Almost everyone agrees with the theory of web performance optimization, but what theory is needed in practice?
With the new Google Core Web Vitals update scheduled to debut next month, using the offer to evaluate the real network user experience is a big advantage when it’s time to follow up SEO the performance of your website.
The Core Web Vitals tool appears how your pages work, based on real – world usage data that tracks speed, responsiveness, and visual stability metrics to help website owners measure the online user experience.
Just before the Core Web Vitals investment tool was launched, Carlline’s technical director, Trainline, sat TechRadar Pro (virtually) to discuss how his team has worked on the launch.
Why is it important to measure network performance by placing customers?
Relevance is still obviously number one on page rank, but now that you also consider your site’s online performance through load time, interactivity, and visual stability, it’s important to optimize them so you can increase your reach to more customers and provide a better user experience for more people.
How has your team worked towards Core Web Vitals as an investment message and why is its implementation important?
We have always focused on optimizing our site for the user experience; However, the introduction of Core Web Vitals has provided us with the opportunity to leverage our investments to ensure we are ahead of their curve when they take effect. I manage Trainline’s front-end Teams, but it’s been a multidisciplinary effort – working closely with Back-End to ensure we optimize performance at all levels.
We’ve been working to set up the basics because it’s the most important step, and then we built from there, layering over the interfaces and then the back end of the website – the user experience is the accumulation of all of these. The baseline allows you to analyze your performance, giving you the basis for building your hypothesis.
For example, we realized that we could optimize our communication with our data connection platform by polishing the HTTP connection in our web application, which allowed us to run seconds from the total flow of the reservation. Measurement was then the key to ensuring that we could learn from our approach and that we could replicate what we had built.
Creating an optimal user experience depends on several factors, including perceived load time, interactivity, visual stability, and creating a balance between these so that the site behaves and loads as the user expects. The key thing is that there is no single speedometer that teams should target themselves, as in our experience, optimizing one often means compromising on another, which ultimately affects the user experience.
We focused our approach on user priorities, for example, on our home page, we’ve focused on making the Travel Search widget interactive as soon as possible so customers don’t have to wait for other elements to load before they can start typing a query. Search is a key part of the Trainline user experience, which is why we focus on this.
What has been the performance optimization journey of your network so far?
We’ve been working to optimize our web performance for the customer experience for a long time, but it’s been great to see that our approach has been strengthened with the introduction of Core Web Vitals.
It’s an ongoing process because the more we build, the more we need to evaluate and adapt to optimize performance. We are constantly building a product with over 300 releases a week, while preventing it from degrading performance or even better, making it faster. Measurement has been a key factor throughout the journey – ensuring that we continuously collect data in a consistent and reliable way so we can see how performance is evolving and how we can improve.
How has your team learned to distinguish practical insights from reporting noise?
It’s about getting the right meter in place – in the beginning we measured everything, but the breakthrough came when we started correlating our metrics to both our business metrics and network installations: we saw some releases have a negligible impact, while others would slow down or speed up performance to fine-tune optimization. Once they were linked to our business metrics, we can then prioritize our operations.
Second, it’s about focusing on the right areas, for example, instinctively addressing issues that affect customers who get the slowest experience – but for the impact to be greatest, you need to look at what affects most users. This way, you make a more meaningful difference to a larger number of customers.
Finally, I recommend that averages can be very misleading in terms of web performance, as they don’t give you a clear picture and can mask the extent of the problem affecting speed. Instead, the use of percentage points to focus on specific user groups proved important for this journey.
How can the needle be moved in terms of network performance?
We can optimize by measuring. The combination of synthetic measurements and real user monitoring (RUM) is key. Synthetic measurements allow us to compare improvements in exactly the same test environments under the same conditions, which means we can compare the performance of each version of our code.
We then test these potential improvements in the field, recording user data that gave insights into their experience, as this matters – and gains in the lab environment don’t always mean gains in the field.
Finally, we focused on where we could make the biggest difference to the majority of our customers to achieve the greatest overall impact. Moving the needle results in a combination of smart metering that guides you to change the user experience.