Skip to main content

Guides

Answers to the problems people actually have

Each of these started as a recurring complaint in hosting communities rather than as a keyword. They are written to be read once and acted on — short answer at the top, method underneath, sources at the bottom.

6 guides · no gated content · no newsletter wall

Meeting a term for the first time? The hosting glossary defines everything these guides use.

All guides

  1. 01 6 min read

    Is your site slow because of your host, or because of your site?

    A 20-minute diagnostic that tells you whether your hosting is the bottleneck or your own site is — before you spend money moving, or months optimising the wrong thing.

    Short answer: Measure TTFB on the same URL cached and uncached. If both are fast, your problem is in the browser, not the server. If the uncached number is slow, it's your application. If even cached pages are slow, it's your host.

    Read the guide
  2. 02 5 min read

    What hosting actually costs: reading past the intro price

    Introductory hosting prices routinely double or triple at renewal. Here's how to work out the real three-year cost before you commit, and what to do if you're already caught.

    Short answer: The advertised price is usually a first-term discount. Calculate the 36-month total using the renewal rate, not the intro rate — for some hosts that turns the cheapest option into the most expensive one.

    Read the guide
  3. 03 5 min read

    How to switch hosts without losing rankings or going offline

    Changing host doesn't hurt your SEO — botched migrations do. A step-by-step method for moving with zero downtime, plus the four mistakes that actually cause ranking drops.

    Short answer: Google doesn't care who hosts you. Build the site on the new host first, test it there, lower your DNS TTL a day ahead, then switch — and keep the old host running for a week.

    Read the guide
  4. 04 5 min read

    Do you need a CDN, or a faster server?

    A CDN and a faster host fix different problems, and buying the wrong one is common. How to tell which bottleneck you actually have — with a worked example where the slower server delivers faster pages.

    Short answer: A CDN speeds up cached content for distant visitors. A faster server speeds up everything else — checkout, admin, search, logged-in pages. If your audience is far away, get a CDN. If your uncached pages are slow, no CDN will help.

    Read the guide
  5. 05 6 min read

    AI crawlers are eating your hosting resources

    Bots now generate more traffic than humans on much of the web. How to tell whether AI crawlers are behind your resource spikes, and what to do about it without disappearing from AI search.

    Short answer: If your CPU and bandwidth are climbing while human traffic is flat, check your server logs before upgrading your plan. AI crawlers are a documented cause, and rate-limiting them costs nothing.

    Read the guide
  6. 06 5 min read

    Static hosting vs managed WordPress: when each actually makes sense

    Static hosting is genuinely faster and often free — and it cannot run WordPress. A straight comparison of what you gain, what you give up, and why static platforms are not in our ranking table.

    Short answer: If your site is content that rarely changes and you are comfortable with a build step, static hosting is faster and cheaper — often free. If you need a cart, logins, comments or a non-technical editor, managed WordPress is still the right answer.

    Read the guide

Editorial policy

How we pick what to write

A topic has to clear three tests before it gets written. It has to be a problem people genuinely report having, in their own words, repeatedly. It has to be something people search for, so the guide can actually reach them. And we have to have something specific to say about it.

The third test is the one that rules things out. "Best hosting for beginners" passes the first two easily and fails the third — we would be writing filler that duplicates the ranking table. Individual provider reviews are deferred until we run original tests, because writing them now would mean restating other people's benchmarks as though they were ours.

Where a guide recommends a provider it uses the same disclosed links as the rest of the site, and the reasoning is always traceable to a column in the table. See the affiliate disclosure for what that means in practice.