How to set up a GA4 instance for successful ecommerce marketing with Enrico Pavan

Anna Shutko:
I’m your host Anna Shutko. And today our guest star is Enrico Pavan, who is a co-founder and president at Analytics Boost choice. In this episode, you’ll learn how margins can change their mindset to adapt to using Google Analytics 4 how the new Google Analytics exploration mode can be used for e-commerce, and what kind of insights one can get from Google Analytics 4, I hope you’ll enjoy this episode. Hello Enrico and welcome to the show.

Enrico Pavan:
Thank you, Anna.

Anna Shutko:
Yes and again, we have a super interesting topic. Today, we’re going to be talking about University analytics and migration to Google Analytics 4. So we’ve talked about Google Analytics 4 in this podcast before, but now we’re going to dive even deeper into this interesting topic. So my first question to you and Rico is what are the main differences between Universal Analytics and G4 for the users who aren’t familiar with Google Analytics 4?

Enrico Pavan:
Great question. Well, there are many. We can talk about I think about that for three or four hours, but in summary, I can say that the biggest difference between Universal Analytics and G4 is the measurement model. Universal Analytics uses a measurement model based on session and page views. Google Analytics 4 uses a measurement model based on events and parameters. The principle here is that any interaction can be captured as an event. So as a result, all Universal Analytics, it will be translated to events in G4 and another significant difference between Universal Analytics and G4 is the removal of monthly hit limits. The free version of Universal Analytics had a monthly limit of 10 million hits and for G4, that’s gone. Other stuff that could be very interesting related to the difference between UA and GA4 or GA3 versus GA4 depending on what you want to call it is that in the reporting interface or data collection, we don’t have anymore, the bounce rate metrics.

And here you can say UP or you could be so sad. It depends on what you use bounce rate, but we can talk about in G4 to engage this session that is the complete opposite of the bounce rate. In G4, there are no more views, but only properties and each event collected could contain the info parameters with values that could describe the event and could give you more actionable insight about the action taken by the users when you are going to track it and so we can say bye to the structure as we know, like event category, event action, and event label. Another difference is set for session and users. The session count in Google Analytics 4 property may be lower than session count Universal Analytics property because G4 does not create a new session when the campaign source change mid-session while Universal Analytics does create a new session under the circumstance.

So there is a little difference between sessions in GA4 and GA3 and if a session crosses the day boundary. So for example, if I start my navigation at five minutes to midnight and then end five minutes after midnight, it is considered a single session though it is counted once per day also GA4 try by default to duplicate and track the user across platform and devices in the report, and it’s count only active users by default and not all users as per Universal Analytics and it’s something that is the missing piece from GA4 and GA3 because right now, we are focused only to the active users not to all the users that land into your website or use your app or something like that.

So and I think the final difference I’m going to cover is GA4’s connection to BigQuery. For years, the connection was available only for the 360 accounts so paid account of Google Analytics, and now with GA4, we can connect our property for free to BigQuery and the latest news. We can choose which type of event to send to BigQuery instead of pushing all the information that we have into GA4 to BigQuery I think I forgot one and okay, I think we don’t have sampling into the main reports of GA4 so that’s quite huge against the GA3 version.

Anna Shutko:
All right, this all sounds very useful for marketers and you’ve mentioned many different things there so now the next question is how should marketers change their mindset in order to adapt and start using GA4?

Enrico Pavan:
Good question. I think marketing are obsessed with collecting valuable user data from their website or app and however, to capture the, I can say full picture of user behavior, marketers needs the ability to collect data across domain, devices, and so on. So for example, crossing domains, crossing from mobile to desktop from website to app and vice versa and something like that, and the customer journey right now in the moment that we leave is messy and complex, especially for companies with multiple platforms like website that spans for different domains for India, for the US, for APAC and something like that and then they have multiple apps related to the market that they want to cover and they want to be being informed on how you interact with the full digital ecosystem is critical and I think that’s why GA4 is updated to track modernly consumer behavior and organize it in one place for analysis.

That is the main point that we want to use GA4. One place where we put all the data collected and where we can manage analysis, stuff like that. The first point I think is if you don’t set it up for Universal Analytics is to create a measurement plan in order to monitor each event parameters you create in GA4 because, in GA4, we can create events, not only via the tech management system, for example, tech manager, or other types of stuff you target and stuff like that, Tealium and so on, but you can create events and modify events also into the platform of in the main section of the GA4 and using, creating or using a measurement plan will be a huge method to create a logbook or a personal playbook in order to retrieve information about data collection and you need to update the measurement plans in the future in order to stay updated with all the tracking that you set up for GA4.

And the biggest part of the shift is switching from the user session heat model that you have in GA3 to full event tracking and I think how to use the new metrics to track business performances. Avoid comparing that among the different software because as mentioned before, there are some changes in counting so you cannot right now take GA4 and GA3 and say okay, this week we have X session in GA3 and we have a lower count of session for the same week in GA4, stuff like that because there are two different methods of tracking two different analytics system. So I think that we need to change this type of mind shift that we need to set for marketers.

Anna Shutko:
All right, this all sounds very on point and I do agree with you that using GA4 as a centralized hub for all the data is very critical and marketers should indeed change their mindset and move from user recession tracking to more of an event tracking just like you mentioned. So now let’s dive deeper into this. So what should marketers keep in mind when they are adjusting their Google Analytics 4 for eCommerce to make sure they track the events correctly because you’ve mentioned that now that there is these event structures, everything is an event and so you need to know when a certain event happens to be able to report it correctly.

Enrico Pavan:
Yeah. Well, first of all, when you’re going to track your eCommerce via GA4, be mind to set all the required AKA mandatory parameters in order to obtain data in your report. As an example, usually in Universal Analytics commerce, the currency isn’t mandatory for the purchase or other checkout steps. If you miss this value in GA4, the reports show no data because you’re going to break the data collection. So be mind to set all the parameters required. This is the first thing that we need to set when we are going to track GA4 for eCommerce also please don’t use shortcuts to track GA4 eCommerce.

It’s better to set up a new data layer dedicated to GA4 data collection than to use some script to transform the old trucking version. I mean Universal Analytics to GA4, I mean, for example, use a custom Java script or something like that to bring the data coming from the data layer you already set for Universal Analytics and transform it into the new value for GA4 is something that could break the data collection and you could have no data into your reports and it’s a pity because you need to have all the data in place to take the proper action in order to improve your business.

And another point I think is important point is to understand how each new parameter could be used by marketers to obtain new valuable data. As an example, right now in GA4, you can set up at least five item categories instead of one. So thinking about it, you could track all the breadcrumbs for each product page or purchase or whatever you want. For example, set any other category type you need to track into the default parameters that you could use for GA4 also please do all with this, refer to the event tracking types set up in GA4. Google set four main types of events. We have automatically collected events, the announced events, recommended events, and custom events.

The custom event is something that we are going to track. If any other type of event is covered by the ones that set up Google by default and before setting up a custom event check if the event could be listed in one for the announced event or recommended event. This is because by using announced events or recommended events, you can skyrocket the power of machine learning that is the core of GA4 and take advantage of new features and reports because you’re going to feed GA4 with something familiar and the software can release new reports or use this type of data to improve modeling behavior and other huge stuff that could be very useful for you to understand your business and to improve your business.

Anna Shutko:
All right, this all sounds great and you’ve mentioned a couple of different scenarios, and how data can be used in GA4. So let’s dive into this topic a little bit deeper on a more technical level. How can marketers use GA4’s exploration mode for eCommerce now that they’ve set everything correctly and they have all the parameters set up and marketers know what to report on, how can they take advantage of this exploration mode?

Enrico Pavan:
Well, I think the explore section could be one of the killer features for GA4. Just because with exploration reports, you can dive into deep analysis and understand your user behavior at the most detailed level. As per eCommerce marketers, you can relate to a lot of new types of data exploration with GA4, but I think there are three measure reports right now that make the difference in comparison with other tools and Universal Analytics. These new reports, one of the new reports is the custom funnels. So with GA4, we can easily build a funnel using any combination of event or page views. We can filter it by any event property with a clever feature like measuring a lapse time through the funnel. So we can understand how many times people jump from step one to step two or from step two to the purchase, something like that and is something that is completely new for Google Analytics and you are not any more related to a destination goal like a URL to create a funnel as we did for Universal Analytics.

And you can simply mix events as your puff steps. So for example, you could understand the fallout from your checkout and the time to complete each step or could try to understand if users are going to add products after they get the proper cut value to avoid the shipping costs and many other super useful things. The second one is the path analysis and as per funnel, now in GA4, we can mix page views and events and understand the complete users pathing through the website and or app. GA4 gives you the opportunity to test not only the forward pathing but also the backward pathing as well. So that means that we can set the last point of interest. For example, I don’t know the ad or the product page views or the beginning checkout section, I don’t know, whatever you want and try to understand which path users follow to perform this event.

It is very huge because we can not only understand the starting point and where people go after the starting point, but from the ending point, I want to improve my pathing or the user habit before they perform this type of action in order to improve the advantage of the value to improve the purchase to improve whatever I want the customer experience at the website and it is something huge that is with Universal Analytics. We cannot perform using the interface that we have right now and the last thing is the overlapping analysis report that is a new report for the free version of GA4 and also for the paid version, but with the overlapping analysis, we can set up useful segment and try to discover hidden gems to optimize our commerce.

So for example, we can build up to free segment and understand if there are any interactions between them using an example, we can set up a segment related to free items, view it, and purchases by user. Having this information, we are able to discover correlations and maybe create bundle products, set a discount, for example, the second product that you buy is discounted by 50%, stuff like that, and much more. So as you can see, GA4 is very flexible compared to GA3 and I think the exploration section could be your Swiss army knife to improve business and performance.

Anna Shutko:
All right, this sounds very interesting and it does sound like there are quite many big shifts marketers have to adapt to and I think one of them is the use of inquiry because I presume with the new integration, which is completely free, more and more markers are going to start using this. So my next question here would be how can we merge data coming from Google Analytics 4 with data coming from other different sources, maybe some offline data in BigQuery, and what kind of benefits can this information and can this integration provide us?

Enrico Pavan:
Good question and is more simple than it appears to integrate data into BigQuery. So with Google BigQuery, we can easily solve such, I can say pressing problems, marketers have such as a real-time management and segment automation. As we know, marketers performance success depends on quick decision and quick reaction to market changes, make personalization, create automation and create user clusters to understand where and how to optimize business, not only online, but also offline. So once the data is imported, it’s easy to take it across the wall ecosystem to do predictive modeling or share it with your CRM. So for example, we can import data coming from our CRM system to BigQuery and using the correlation with Google Analytics 4 like the user ID, we can understand what people do on the website or app, and then we can make some automation using the CRM data with the primary keys, the user ID that we set and we can create automation.

We can create improvement in the main marketing and something like that and I think that one of the easiest ways to do that is to use the Supermetrics data connector to import that into BigQuery. So for example, thinking about Facebook ads with the connector set by Supermetrics, you could import the data in minutes and then merge it to GA4 data in order to analyze cross-market information. So we can for example understand the post click for campaigns into BigQuery that is not possible in the easiest way like we can do into BigQuery in any other stuff like also GA4, it’s better to move on BigQuery because you have more capability of data and user capacity to query on data and also you can create, for example, your own connector, but I’m not too in reinventing the wheel if we found some connectors ready to use in order to simplify my daily work and Supermetrics right now has a lot of data.

And could be very useful to take it into the marketplace of BigQuery or Google Cloud Platform to set it up and bring all the data coming from that ecosystem into BigQuery. Also, you can for example import data coming from your physical store into BigQuery. For example, you can send the data coming from the purchase made by users into your store and if you set the user ID as a fidelity card or something like that, you can use this key as the primary key to match the data coming from GA4, coming from your offline purchase, coming from your CRM, merge it and for example, you can understand if people are browsing something into your Apple, into your website and then buy something different into the store and then you can set up, for example, automation that could bring the user attention to complementary products that could be purchased into the store or into the eCommerce and something like that.

So you only need to define which is the key and the data that you want to set into BigQuery and then you can query on and get a lot of information to improve your business.

Anna Shutko:
All right, it’s awesome to hear that you’re enjoying using our tool and now I would still like to talk more about the reporting part. So you’ve mentioned how to get data into BigQuery would kind of scenarios or reporting scenarios marketing could cover and how to do it using a different kind of tools. So what kind of eCommerce reports can marketers build once they have all the data cleaned and imported into Google BigQuery correctly my second question here would be what kind of insights can users get from these reports?

Enrico Pavan:
The answer is it depends, but integrating the GA4 data into query under other stuff like CRM data, ad campaigns, not only Google ads or Google ecosystem as data, but coming from retail, something like that and also don’t forget to integrate first-party data. We can easily try to define the world customer journey. I understand user needs that is the main point that we usually use BigQuery in order to understand what people do before they purchase something or they fill out from online, or offline, and understand their needs in order to solve problems first and sell second in order to obtain a better customer experience side data, the website or app, or even the store and query on data, you could discover where you can set an optimization maybe as I mentioned before, not only online, but also offline.

For example, you can try to understand the pre and post-clicks coming from the mail chain and create better automation, or you can improve the recommendation online, starting with the data coming from offline purchases if you have a physical store. So using the data into BigQuery coming from online, or offline, you can get the big picture of the digital word that you are going to manage, to improve your business and regarding the insight coming from the report that you can set up as mentioned before, you can use GA4 as standalone software and use the data you are tracking into your website and or app to get the most of the value coming from the exploration section or other report staff in order to understand user behavior, create better, and targeted audiences and other huge information that you can set up.

So for example, you can set up audiences like never before to import into your ads campaign because you have more type of information that you can set up for that type of audience, but if you switch to BigQuery, merging the online, offline data, you could also improve the offline performance in discover user clusters to submit, to test in order to improve your, for example, an average of the value. You can create for example your metrics. You can try to understand which type of cluster of users you can have, for example, first-time buyers to repeat buyers to seven to 10 repeat buyers, most likely in the seven days or something like that, you can merge all these types of data and if you want, you can move a step over and bring all the data and maybe bring all the data queried on BigQuery and move it into something like Python script and use Markov chain or something like that in order to create your own attribution model or to understand, to create some predictive model that could be useful to improve the purchase or the customer journey inside your website or app.

Anna Shutko:
All right, this all sounds very cool and very useful. Thank you so much for sharing and now if the audience would love to learn more about you and I’m pretty sure they will, where can they find you?

Enrico Pavan:
Well, you can find me on Twitter as IndioImperator and on LinkedIn, Enrico Pavon, or maybe the better way is to subscribe to the Analytics Boosters newsletter in order to not miss any updates related to data, GA4, eCommerce analytics, and other stuff related to conversion rate optimization.

Anna Shutko:
All right, thank you so much for coming in the show Enrico.

Enrico Pavan:
Thanks for having me.