As web developers, we all strive for a high Lighthouse score. However, despite following best practices on Core Web Vitals, some of us still struggle with a good score.
This talk will explore the differences between field data and lab data, the importance of the latter and practical ways to improve it, all backed up by live examples.
We will talk about the definition of TBT and long tasks, about JavaScript execution by v8 engine and the impact of Webpack on it. Finally we'll see how this can be improved and how indeed one can reach a score of 100 on Lighthouse.

The audience will learn:

- Difference between field and lab data and how it correlates with real user experience
- Types of long tasks and how to deal with them
- Fundamentals of JavaScript evaluation by browsers (parsing/compilation/execution) and how it affects lab scores
- Techniques for breaking up long tasks and improving TBT
- The truth about bundle sizes (which do not necessarily mean bad performance)

During the past year I’ve been leading cross-company effort to improve the Lighthouse Score of all Wix sites (which is a lot). I’ve learned a lot and I’m willing to share this knowledge with others. In fact I am already doing it here https://www.justjeb.com/post/how-to-score-100-on-lighthouse, but I’d love to do that in person.

Sneak peek: https://www.youtube.com/watch?v=892xU3TIudM (LibertyJS, Philly, 2023)