Hi <<First Name>>,
Welcome to the April edition of Speed Matters! This month I'm excited to share research and best practices that offer surprising answers to interesting questions, such as:
- Are you optimizing for developer vanity at the expense of your business?
- Are your SVGs as fast as you think they are?
- Does a poor CLS score hurt other UX metrics?
- What level of "performance maturity" have most companies achieved?
If you have any questions or feedback, I'd love to hear from you. And if you have any great resources you think belong in the next edition of this newsletter, please send them my way!
Until next month,
Tammy
@tammyeverts.com
|
|
|
Performance audit: Costco.com
Can Cumulative Layout Shift (CLS) affect Largest Contentful Paint (LCP)? The answer is: yes, more than you'd think. Aaron T. Grogg shows how layout shifts can push new elements into view, effectively changing what counts as the "largest”" paint and delaying LCP. The big takeaway is that these metrics aren’t independent — if your page is janky, you might be quietly tanking your LCP, too.
|
|
|
UX Speed Museum
The UX Speed Museum is a fun project from longtime performance and UX advocate Sergey Chernyshev. Instead of just explaining concepts like input latency and layout shift, you can experience them firsthand through "exhibits" that simulate slow responses, janky layouts, and frustrating delays side-by-side with optimized versions. It's a surprisingly powerful teaching tool for developers, designers, and anyone working on digital products.
|
|
|
Web Performance Week
Mark your calendar: May 4-8 is Web Performance Week! This is a FREE virtual mini-conference focused on the latest and greatest in web performance and frontend observability — from Core Web Vitals to debugging user experience issues in production.
One registration covers the full week. Attend the live sessions that work for your schedule. Recordings, posts, and the AMA summary will land in your inbox when the week wraps up.
|
|
|
Your images are (probably) oversized
This post makes a simple but often overlooked point: many websites serve images that are way larger than they need to be, wasting bandwidth and hurting performance. Among other things, there's a walkthrough showing how even "optimized" setups (like using Next.js’s image component) can still deliver oversized images if you’re not properly using things like sizes and srcset.
|
|
|
Report: The AI and observability gap for frontend teams
We surveyed 300 frontend engineers to create the FIRST industry benchmark that connects observability maturity to AI readiness, specifically for frontend web and mobile teams. Among other things we found that 89% of engineers are using AI for coding — but only 8% are using it for observability.
While most teams can detect issues, very few can explain them. We found:
- 74% are stuck in mid-level observability maturity
- Only 5% can actually connect frontend experience to backend root cause
- Nearly half of mobile teams don’t even know that AI *can* help with debugging
Interested in learning more about AI-powered observability? Check out this upcoming live session (April 29).
|
|
|
Introducing Waterfall Tools
I’ve been checking out Pat Meenan's new Waterfall Tools – from the godfather of synthetic testing himself! – and it’s a really practical way to clean up and work with network waterfall data from different sources. It takes messy trace formats and turns them into something consistent and easier to explore, which makes digging into performance issues a lot less painful.
|
|
|
Why you need to know your site's performance plateau (and how to find it)
Have you ever asked yourself these questions?
- "I made my pages faster, but my business and user engagement metrics didn't change. WHY???"
- "How do I know how fast my site should be?"
- "How can I demonstrate the business value of page speed to people in my organization?"
The answers might lie with identifying and understanding the performance plateau for your site. That's what I presented for Episode 10 of the PerformanceObserver. Watch the full video and demo!
|
|
|
Want to improve INP? Yield to the main thread
If your JavaScript does too much work at once, it blocks the browser and makes your site feel slow. The trick is to break that work into smaller chunks and "let the browser breathe" in between. Joan León walks through different ways to do this, but the big takeaway is that small changes in how you schedule work can make your app feel dramatically more responsive, without rewriting everything.
|
|
|
Measuring SVG rendering time
Don't assume SVGs are always fast. When files are big or complex, things can degrade quickly. Stoyan Stefanov tested hundreds of images and found that rendering time stays pretty flat until certain size thresholds (around a few hundred KB), where it suddenly jumps. And for really large files, PNGs can actually render faster.
|
|
|
|
