UX event data reflects user interactions with your website or application. This data allows you to better understand user journeys, detect common patterns, friction points, and improve your KPIs instead of just measuring them.
Before we go into it, let’s see the bigger picture of what type of data an online business should track
I usually split the data into three categories:
- Core events, the events that are directly linked to the value of your business
- UX (User eXperience) data, which is how people interact with the interface of your product,
- 3rd party data, which is the data from the other tools you have in your marketing stack.
In this podcast episode, join special guest Daniel Bashari, Co-Founder & CEO at Convizit, and our host Claudiu Murariu, CEO and Co-Founder of InnerTrends, as they dive into Why UX Event Tagging is Dead As We Know It.
Here’s what they talked about:
- What is the purpose of UX data?
- Why is the old way of tracking UX data failing
- Why companies need to track complete UX data
- How to track complete UX data in an efficient and useful way
Listen to the whole episode here:
Subscribe on your favorite streaming platform for more episodes:
Listen on Apple Podcasts Listen on Spotify Listen on Google Podcasts Listen on StitcherWhat is the purpose of UX data?
[D]: Most companies track core event data – that is required to measure their KPIs, – but many don’t have the ability to properly track UX event data which is essential for improving these KPIs.
UX event data reflects user interactions with your website or application. This data allows you to better understand user journeys, detect common patterns, friction points, and improve your KPIs instead of just measuring them.
[C]: Indeed, all the significant KPIs out there are about hitting certain milestones for the business. But to change those metrics, you need to interact with the product. So that’s where the UX data comes in.
How do we track UX data today?
[D]: When it comes to UX data, most companies rely on page views provided by default with any analytics tool. But their insights, using this very basic data, are minimal.
As we all know, modern websites and SaaS applications contain many models and different functionality on a single page. If we rely just on page views, we’ll get no insight into what visitors are doing with the web page itself, how they interact with different elements, and where they get stuck when trying to accomplish tasks.
Limited visibility into user journeys leads to minimal KPI improvement. Therefore, we all agree that UX event data must also include on-page activities. And the most advanced conventional approach to tracking UX event data in pages is actually to sit down and create a tracking plan.
That means deciding in advance what specific events you want to track, how to call them, and their relevant properties. Then someone in the organization, usually a developer, needs to add tracking codes to track each event with its properties. That’s usually done by using a JavaScript SDK.
Other approaches include Tag Manager, like Google Tag Manager.
Although this approach involves less code, it’s still very manual and requires you to understand CSS selectors, new processes, or visual tagging tools.
These tools allow you to tag different events using a user interface. But they come with their downsides, mainly around reliability over time and the fact that they still require some technical understanding.
So, to track properties, you still need to understand the structure of the web page, CSS selectors, and more.
[C]: We are not saying here that you don’t need a tracking plan. But by putting all the UX data in a tracking plan means automatically you will make your tracking plan a mess.
So the idea is not to eliminate the tracking plan but to take the UX data out of the tracking plan.
A better approach to getting the UX data
[D]: Convizit uses AI to understand user interactions on webpages or applications automatically and track every single interaction, extract the relevant meaning from the webpage, just like a human being does, and structure this data and send it into any tool that you integrate with.
you get complete UX event data effortlessly, without manual tagging, coding, or any of these things.
What’s unique about UX event data is that it tends to change a lot. Websites and applications change all the time. You cannot really keep track of all these changes when you do things manually. And you can get much more granular data by having a system that is intelligent enough to understand the webpage’s structure and which properties are relevant to which event without manually defining every single event.
What Convizit does is understand the webpage’s structure, the relevant context, properties, and the surroundings that may provide us with the necessary information for each event.
We track this information without having to manually define an identifier for a specific element you want to track or relying on CSS selectors, which is exactly what other tools do.
So you can get all the data you need without manual instrumentation.
It is important to keep in mind that for core event data, like for measuring your KPIs, you do need a tracking plan. You want to be consistent and measure the same things over time.
But UX event data reflects the user interface the customer sees. And this user interface changes constantly; you need an automatic tool to track the events to explain user behaviors better, understand the user journey, and detect relevant patterns. You cannot really keep track of all these changes manually.
How much UX data should we collect?
[D]: Unlike core event data, which should be very focused, when it comes to UX event data, it is important to track as many events and properties as you can.
Having complete UX event data allows you to detect different opportunities that you can’t find otherwise, especially with the sophistication of analytics tools today.
You can’t possibly know in advance which events and properties are the most important to track. And usually, when you find out that you need this information, it is too late.
Generally, it’s better to collect complete UX event data than to track just a few events.
The problem was that, until very recently, you had to manually select and keep track of all these website changes, making it a real hassle for companies.
The good news is that you don’t need to use these conventional approaches. Using AI, you can detect all the changes automatically and track complete UX event data.
You don’t need to look at all the data at once, but when you need the data to investigate opportunities, find insights, and understand friction points, you can get the exact data you need to find the right insights.
One last thing to remember is that you want to track as many UX events as possible. You don’t need to keep this information for too long. I would even say that just a couple of months of data is enough. Assuming that you have enough traffic. This should be enough to understand user patterns, find friction points, and answer your questions. But it is essential to have enough complete data to make the right decisions and improve your KPIs.
Another plus is that you don’t even need 100% accurate data. Most of it is tracked client side, JavaScript blockers will stop off some of that data, and nothing can go around that. But it’s all about statistics; parts of our product, sections, and buttons are more likely to influence people to go one way or another.
[D]: UX event data allows you to understand what’s happening between the final stages and what drives the next action. Although accuracy is essential in tracking data in general, the most important thing here is a complete understanding of what’s going on.
Accuracy doesn’t come from tracking all the users but from tracking more interactions. The more data I track, even if it’s not from all the users, I still get the whole picture.
What improvements can you see from correctly tracking UX data?
[D]: From a product perspective, the primary use is to uncover friction points or to get insights into different groups of users.
Many companies try to replicate success. They try to understand what drives desired behaviors to replicate them.
For example, suppose I see a specific segment of users with a high conversion rate. In that case, I’ll look at the UX event data and find the specific actions that led them to eventually convert or purchase the product or things they did in the first session.
Without having complete UX event data, you cannot find these patterns. It’s pretty hard to guess what makes them eventually buy a product.
Another use case is detecting friction points. Many companies use UX event data to explain drop-offs between final stages. it is much easier to do this when you have complete UX data.
Many analytics tools today have very sophisticated features.
Some of them, like InnerTrends, can also give you insights, which is quite remarkable. And I think having the correct data is the fuel for making these systems work for you.
On the marketing side, the main use cases that I’ve seen is creating segments for retargeting. Companies are looking to create more personalized experiences. We can see an increase in the campaigns ROI by using more granular data to define their segments.
These are the typical use cases, but the sky is the limit.
[C]: If I look at the data, core events are the data from the backend. That’s why we build the product. That’s how we deliver value. UX data presents the activity in the front end, and the front end is what links people to value. That’s where your product can deliver fantastic value. But with a poor interface, people would not get it. So UX data is crucially critical in showing you how good your interface is at delivering the promise that it makes and how people interact with your product.
Final thoughts
- UX data is essential for understanding how users interact with your webpage or your product, and how they get to the value you promised
- Tracking complete UX data will paint a better picture, even though it is not 100% accurate
- UX data is constantly changing, meaning you don’t need to look too far back in time to get useful insights
- It is almost impossible and certainly not cost-efficient to keep up with the UX changes manually, the solution is to track UX data automatically with AI tools.
About Daniel Bashari
Daniel Bashari is an entrepreneur, leader and technology visionary, currently serving as CEO of Convizit, a high-tech startup company she co-founded in 2017.
Throughout her career, Daniel provided strategic and technological consulting services to various organizations.
Listen to the whole episode and subscribe to the Data-led Podcast on your favorite podcast platform:
Listen on Apple Podcasts Listen on Spotify Listen on Google Podcasts Listen on Stitcher