New series: what assistive tech really hears on your checkout page. Start with part one →
Shipwright
Home
Performance Sep 20, 2026 1 min read 927 views

Before and After: Lighthouse Reports from a Month of Tuning

Priya Raman

Priya Raman

Database and performance engineer. Priya reads…

Share
Cover illustration for the article Before and After: Lighthouse Reports from a Month of Tuning

Four weeks of reports showing which changes moved the score and which did not. Performance work goes wrong when it starts with a guess. Every change in this post began with a measurement on real hardware and ended with a second measurement to prove it helped.

Week 1: performance score 41
Week 1: performance score 41
Week 1: LCP breakdown
Week 1: LCP breakdown
Week 2: after resizing images
Week 2: after resizing images
Week 3: third-party tags deferred
Week 3: third-party tags deferred
Week 4: performance score 94
Week 4: performance score 94
Final layout shift trace
Final layout shift trace

Reading the reports

Field data from real users tells you where the problem is; lab tools tell you why. We start with the Core Web Vitals report, pick the worst page template, and only then open the profiler.

As always, the examples are simplified to fit on a page. The production version has more error handling and logging, but the shape of the code is the same.

Questions, corrections, and war stories are welcome in the comments. Several of our best follow-up posts started as a reader pointing out something we had missed.

We will revisit this in six months with fresh numbers. If the results change, we will update the post and note what changed at the end.

If you try this on your own project, start with one page or one service rather than the whole codebase. A small experiment gives you real numbers to argue with, and it is easy to roll back if the idea does not fit.

What we would change next time

If you try this on your own project, start with one page or one service rather than the whole codebase. A small experiment gives you real numbers to argue with, and it is easy to roll back if the idea does not fit.

  • Start from field data, not a lab score.
  • Fix the largest resource first.
  • Compare percentiles, not single runs.
  • Re-measure a week later.
Priya Raman

Priya Raman

Author

Database and performance engineer. Priya reads query plans for fun and believes every slow page has a story worth measuring.

Related Posts

Comments 9

Ben Carter

This finally clicked for me after reading the docs three times.

Fatima Zahra

Came for the snippet, stayed for the explanation.

Ben Carter

Good question. We measured on a mid-range Android phone and took the median of five runs.

Chloe Martin

Good question. We measured on a mid-range Android phone and took the median of five runs.

Alex Moreno

Came for the snippet, stayed for the explanation.

Sofia Rossi

Counterpoint: for small teams the simpler approach in the first section is usually enough.

Elias Lindqvist Member

The quote in the middle is doing a lot of work here, and it earns it.

Sofia Rossi

We tried something similar in 2023 and rolled it back. Curious whether the tooling has caught up since.

Tomás Herrera Member

The part about naming things made me laugh, then made me go rename three functions.

Leave a comment

Your email will not be published. Sign in for a faster experience.

Supports **bold**, *italic*, and `code`.