This interview is with Matt Suffoletto, Founder & CEO, PageSpeed Matters.
What experiences shaped your path to Founder & CEO focused on website performance and pagespeed optimization for tech startups and e-commerce?
Honestly, it wasn’t some grand plan from the start. I got into web development pretty young and just kind of learned by doing, you know?
Early on, I was building sites and noticing that half the time the actual code wasn’t even the problem; it was how the site was being served and loaded.
That stuck with me. I spent years just tinkering with optimization on the side before it became the main focus. The real turning point was when I started working with eCommerce clients and saw firsthand how a few seconds of load time could literally tank someone’s conversion rate.
We are talking about real money disappearing because a page took too long. That’s when I knew this was the thing I wanted to pursue wholeheartedly.
I started PageSpeed Matters because I kept seeing the same problems over and over, and there just wasn’t a team out there doing it right for the types of companies that needed it most: B2B and eCommerce sites that were bleeding performance without even knowing it.
So, yes, it was less of a lightbulb moment and more like I just kept getting pulled toward it until it made sense to make it official.
Building on your use of inflection-point analytics, how do you convert a sudden rate-of-change alert into a 24-hour engineering action plan?
So this is actually something we’ve dialed in over the years through a lot of trial and error. When we get a sudden spike in the rate of change on any key metric, the first thing we do is not panic.
It sounds obvious, but it’s easier said than done when a client’s bounce rate just jumped overnight. We triage quickly; basically, within the first hour, we’re figuring out if it’s a real performance regression or just a traffic anomaly.
If it’s legitimate, we map it back to whatever changed—a deployment, a third-party script, a CDN issue, or whatever it is. From there, it’s pretty straightforward, honestly. We prioritize by impact, so whatever is affecting conversions or Core Web Vitals the hardest goes first. In my experience, the fix is usually pretty contained once you actually isolate the root cause.
The 24-hour window keeps everyone honest, too. No one is sitting on it over a weekend or whatever. We’ve built that cadence into how we operate at PageSpeed Matters, and it’s honestly one of the reasons our clients see results so quickly.
Staying with measurement, what is your process for turning real-user monitoring insights into case-study content that attracts qualified buyers while preserving client confidentiality?
Yeah, this is something I think about a lot, actually. Real-user monitoring data is gold for storytelling, but you obviously can’t just throw client numbers out there. So our process is pretty simple: we anonymize everything first. No company names, no specific revenue figures, and nothing that ties back to a real client unless they explicitly say, “go for it.”
From there, we look for patterns across our 1,500-plus optimizations and pull out the ones that tell a compelling story.
If we’ve seen the same issue tank load times across 20 different Shopify stores, that’s a case study right there without ever naming a single client. In my experience, the best content isn’t about flexing specific wins anyway; it’s about showing someone, “Hey, this thing is probably happening to your site too.” That’s what actually converts qualified buyers.
People don’t care about someone else’s numbers; they care about recognizing their own problem in what you’re saying. So we keep it pattern-based and outcome-focused, and honestly, that approach has been way more effective for us than traditional case studies ever were. It just builds more trust from the jump.
On the deployment side, within your zero-touch live deployment philosophy, which single safeguard has most consistently protected Core Web Vitals during media-heavy feature launches?
So look, when you’re pushing a media-heavy launch, like heavy images, videos, and new interactive elements, everything can go sideways fast if you’re not careful. The single biggest safeguard for us has been staged rollouts with automated Core Web Vitals checkpoints baked right into the pipeline.
Basically, nothing goes fully live until it passes our performance thresholds automatically. No one is manually signing off at 2 AM hoping for the best. That’s the whole point of zero-touch; you set the guardrails and let the system do its job.
The moment you add a manual step to that process, someone cuts a corner under pressure, and your LCP just craters. We’ve seen it happen. The staged approach means that if something fails the checkpoint, it just stops—simple as that.
It doesn’t matter if it’s a single image that’s too heavy or a whole feature set; it catches it before it touches your real users. We’ve been running this way for a while now at PageSpeed Matters, and it’s honestly the one thing I’d tell any dev team to implement immediately if they haven’t already. It’s not flashy, but it’s bulletproof.
Shifting to business growth, how did you structure your performance-backed funding model so incentives stay aligned and scope remains controlled?
So this one took us a while to get right, not gonna lie. Early on, we tried the traditional agency model: you know, fixed price, do the work, send the invoice. And it just didn’t sit right with me because there was no skin in the game on our end once the project was done. So we flipped it.
Our model now is tied directly to performance outcomes. If we don’t move the needle on the metrics that actually matter—conversions, Core Web Vitals, bounce rate, whatever we scoped out upfront—then we’re not hitting our targets either.
That keeps everyone honest. Scope creep was the other thing we had to nail down early. In my experience, that’s where most agency relationships fall apart. So we got really tight on our agreements, with clear deliverables, clear timelines, and clear boundaries on what’s in and what’s out. It’s not about being rigid; it’s about protecting both sides.
The funding side of it naturally followed once we could actually prove the model worked. Investors like seeing results tied to revenue, not just promises. We’ve done over 1,500 optimizations at this point, so the data kind of speaks for itself. It just made sense to build the whole structure around that.
Architecturally, which decision has produced the largest speed-to-revenue lift in your work—edge compute, SSR/ISR, or a decoupled storefront?
If I had to pick one thing that consistently delivered the biggest speed-to-revenue lift for our eCommerce clients, it has been decoupled storefronts. And I say that as someone who has worked with all three pretty heavily. Edge compute is great, and SSR and ISR have their place, but when you fully decouple your storefront from your backend, you’re removing so many bottlenecks at once that it’s hard to overstate.
We had one WooCommerce client where we moved to a decoupled setup, and their LCP dropped by almost 60 percent. The conversion rate ticked up almost immediately. From what I’ve seen, the merchants who resist decoupling are usually the ones still thinking about it as a rewrite, when really it’s more of a shift in how the front end communicates with everything else. It doesn’t have to be scary.
The edge solutions are appealing and generate a lot of hype, but for pure speed-to-revenue impact on eCommerce specifically, decoupled has been the winner for us more often than not. We’ve seen it across WordPress, WooCommerce, Shopify—pretty much all of it. That’s where the real gains reside.
From a team perspective, what hiring profile and operating cadence have helped you keep performance regressions near zero at scale?
Yeah, so this one is actually something I’m pretty proud of, honestly. When we first started scaling the team, I made a mistake a lot of founders make: I hired generalists and hoped they’d figure it out. That didn’t work.
Performance optimization is one of those things where you really need people who genuinely geek out about it. So now we hire specifically for that—people who already care about Core Web Vitals and page speed before they ever walked through our door. That’s non-negotiable for us.
On the operating cadence side, we run tight. Every optimization goes through a review before it touches production—no exceptions. We’ve built that into our process, so it’s just how things work here. Nobody’s eyeballing it and hoping for the best. Regressions don’t really sneak up on you if you’ve got the right checkpoints in place; they happen when someone skips a step or rushes something through.
We’ve basically eliminated that by making the process the thing everyone follows, not just a suggestion. At PageSpeed Matters, we’ve done over 1,500 optimizations, and keeping that regression rate near zero at scale has been one of the things that sets us apart. It’s boring stuff, honestly, but it’s what actually keeps clients happy long term.
Looking ahead, which emerging pagespeed or browser capability are you most eager to pilot for e-commerce conversion gains in the next year?
So there’s a lot happening right now, but the thing I’m most excited to pilot is the Speculation API, specifically prefetch and prerender. It’s still pretty early, but the potential for eCommerce is massive.
Basically, you’re loading the next page before the user even clicks, and when you nail it, the perceived load time is basically zero. For conversion-focused pages like product pages and checkout flows, that’s huge.
We’ve already been experimenting with it on a few Shopify and WooCommerce clients, and the early results are genuinely exciting. The biggest conversion gains don’t always come from the flashiest new tech; they come from shaving off those small moments of friction that users don’t even consciously notice. And that’s exactly what this does.
Browser support is getting there fast too, so it’s not like we’re waiting around forever. I think within the next 12 months, this is going to become a pretty standard part of how we approach eCommerce optimization at PageSpeed Matters. It’s one of those things where if you’re not watching it now, you’re going to be playing catch-up pretty soon.