Funnels are one of the primary things conversion rate experts and data analysts are working on. Because if you can improve the number of users visiting your thank you page without changing the number of users on checkout pages, you will improve business revenue without spending additional budgets on the acquisition. A higher conversion rate means you can access more paid traffic sources and overcome competitors.
Thus, building a funnel exploration report in Google Analytics 4 is the first step to improving every business’s bottom line.
Unlike Universal Analytics (free version), Google Analytics 4 allows you to build funnels using historical data and adjust them on the fly. This functionality was available in Google Analytics 360 (premium edition) but not Universal Analytics. Therefore it’s one of the differences between GA4 and UA.
I dedicate this article to explaining what funnels stand for, how you can build your first funnel in Google Analytics 4 and this feature limitation. If you are interested more in actual funnel analysis, you can look at the case study of one of my clients. Let’s get started.
What’s a funnel?
As I mentioned early, let’s start with the funnel definition. If you are familiar with that, you can skip this article section and move to the next one.
Funnel is the number of steps someone (in our case, user or player) should complete to reach the goal.
For instance, ecommerce funnel usually consists of a product category page (listing page), product page, add to cart step, checkout steps and thank you or confirmation page.
For SaaS, the funnel is usually split into two or three parts or becomes much longer. So, we have the acquisition and product onboarding funnels. The acquisition funnel consists of the following steps:
- Landing Page
- Sign-up page (free trial or freemium model)
The product funnel consists of the following steps:
- New registration (New Trial User)
- Monetisation (Upgrade to Paid plan)
As you see above, in any case, the funnel is a series of steps a user should complete. Optimising each step’s conversion rate can improve the business’s bottom line without spending more resources on paid advertising. It means you can have similar costs but more profit!
Therefore, the ability to build funnels in Google Analytics is so valuable for every business. Universal Analytics funnels weren’t sufficient to meet business requirements. Therefore many SaaS companies selected other tools (such as Mixpanel and Amplitude), but what does GA4 bring to the table?
How to build your first funnel in Google Analytics 4?
So, Google Analytics 4 allows you to build funnels the same way as you can build in Mixpanel or Amplitude. Google improved its tool’s functionality; therefore, you can consider selecting GA4 instead of Mixpanel or Amplitude.
As a result of this improvement, you can build funnels using a drag&drop interface, you can build funnels using your historical data, and due to the event-based measurement model, you are not limited to sessions anymore.
Let’s build our first funnel in Google Analytics 4. For this article, I will use Google Analytics 4 Merchandise Store, a free and publicly available property, and you can replicate the same steps for the SaaS funnel.
Thus, to build a funnel in GA4, you should take the following steps:
- Open Google Analytics, go to Explore
- Select the Funnel Exploration template
And you built your first funnel in GA4. After that, we should customise our funnel to show what we are interested in, but before doing that, let’s overview the critical functionality of funnel report settings.
Once we make it, we get back to polishing our funnel to turn it into the ecommerce funnel that I mentioned above and includes the following core steps:
- Listing page (category page)
- Product page (product pages)
- Add to cart
- Checkout steps
- Thank you page
Funnel Variables (segments, dimensions, metrics)
The first element allows us to change the name of the overall exploration. We can build multiple funnels by adding additional tabs and naming each of them differently, but to name the overall report, we can use this section.
The second section allows us to change the date period of our funnel. If your period doesn’t include more than 2 months and you have GA4 for more than two months, remember to adjust GA4 data retention settings. It will help you to analyse data for more than 2 last months.
The third section allows you to build segments. It works close to what we had in Universal Analytics. For instance, you can build a segment which includes US users that visited your website on mobile.
After you build the segment, you can apply segments by moving them to section #9.
You can import select available dimensions and metrics in the fourth and fifth sections and then use them for breakdown or filter purposes.
For instance, you can build the segment that includes US traffic, or you can import “Country” dimension and apply the filter “Country” equals “United States”. Both variants will give you the same results.
Although these variables are the same across all explorations reports (including the User Explorer report) in Google Analytics 4, the second panel is more specific for the Funnel Exploration report; therefore, let’s look at it as well.
Funnel settings (visualisations, open/closed funnel, segment comparisons, etc.)
Section #6 allows moving between all available GA4 explorations. You can read more about other explorations on my blog.
Section #7 enables us to switch our funnel between two versions of the funnel: trended and standard. However, I couldn’t find trended funnel so helpful because of the fact that it doesn’t show you the conversion rate between steps but only the absolute number of users on the specific step. It still allows you to see the trend but not so prominent.
You can change the funnel from closed to open in section #8 of the funnel report. It can be interesting to have an open funnel sometimes. Still, in most cases and from my experience analysing the funnels, people usually use closed funnels to see trends better.
However, if you have an ecommerce and some of your traffic lands on the product page. It can’t make sense to keep the funnel closed when the first step is a category page. But even in this situation, I prefer two separate funnels, one starting from the category page and another starting from the pricing page, without including category page traffic.
I will discuss the open and closed funnel differences more carefully later in this article. Still, you should remember that open funnels mean that users can enter your funnel on each step from all places and not only from the previous step. It’s a core difference between them.
Before we proceed to section #10, let’s talk about features #11, 12, 13, 14, 15, 16 and 17 because section #10 is the core functionality of this report and requires a better look.
Section #11 allows you to use some dimensions as a breakdown. For instance, they can be country, device category, channel grouping and others.
Section #12 limits how many rows you want to see per dimension. If you apply the device category, it has 4 values only (mobile, desktop, tablet and smartTV), but if you apply country, it can have more than 100 values; therefore, here, you can adjust how many countries you want to see.
You can adjust section #13 to show the average elapsed time. Elapsed time is the time a user spends to complete the step successfully. If you see that some step of your funnel has a higher elapsed time than others, it can mean that users get stuck on that and don’t know how to proceed further. It’s not always the case because sometimes a step can require more time to personalise the final product, for instance, but it’s highly often to see this.
Section #14 – Next Action. It’s a new feature that allows you to see the TOP other events your user performed instead of moving to the next step of your funnel. It accepts two properties: event name and event screen.
If you are not interested in the specific events it shows (for instance, scroll or session_start), you can remove these events by creating an Event segment which includes all events without the ones you are not interested in.
You can also apply filters in section #15. You can filter funnel data by multiple dimensions accessible in GA4; remember to import them using section #2.
Apart from all the functionality, you can also use two features. First, you can zoom in and out using section #17 and export your report or give access (read-only) to your colleague in section #16. You can also use section #16 to redo or undo a specific action you did recently.
Now, after we discuss all the above-mentioned features, let’s look at section #10. This helps us specify the funnel’s steps and give them names.
Specifying steps for GA4 funnel
To better understand section #10, let’s change our funnel from the one we received as a template to the one we want to have – ecommerce funnel I mentioned above. Let’s click on “edit” (pencil) icon and select the steps for the e-commerce funnel.
This page has 4 sections that we are going to discuss further.
The first one is responsible for the name of the step. You can write your own; for instance, the first step of ecommerce funnel can be named“Listing Page”.
The second section allows you to select the event and dimension that you want to use to specify the step. In our case, we will use the ecommerce events such as item_list_view, item_view, add_to_cart and others.
The third section allows you to select weather the event is followed directly or indirectly after the previous event. According to the documentation, it means:
- Steps that indirectly follow the previous step can have other intervening actions.
- Steps that directly follow the previous step must occur immediately after the previous step.
Since a scroll event can follow most steps, I recommend using “indirectly” in most cases.
And the fourth step is responsible for the time you give you users to complete the action. Sometimes it’s relevant to specify it when you see that users on checkout pages spend (elapsed time) more than 1 hour. Most likely, this result is based on outliers, and you want to remove them.
After I specified all ecommerce funnel steps, I received the page with these settings.
If you configure the funnel according to what is mentioned above, you should see close to the funnel in our screenshot below.
Difference between open and closed funnel
Now, let me explain more carefully what it means to have an open and closed funnel. Again, from my experience working with clients worldwide, I found that closed funnels are used in 95% of cases, and even if you want to use an open funnel, very often, a closed funnel with minor adjustments can give you much better results.
Open funnel means users can enter this funnel at each step, while closed funnel means that users can enter your funnel only at the first step.
So, for instance, in our case, if a user enters your website from the product page, a closed funnel will eliminate this user, but an open funnel will show you this traffic this way.
Google Analytics 4 breakdown such users for the step when they entered and allows you to see how many users entered the funnel at each step, but after that, it combines all of the traffic.
You can play with that in GA4 and see if you can learn anything from that. From my experience working with Ecommerce and SaaS, much more beneficial to use closed funnels. Many product analytics solutions, such as Mixpanel and Amplitude, use closed funnels by default.
After you know everything about GA4 Funnel Exploration, let me tell you about its limitations.
Funnel Exploration Report Limitations
There are a few limitations you should consider when you set up funnels in GA4:
- The funnel can use data from Google Analytics 4 up to yesterday. Moreover, you can’t check yesterday’s data in the morning. GA4 requires more than 24 hours to process all your data. It’s the same for all GA4 reports. If you want real-time analytics, you should consider reading my other article about it and integrating GA4 with BigQuery and using Google Data Studio.
- “Next Action” feature can use only two dimensions: event name and screen name.
- If you build funnels with signup events or any other event when you assign user_id to this event, you should assign user_id to all events this user generates after this event. If you don’t assign to one of them, GA4 funnels won’t work correctly.
The funnel is the essential element of every business. Therefore, optimising it can help you to receive more conversions by keeping the costs at the same level.
Google Analytics 4 offers a new way to build funnels compared to Universal Analytics. It allows us to use historical data and build funnels on the fly. We can also build funnels based on users and not on sessions.
If you find this article helpful or have questions, feel free to comment below.
Frequently Asked Questions
To build a funnel in GA4, you should open GA4, go to Explore and select “Funnel exploration” template. After that, change the settings according to your case.
Funnel Exploration report allows you to specify the steps of your funnel, type of visualisation (trended or standard), see next events and elapsed time.
To analyse the funnel in GA4, you should open GA4, go to Explore and select “Funnel exploration” template. After that, change the settings according to your case.
The analysis usually starts with checking every step’s drop ratio and elapsed time. After that, you can apply different segments (mobile, desktop, traffic source, countries) to understand why users leave your website at a specific step. You can also consider watching session recordings in a tool like Microsoft Clarity, Hotjar or LuckyOrange. All of that should help you understand what you can change on the website in the product to convert more users.