ecommerce conversion rate (number of visitors who turn into customers):
Now ask yourself: how does knowing my average conversion rate help my business exactly? Is it good or bad? How do I know? If it’s bad, what should I do?
That average number won’t provide answers to any of those questions.
You need to dig deeper. That’s where segments and filters come into play. Essentially, these do the same thing (although there are technical differences): they allow you to look at specific types of visitors in your reports.
So, if you look a bit deeper, the conversion rate for mobile visitors is far worse than for desktop users:
(NOTE: This is the case for almost all merchants now, by the way, so don’t panic!)
Now, how does knowing this benefit your business? You see that there are far more mobile visitors but they are not converting into customers so readily – that’s an insight. Now you know that it is worth investigating your mobile UX (user experience) and figuring out how can you improve it.
From this oversimplified example, you can see that aggregated data is quite useless. Segmentation provides more context, which leads to better insights.
Beware of insufficient data and outliers
A large part of this guide will focus on how to make sure that you are getting clean data. Google will automatically record stuff that’s useless and misleading, so, if you don’t filter it out, you’ll see wrong numbers.
On top of that, you’ll likely see data that represents either out-of-norm events (called outliers – special events, promotions, holidays, etc.) or there’s just too little data to draw any conclusion (especially the case when you’ve recently set up your account).
An outlier is essentially a set of data that is either way smaller or way bigger than most of the other data points. It doesn’t represent the normal behaviour, so you can’t really rely on it as a valuable insight (insights need to be based on trends).
Ignore statistically insignificant data: this is best explained with an example:
Wow! There’s a very high conversion rate for traffic coming from Bing. That must be really valuable, right?
No. There are 8 conversions out of 60 sessions. This data is not statistically significant as the sample size or amount of data is too small. You should ignore it for now.
How to add Google Analytics to Shopify
The official Shopify manual covers most of the steps, including how to enable ecommerce tracking and how to create goals and funnels. Here are a few important things that are not properly explained and stuff that people tend to forget, leaving them with incomplete data.
Your account is broken down into three levels:
Account – mainly focused on user management and access to different levels of data. If you are the owner, you have access to everything but you can invite other people and limit their access.
Property – this is where you configure where and how data is collected. The code snippet you implemented on your site contains a property ID. Based on that, Google collects the data and reports what’s happening around that ID.
You can have two different sites under one account but if you use the same tracking code and the same ID, data from both sites will mesh together under one property ( and you want to avoid this).
Use one property per domain (you can have subdomains under one property).
View – when data is processed, it then shows itself in a ‘view’. This is where you examine your reports. By default, you will have one view that is called “All Web Site Data”.
But you can create more views (25 total) and apply filters so that the view is processing data for only certain segments of visitors.
A smart thing to do is to have at least five views:
- All Web Site Data – your default view with a few filters for cleaner data (I’ll explain later)
- RAW view – your untouched data: this will serve you only as a backup because once you have applied changes to how the data is processed, it cannot be undone in GA
- Test – your ‘play’ view: since you cannot ‘undo’ filters, this where you try them out before applying to other views
- Desktop – you filter out all visitors not browsing your store using a laptop or a PC (plus cleaner data filters)
- Mobile – you filter out all visitors not browsing using a mobile device (plus cleaner data filters)
To create a new view go to: Admin > click “Create new view”
Name the view, set your timezone and click “Create View”.
You are all set but you still need to create and apply filters to make your views useful.
Create and apply Filters
Filters are created in the admin interface. You can create a filter on the account level (to apply it to several views) or directly on the view level (for one specific view):
Honestly, it doesn’t matter where you create it; you can still apply every filter you create to every view where you created it.
Select the view you want to create a filter for > Go to “Filters” > click “+ Add Filter”.
Filter Your IP, your staff’s IP, agencies & third parties
Since it’s your website, you are going to visit it quite often. However, Google can’t tell the difference between you and your visitors. This means that if you don’t filter yourself out, you will slightly skew your data.
This issue becomes more important if you have a relatively big team. It can easily amount to considerable non-relevant visits.
Go to: Filters > click “Predefined” > Choose filter type “Exclude” > add your IP address > Click “Save”
This can get tricky if you need to filter out more than one IP address; you’ll probably need to use Regex.
It’s only relevant if you have a sizeable staff. If you have 10 people, for instance, you can just write down all the IPs like this and be done with it:
(NOTE: separate each IP using the pipe “|”)
However, there’s a 255 character limit (that’s approximately 15 IP addresses). So, if you have more than 15 IP addresses, you won’t be able to include them all in one filter.
In this case, Regex can help, as it uses special characters to describe a pattern.
If you have a large staff and they share an office, all of your IP addresses will follow a pattern because they are under one network. So, instead of having to type them all in like in the example above, you use special characters to instruct Google to filter all of the IPs that follow this specific office pattern:
You don’t really need to learn Regex to be able to do this. This tool comes in handy: just enter the first and last IP addresses and it will generate your Regex for you.
This can also apply to all of the other people working on your site – agencies, contractors, third parties and your employees’ home offices (if they work from home or if you have multiple offices).
NOTE: You should apply this filter to all views (except for RAW, the unfiltered view). You don’t need to create it separately for each view. Once you create it, you will be able to select it from the existing filter list when you are adding it to the new view.
Filter out spam and bots
You may get traffic from spam bots. This is ‘fake’ traffic. Sometimes it can even be disguised as traffic coming from reputable sources like reddit.com, lifehacker.com, vice.com and others but it is still fake (no one actually visited your store from those links):
(NOTE: all traffic sources in the image above are fake)
How do you know Reddit is fake? If there are legitimate posts with a link back to your site then no they are not fake (probably).
A simple way to find out for sure is to go to Google Analytics: Audience > Technology > Network > Hostname
If you see Reddit.com (or any other popular and seemingly reputable sources) then it’s fake.
The only valid hostnames are your own domains, subdomains, and third party tools you use to run your site (including Shopify). As you can see in the example above, there are quite a few fake ones.
You need to configure your ‘view’ settings to exclude this kind of traffic:
NOTE: By doing this, you will stop recording future spam hits but they will still show in your past reports.
It may not completely eliminate the problem as there are many types of such bots. You may need to filter out some of them separately, similar to how you filtered out your IP addresses.
This is only a problem if you still get a lot of these spam hits after you checked the bot filtering box, which usually works well. Here are deeper step-by-step guides on the best-known bots:
Filter by Device
To create a mobile filter: Name the filter “Mobile” > Choose filter type “Custom” > Click “Include” > scroll till “Device category” > Choose ”Mobile” > Click “Save”.
That’s it! Now you have a new view where you will see data only from mobile visitors.
To create a filter for only Desktop visitors, follow the same steps but choose “Desktop” from the Device category.
NOTE: There is more to it. As you get more advanced with Analytics, you can also create views for even deeper segments like only registered members, a specific traffic source (e.g. only organic traffic), or based on their location (e.g only US visitors – makes sense if 90% of your purchases come from the US).
Just like filters, segments allow you to look at your reports in a more detailed way. They are basically like mini-filters that you can apply directly to your reports:
You can create your own segments from simple stuff like device, browser, demographics, traffic sources and more. Most of these will be automatically created by default.
But there are more-complex conditional and sequence segments: for instance, visitors from a specific demographic who come from a specific traffic source and complete a specific action.
What’s the difference between segments and filters? There are fundamental technical differences.
For starters, you can use segments to look at your historical data and they don’t have a permanent effect. If you screw up, no biggie! You can delete the segment and it won’t affect the overall data in any way.
So why not always use segments instead of filters?
It’s much more convenient to use filters for some of your bigger segments because of sampling. For example, if you have 60% of visitors on mobile devices and you segment by mobile in your reports, you will experience this:
Google will report only a small fraction of the data. In the example above, it’s just 9.07% of all the data. This normally happens when you look at historical data and for a time range more than a month.
Although this is covered in the official Shopify manual, a lot of merchants tend to forget to exclude their own domain, Shopify, and all the third-party payment options from the referral list.
If you don’t do it, here’s how the referral report will look:
Most of the transactions will be attributed to Paypal (or other third party payment methods you use), Shopify, or your own domain. This is not accurate at all: just look at the ecommerce conversion rate of 60.41%.
It is basically telling you that your own site is sending you visitors and you won’t be able to tell where the shoppers are actually coming, rendering the data useless.
To fix this, go to: Admin > Property > Tracking info > Referral Exclusion List:
Click “+Add Referral Exclusion” and add these domains:
- yourwebsite.com (the primary domain of your site; you may have more than one)
- paypal.com (or the URL of other third payment providers you are using, such as apple pay or amazon payments)
Enable Google to track additional data
Now that you have filtered out the ‘bad’ data and set up your account the right way, it’s time to tell Google to collect more of the data that is important to your ecommerce business (stuff that’s turned off by default).
If your store has a search form, you need to enable this. It will give you valuable information about what types of products your visitors are searching for and whether they are finding what they are looking for.
Go to: Admin > View > View Setting > Scroll down to Site Search Setting
Turn “Site Search Tracking” on.
You also need to enter the GET variable used for search queries in the Query parameter box. Simply perform a search on your store and look at your URL. Normally it is something like this:
In this case, “q” is the GET variable and you enter it in the box.
Finally, check the “Strip query parameters out of URL” and leave Site Search Categories off (you won’t need it unless you have a huge list of products with various categories and subcategories).
Don’t forget that, since this happens at the view level, you need to configure this on all of your views separately.
This is important because it will give you a clearer picture of the types of people visiting your store and how their behaviour differs, depending on age and gender.
Go to: Admin > Property Settings > Turn on “Enable Demographics and Interest Reports”
Ecommerce (and enhanced ecommerce)
This is just a reminder not to forget about your ecommerce data because it will not be turned on automatically. However, the Shopify manual covers all the steps in full.
If you are not tracking your ecommerce data in Analytics, why are you using analytics at all? This is by far the most critical data available from GA because it tells you exactly how much money your store is making and how it correlates with other website usage metrics.
Plus, today Shopify makes it so easy to implement this with just a few clicks that it’s really a crime if you don’t use it.
Just a little while ago, you needed to pay people to enable this for you because it required technical skills.
Goals and funnels
It helps to look at your whole site as a ‘funnel’: each page has a specific goal that leads to the next page (or the next stage in the funnel) until your visitor becomes a customer, which is the ultimate goal.
With Google Analytics, you can see how effective your funnel really is. It gives you more insights into where people are dropping off and which pages you should pay more attention to – where exactly you are leaking money.
The Shopify manual on how to create goals and setup funnels and almost all of the other guides will show you how to create a funnel just for your checkout pages, starting from your cart page or product page.
However, this is only relevant if you are on Shopify Plus because then you can actually edit the checkout pages (though there are still some limitations).
I found it more useful to create a funnel including all of your key pages. For a typical Shopify store, it looks something like this:
- Category/Collection page
- Product page
- Cart page
- Shopify checkout (4 additional steps)
This way you can see the ‘big picture’: which pages get the most traffic (and from where), how they correlate with each other; and where the biggest dropoffs are happening.
But, in order to be able to see this, first, you need to create a purchase goal.
Go to: View > Goals > Goal setup, choose “Custom” > Goal description (give it a name: purchase, payment, completed checkout … whatever) > Choose type “Destination” > Click “Continue”.
In ‘Goal details’ choose ‘Regular expressions’ (yes, that’s Regex again because the full URL is different for each visitor but it’s still following a pattern) > Enter the URL of the visitor’s final page visit after making a purchase (leave ‘case sensitive’ unchecked).
For your Shopify store it’s the following:
You can just copy this URL, even if you are not on Shopify Plus.
Turn on the funnel switch > Add these steps (just copy them too as all Shopify stores have the same URL structure unless you do some customization):
You can check if you entered the destination URL correctly by clicking “Verify this Goal”. If all is good, click “Save”.
- It will take up to 24 hours to see any data collected
- Goals are configured at view level, so you will need to do this for each view separately
- You can create and track all kinds of goals: not just purchases but any kind of action visitors take while they are visiting your store e.g., when they subscribed to your email list, clicked a specific button, shared socially, received 404 errors, completed a form or scrolled to the bottom of the page.
For your Shopify store, you probably don’t need all of that. This goes against the popular advice where you should track everything but I’ve seen companies spending tons of money and effort making sure they track every single detail and then rarely looking at the data.
Keep it simple and focus on metrics that really matter. On top of that, enhanced ecommerce tracking (which you already enabled, right?) automatically tracks almost all of the important ecommerce data by default.
Further down the road when your business gets more sophisticated, it may be a good idea to create and track more sophisticated goals but, by then, you will be better off hiring people than doing it yourself. After all, creating complicated event goals requires writing code.
Group your content
Since you’re running an ecommerce store, you likely have more than one product and more than one category.
Instead of seeing many pages like this:
You can create a product page group and see it in Google Analytics as one page with content groups.
This comes in very handy for certain reports. I recommend that you at least create content groups for your key funnel pages.
Go to: Admin > View > Content Grouping > +New Content Grouping
Name it “Content directory” or “Funnel” (or whatever you like) > Group using rule definitions > + Create a rule set.
You can use the same regex URL that you used for Goal funnels.
- It will take at least 10-20 mins to see content grouping data (until Google records new data). You will not be able to see past data with the new groups.
- “Not set” will likely appear in your reports: this is just traffic that’s not set in a particular group
- You can create up to five content groupings per view in GA
- You can also create more specific groups for products and collections to see the performance of each product and category but you can also see that in ecommerce report (and you need to group using extraction, which is a bit more complicated)
Webmaster Tools (Google Search Tool)
The Google Search Tool is really important for your SEO as it will tell you which keywords bring visitors to your store and where you rank.
But you need to create a separate Google Webmasters account and then verify your store. The Shopify manual covers the steps you need to take in full.
The most useful Google Analytics reports for your Shopify store
The big picture
Remember that it is useful to view your store as a funnel, with a number of steps your visitors need to take to complete their purchase.
Another healthy way to look at this is that each step is designed to assist the visitor in making a purchasing decision.
When you look at the whole funnel, you can easily identify the problem areas – the steps that need improvement.
Here are two reports that can help you with just that…
Shopping Behaviour report
Report: Conversions > Ecommerce > Shopping Behavior
How to read it: it helps you visualize the visitor’s overall shopping experience in your online store. You can see how your visitors move from one stage to the next or where they abandon the process.
In the example above, notice how only half of all the visitors view product pages. This indicates that the store is doing a subpar job of directing people to view its product pages.
What’s even worse is the next step: adding products to their carts. High numbers of visitors simply abandon the product pages, which may indicate the need for design improvements.
The report above is nice enough but it doesn’t really tell you what happens before visitors view product pages. This is where the ‘goal flow’ comes in. If you configured it correctly, it can provide an excellent visual overview of your whole funnel:
How to read it: This report helps you visualize the “big picture”: the path visitors take towards making a purchase, featuring each of the key steps. Check:
- Where the visitors enter the funnel and which traffic source they originate from
- High dropout rates – where people exit the funnel (the leaks)
- Loopbacks – where people turn back to the previous steps of the funnel
You can see that visitors rarely start their path from the homepage. It’s the collection and product pages.
But that’s also where the biggest dropoffs happen and there’s a big loopback between product and collection pages.
This highlights the potential to significantly increase your revenue by making improvements to these pages but first, you need to investigate exactly what’s wrong there – and perhaps examine alternative ways to direct the traffic.
You can also use segments to see the big picture for different types of visitors. Answer questions like:
- Where are the biggest drop offs for mobile visitors?
- What’s the difference between new and returning visitors?
- Do they drop off in the same places?
The more effective your funnel is, the better your online store will be at converting sales and making money.
- The Funnel Visualization report is similar but less useful. If you read more guides, you may see it referred to. There are important differences. ‘Goal flow’ is more flexible and you don’t need to configure anything extra if you have the goals and funnels created.
- The money is always closer to the checkout so first look at the cart and product pages (or the full checkout if you are on Plus). See if there are any significant leaks (like half the traffic dropping off from cart to checkout). Visitors who make it this far show more interest in buying and so there’s a higher chance of converting them.
Traffic & marketing campaign performance
Before we get into traffic reports, it’s important to understand the UTM tags (Urchin Tracking Module). Despite the name, it’s not that complicated. It’s simply an extension of your URL for better tracking.
You have probably seen a URL like this:
This particular URL was created for email so you see: medium = email and campaign = instagram marketing strategy because it’s a blog post about Instagram marketing strategy.
A newsletter email was sent out to the list and it allowed to tell how many email subscribers read that blog post (and how many of them became customers afterwards).
You should create these UTM tags for your guest posts, social media links, newsletter links, affiliates, and paid ad campaigns.
Google has a handy tool called Campaign URL Builder, where you can easily add proper UTM tags.
This is by far the most popular report in Google Analytics. Chances are that you have already seen it and it might be the only report you somewhat understand. It tells you where your traffic is coming from.
Report: Acquisition > All Traffic > Channels
How to read it: figure out which traffic sources are leading to purchases and better conversion rates. See which traffic sources are performing poorly and double down on what’s performing best.
The reason why I introduced you to UTM tags is that, without them, your traffic might end up in the wrong categories. This is especially the case when you are running several different campaigns. It can get messy (in fact, most accounts are really messy in this way), making this report unreliable.
If you are not sure what each of the channels, sources and mediums mean, here’s a detailed explanation.
Products & sales performance
You need to sell people what they want to buy. With these reports, you can see which products people are searching for and what gets very little attention.
Site Search Report
Report: Behaviour > Site search > Search terms
How to read it: Look at terms with high search volume but high exit rates. This indicates that people are not finding what they are looking for.
Report: Conversions > Ecommerce > Product Performance > Shopping Behavior
How to read it: Find your best sellers and look at products that bring in the most revenue. Identify products with high views and low buy-to-detail rates.
Get rid of products that no one is interested in and put more emphasis on your bestsellers and products that generate the most revenue – display them on your homepage and make them hard to miss.
What we covered here is only a fraction of what you can do with Google Analytics.
But, just like most things in life, it’s wiser to start with smaller steps and then gradually move to the more complicated stuff. Otherwise, you get overwhelmed and end up taking no action at all.
Most Google Analytics accounts are misconfigured (or completely ignored). If you follow the steps above you’ll have a better set up than 90% of other Shopify stores – and that will give you an important competitive advantage.
Most importantly, you’ll know how to actually use Google Analytics to grow your business.
Author: Emils Veveris
Emils Veveris is a certified CRO strategist and consultant. He helps ecommerce stores get better at converting visitors into customers using UX research, digital analytics and A/B testing.