FindBestWebHosting has been covering web hosting since 2009. It's written by Tom George, who has been working with hosting infrastructure for 18+ years — as a developer, not a journalist. That context shapes everything about how we evaluate hosts.
This page explains exactly what we look at, where our data comes from, and how our affiliate relationships work. If you're deciding whether to trust a recommendation here, this is the right place to start.
What We Evaluate
⏱ Performance Data
We do not run synthetic single-point benchmarks and call them "our tests." That methodology is too easy to game and too easy to get wrong. Instead, we cite published benchmark data from two categories of source:
- Hosting companies' own documented infrastructure — PHP worker counts, server configurations, CDN architecture, caching layers. These are verifiable claims hosts make publicly.
- Neutral third-party benchmark publications — organisations with no affiliate relationship to the hosts being measured, using documented and repeatable methodology.
When we reference a performance figure, we cite the source. If we can't cite it, we don't use it. We hold every figure to the same bar: verifiable, attributable, and reproducible from a documented source.
📡 Uptime Claims
Every host promises uptime. We separate two things that most review sites conflate:
- SLA guarantee — the contractual uptime commitment, what the host will compensate you for if they miss it, and what the fine print excludes
- Publicly documented reliability — each host's own status page history, publicly reported incidents, and community-reported outages that reveal the gap between the promise and the reality
We report the SLA as stated and flag what the fine print excludes. We do not run independent uptime monitors — we rely on vendor-published status records and publicly documented incidents. A 99.99% SLA that comes with 14 exclusion clauses is not the same as a 99.9% SLA with straightforward terms. For a full explanation of how uptime guarantees actually work, see our dedicated guide.
💬 Support Assessment
Support quality is the hardest thing to measure fairly. We evaluate it through three lenses:
- Direct experience — interactions with support teams accumulated over 18 years of active use across multiple hosting platforms
- Documented response policies — stated SLA for first response, escalation paths, channel availability (live chat, phone, tickets), and what's restricted to higher tiers
- Community reputation — patterns in developer forums, Reddit, and verified user communities that reveal systemic issues individual interactions might miss
We do not run anonymous support ticket tests on a schedule. Assessments reflect accumulated experience and publicly available information. We flag when support quality has declined over time — acquisition, staff cuts, and offshoring are real factors that affect real users.
💰 True Cost Assessment
Promotional pricing is not the real price. We research and report:
- Sign-up price (what you pay today)
- Renewal price (what you pay after the intro period)
- The multiplier between them — expressed as a percentage increase, not buried in a footnote
- Add-on costs that are not optional in practice — SSL, backups, migrations, staging environments
- Overage charges and visit limits that create unpredictable bills
All pricing data is sourced directly from host pricing pages, verified at the date shown in each review's "Last Verified" timestamp.
🛠 Feature Verification
We verify features against primary sources — host documentation, support confirmations, and official changelog entries. We flag when a marketed feature is:
- A paid add-on on lower tiers (e.g. Redis on Kinsta, Smart Plugin Manager on WP Engine)
- Available on paper but limited in practice
- Recently changed or removed without prominent announcement
What We Don't Do
- We don't invent lab tests. "We tested 47 hosts over 6 months" is a claim most affiliate sites make without the infrastructure to support it. We don't make that claim.
- We don't cite figures we can't source. Every performance or pricing claim in a review traces back to a documented, verifiable source.
- We don't hide renewal prices. This is the most common trust violation in hosting reviews. We put renewal pricing prominently in every review.
- We don't recommend hosts we wouldn't use. Every host we actively recommend is one we'd point a client or colleague to without hesitation.
Affiliate Relationships — Full Disclosure
We earn commissions from most of the hosts we recommend when you sign up through our links.
Hosts we mention without an affiliate relationship — Hetzner, for example — are included when they're genuinely the right answer for a specific use case. Editorial mentions are not for sale.
Does affiliate status affect rankings? The honest answer is: it creates an incentive we actively manage. Our control is transparency, primary source data requirements, and willingness to write negative reviews and "who should not use this" sections that reduce conversions. A review that loses you a reader who would have been a bad fit is better for long-term trust than one that converts everyone and generates refund complaints.
Update Schedule
- Review pages — re-verified quarterly. Each shows a "Last Verified" date.
- Comparison pages — updated when pricing or features change materially, or quarterly at minimum.
- Articles — updated when the underlying facts change. Date shown in each article.
If you spot outdated information, use the contact page — we take corrections seriously and update promptly.
Who Writes This
FBWH is written by Tom George, a hosting analyst with 18+ years of hands-on experience. Full background on the author page.