From Public Flop to Top Tier 3x: The WP Buzz Rebuild

Our friends at WP Buzz, a small hosting company from Sheffield, England, are here today to share the secret to a recent achievement: WP Buzz entered the 2026 Review Signal WP Hosting Benchmarks in all three tiers and posted the fastest worldwide load time in every one, earning Top Tier status three times over. Impressive! We asked founder Haider Shah if he’d be willing to share how WP Buzz has achieved this success, and happily, he agreed.

Without further ado, here he is in his own words. –LC

For readers who haven’t met you yet, what’s WP Buzz, and what actually happened in the 2026 benchmarks?

We’re a WordPress engineering company from Sheffield. Six people, our own bare metal, and a habit of only saying things we can point to a public test for. We let the independently tested benchmarks do all the talking. There’s a bit of British culture baked into the brand philosophy: understatement over hype. These days we host mission-critical sites for enterprise clients on that infrastructure, and we’re steadily growing.

In 2026 we entered the independent Review Signal benchmarks at all three price tiers. We came back with the fastest worldwide load times of any host tested, at every tier we entered–255ms, 240ms and 237ms–and Top Tier at all three. Out of 36 city tests across the tiers, we won 30. In the UK, load times were as low as 150ms and never above 193ms at any tier. Six British hosts entered that round, and we outperformed the other five on every core metric: worldwide page load, UK page load, cached response, and hardware score.

You didn’t start on your own infrastructure. What’s the backstory?

We started in 2019 as a reseller, honestly because I was fed up with what the big hosts charged for what they delivered. It worked, we grew, we registered as a UK limited company in 2020. But I’ll be honest about that era. On third-party infrastructure our speed was okay, never exceptional, and logged-in performance was frankly unsatisfactory. The biggest issue was cache purging. It was unreliable, the origin ran Apache while the CDN ran NGINX, and the two never quite agreed. Around 60% of our support tickets were people saying they’d purged their cache hours or days ago and the site still hadn’t updated, which meant manual cache intervention at the origin, ticket after ticket. Support carried us. The machines weren’t ours to tune, and it showed.

Then 2023 happened. We entered the Review Signal WP Hosting Benchmarks as a reseller, and the provider we resold refused to lift a security measure for the test, citing competition concerns. The results showed it. No award, and the aftermath was worse. For months, whenever someone typed WP Buzz into Google, the AI snippet said something like “WP Buzz has great support but folded under the load test as soon as 200 users logged-in at once.” That really hurt the brand, and honestly, it brought us to our knees.

It also forced the question we couldn’t dodge any more. If we’re serious, why is someone else’s platform sitting between us and our customers? So 2024 was the rebuild. Months of testing hardware and software combinations under load. In 2025 the new stack went live on our own machines and we moved customers across as they were ready. And then we waited, because the benchmarks didn’t run again for three years. We spent that time ready. When the 2026 round opened, we went straight back to the test that had knocked us down. The full arc is on our story page, and the 2023 result is still public. We’ve never asked for it to be taken down. It turned out to be the best thing that ever happened to us.

During that rebuild you tested a lot of stacks. Why did LiteSpeed win?

Because it kept winning, and for a specific reason. We tested every major webserver. Apache was the slowest, no way around it. NGINX did very well, genuinely, but it never integrated as a coherent stack. You’re bolting on a cache layer here, a third-party CDN there, and the seams show under load. LiteSpeed is a WordPress dream. The fastest webserver we tested, and it ships with LSCache, which integrates with its own CDN. So it’s not just performance, it’s performance that stays coherent under load, and the seams never show. That held at every tier we entered in particular, and across the field of 31 hosts tested across those tiers in general. Coherence under load is where most of that field came apart.

It’s built as a family. LiteSpeed Enterprise as the server, LSCache as the caching layer that actually understands WordPress, Redis for objects, and QUIC.cloud as the edge, designed to work with each other rather than bolted together from strangers. Most hosting stacks are an arranged marriage between a generic server and a cache plugin. This one grew up in the same house.

And I have to say this part plainly. The LiteSpeed team went above and beyond helping us during our testing, Lauren especially. Without that support I don’t think we’d have taken Top Tier across every tier we entered. Some companies sell you software. This lot turned up.

The results show the fastest worldwide load times at every tier. What role did QUIC.cloud play in that?

The starring one, and the reason is what QUIC.cloud caches. Most CDNs cache your assets, images, scripts, styles, and do it well. QUIC.cloud caches the whole finished page at the edge, HTML included, which is the part that decides how fast a WordPress page actually feels. A visitor in London, Ashburn or Toronto isn’t fetching a page built somewhere far away, they’re getting the complete page from a server near them.

That’s exactly what the worldwide numbers measure. Twelve cities, full page loads, and we won 30 of the 36 across the tiers. The worldwide averages, 255, 240 and 237 milliseconds, are QUIC.cloud doing its job everywhere at once.

I’ll volunteer the honest bit rather than have someone find it. On the raw cached-response tests, a couple of hosts run faster raw edge responses. Where QUIC.cloud won us the round is the whole journey, the full page delivered across twelve cities, not the first millisecond of one. That’s the trade I’d take every time, and it’s the one the results reward.

And the other half, the traffic no CDN can touch. What are the machines doing?

Our machines run 5.7 GHz, the highest clock available, with Gen5 NVMe and DDR5 behind them. In the benchmarks, that showed up as hardware scores of 9.9, 9.8 and 9.9 across the tiers, and the fastest raw PHP compute at two of the three. The load tests threw just over two million requests at us–thousands of users logged in at once–and we finished with thirteen errors total, including a flawless zero at entry tier.

But neither half wins alone. The fastest machines with a bolted-on CDN still lose the click. The best edge network in front of slow origins loses everything uncached. What produced our results is one system, server, cache, objects and edge behaving as a single coherent thing. That coherence is the product.

What would you tell another host or agency deploying LiteSpeed and QUIC.cloud?

Trust the family. Use the pieces that were built for each other, LiteSpeed Enterprise, LSCache, Redis, QUIC.cloud, before reaching for anything bolted on. Half the performance problems we see on migrating sites are three caching products fighting over the same job. And test under real load, uncached, logged-in traffic, because that’s where stacks tell the truth. Cached speed is table stakes. The uncached half is where hosting is actually won.

What’s next for WP Buzz?

Honestly, not just benchmarks. Benchmarks will take care of themselves. What’s next is building a better, faster, more affordable experience for the people who host with us, and being ready for the agentic era.

We’re quality-over-quantity people. Retention matters more to us than acquisition, keeping the users we have happy beats chasing more and more signups, and that shapes everything from how support works to how we price.

Don’t take my word for anything I’ve said here. Every number in this piece sits at wphostingbenchmarks.com, checkable line by line, next to a decade of other hosts’ results, including our bad year. That’s the kind of testing this industry needs more of, and it’s why, when LiteSpeed asked us to tell this story, the answer was “yes” before the question finished.

Conclusion

Our thanks to WP Buzz for sharing their story in this Showcase! Has your business experienced the LiteSpeed difference? Have you created a tool that LiteSpeed users may want to know about? If you would like to share your story with our readers, find me (@Lisa at Litespeed) on our Slack workspace, and we can discuss it. –LC