Your hosting choice can be the difference between a PageSpeed score of 45 and one of 92. Server response time (TTFB), the web server software, and built-in caching are the server-side factors that most directly affect your PageSpeed Insights score.
What does PageSpeed actually measure — and what part does hosting control?
PageSpeed Insights (PSI) assesses your Core Web Vitals and lab metrics using Lighthouse. Not every metric is under your hosting control, but a significant portion is:
- TTFB (Time to First Byte): How long the server takes to respond to the first request. Hosting has direct influence here.
- LCP (Largest Contentful Paint): How long it takes to render the largest visible element. Depends on TTFB plus asset delivery.
- Total Blocking Time (TBT): Tied to JavaScript execution — not the hosting itself.
- INP (Interaction to Next Paint): UI responsiveness, primarily a client-side code concern.
Bottom line: hosting directly controls TTFB and static file delivery speed. A well-configured host can add 15–30 points to your PageSpeed score without a single line of code change.
Hosting factors that raise (or tank) your PageSpeed score
1. Storage type: NVMe vs. SATA SSD
An NVMe server answers database queries and reads PHP files 4–7× faster than a SATA SSD server. This directly lowers TTFB. Look for providers that explicitly say "NVMe" — not just "SSD."
2. Web server: LiteSpeed vs. Apache vs. Nginx
LiteSpeed Enterprise has its own built-in cache (LiteSpeed Cache) that can serve cached pages with a TTFB under 50 ms. Apache with mod_cache works but is slower. Nginx with FastCGI cache is a solid option for VPS environments.
3. PHP version and PHP-FPM
PHP 8.2 processes requests up to 30% faster than PHP 7.4. PHP-FPM with separate pools prevents a slow site from affecting others on the same server. Make sure your plan lets you select PHP 8.1 or higher.
4. Server-side caching
A page cache converts dynamic pages (WordPress, for example) into static responses served without running PHP or hitting the database. The best web performance results come from combining object cache (Redis or Memcached) with full-page caching.
5. Server location and CDN
A server in Dallas or Mexico City has ~10–20 ms of network latency to most Mexican users. A server in Europe can add 100–150 ms on top. Many providers bundle Cloudflare CDN, which moves static assets closer to users regardless of where the origin server sits.
Comparison: hosting features that impact PageSpeed by plan type
| Feature | Basic shared | Advanced shared | NVMe VPS |
|---|---|---|---|
| Storage type | SATA SSD | NVMe | NVMe |
| Web server | Apache | LiteSpeed / Nginx | Nginx / LiteSpeed |
| Page cache | Not included | LiteSpeed Cache | Configurable (Redis) |
| PHP 8.2+ | Varies | Yes | Yes |
| CDN | Not included | Free Cloudflare | Configurable |
| Expected TTFB (WordPress) | 300–600 ms | 100–200 ms | 50–150 ms |
| Estimated PageSpeed score | 50–70 | 75–90 | 85–98 |
Note: scores are estimates for a standard WordPress install without problematic third-party plugins. Your actual result also depends on theme and plugins.
What PageSpeed score do you actually need?
Google doesn't publish a minimum PageSpeed threshold for ranking, but real-world Core Web Vitals data does influence rankings. Here are practical targets:
- 90+ on desktop: Excellent. Very few pages reach this.
- 75–89 on desktop: Good. Sufficient for most businesses.
- 90+ on mobile: Exceptional — hard to achieve without optimized hosting plus clean code.
- 60–74 on mobile: Acceptable if real Core Web Vitals are "good" (green in Search Console).
- Below 50 on mobile: Red flag. Likely hurting conversions and rankings.
Hosting can take you from 50 to 80 on mobile relatively quickly. Going from 80 to 95 also requires image optimization, reducing third-party JavaScript, and inlining critical CSS.
How to evaluate a hosting provider before signing up
- Ask for the storage spec: If they only say "SSD," ask whether it's SATA or NVMe. If they don't know, it's probably SATA.
- Check the web server: Standard cPanel with Apache rarely breaks 75 on PageSpeed with WordPress out of the box.
- Look for reviews with measured TTFB: GTmetrix and WebPageTest report TTFB. Seek reviews with real data, not generic screenshots.
- Confirm the datacenter location: Mexico or Texas is correct for a Mexican audience.
- Check for LiteSpeed Cache or Redis access: These signal that the provider prioritizes performance.
If you need personalized guidance, the team at elenlace.com audits your current site and gives you a concrete roadmap for raising your PageSpeed score through hosting and code improvements.
Key takeaways
- Hosting directly controls TTFB, the single most important server-side factor in your PageSpeed score.
- NVMe storage + LiteSpeed Cache + PHP 8.2 is the combination that most effectively raises server-side scores.
- A well-configured advanced shared plan can reach 75–90 on desktop with WordPress.
- Reaching 90+ on mobile requires combining good hosting with code optimization (images, JS, critical CSS).
- Server location matters: choose one in Mexico, Texas, or Florida for a Mexican audience.
- Always verify storage type and web server before committing — don't rely on marketing copy alone.
Want to know exactly what's dragging your PageSpeed score down and how to fix it? elenlace.com provides performance audits and hosting migrations for sites targeting the Mexican market — reach out for a specific, actionable diagnosis.
FAQ
Does a high PageSpeed score guarantee better Google rankings?
Not directly. Google uses real-user Core Web Vitals data (not the Lighthouse lab score) as a ranking signal. However, low TTFB and fast LCP do correlate with better positions, especially when two pages are tied on content relevance.
Can my current hosting improve my PageSpeed without switching providers?
Sometimes yes. You can install a caching plugin (LiteSpeed Cache, W3 Total Cache), enable Gzip/Brotli compression, and turn on HTTP/2. If the provider offers Redis or Memcached as an add-on, enabling it can add 10–15 points. But if your base TTFB exceeds 500 ms, that signals a hardware or server saturation problem that no plugin can fix.
Does Cloudflare help improve PageSpeed?
Yes, significantly for static assets. Cloudflare serves images, CSS, and JS from its CDN, reducing latency for users across Mexico. It also automatically enables Brotli compression and can convert images to WebP via Cloudflare Polish. However, it doesn't reduce the TTFB of the main HTML document if the origin server is slow — that still requires server-side page caching.
How long does it take for PageSpeed to improve after switching hosting?
Lab metric changes (Lighthouse) are immediate — they show up in the first audit after migration. Field data in Search Console (real Core Web Vitals) takes 28 days to update, since Google accumulates real user data over that rolling window.
Useful resources
Other providers and guides worth comparing:
