How we reach a verdict

What we do

We do the comparison work that a buyer does not have time to do, and then we commit to an answer. In practice that means reading the vendor's own pricing and documentation properly — including the renewal terms, the fee schedules and the fair-usage clauses that the marketing page keeps below the fold — working out what the thing actually costs over the life of the decision, and naming the product we would tell a friend to buy.

Every factual claim we publish carries a source link and the date we checked it. If we say a plan renews at $16.99, there is a link to the page where the vendor says so. You should not have to trust us; you should be able to check us in about ninety seconds.

What we don't do

We do not run a hardware lab. We have not load-tested these platforms at 500 concurrent users, and you will never see us claim a benchmark we did not run. Plenty of affiliate sites do exactly that — plausible-sounding numbers, a confident tone, no evidence — and it is the single most common lie in this industry. A fabricated benchmark is worse than no benchmark, because it launders a guess into a fact.

So when we say a product has a weakness, it is because the vendor's own terms say so, or because a named and linked source reported it. Not because it sounded right.

We total the money, and we show the working

A monthly rate is not a cost. This is the mistake the entire hosting-comparison industry is built on, and it is the reason a $3.79 plan and an $11 plan can quietly change places. So every price on this site goes into a cost model — the promotional rate, how many months it lasts, what happens at renewal — and the site computes the total rather than letting anyone assert it in prose.

The chart on each comparison is that computation: what you will actually have paid, month by month, out to ten years, with the point where the two lines cross marked on it. That crossover is usually the single most decision-relevant number on the page, and it is the one nobody publishes.

We do this because prose lied to us. The first draft of the Hostinger verdict argued Cloudways was cheaper "from the first renewal onward" — a sentence that is true of the monthly rate and false of the money. The arithmetic caught it: totalled properly, Hostinger stays cheaper for 106 months. The verdict was reversed before you ever read it. Numbers you can recompute do not let you get away with a confident sentence.

Then we name one product

Not a shortlist, not a table of green ticks — one. And because a recommendation with no boundary is indistinguishable from an advert, every verdict is published alongside the specific circumstances in which our own pick is the wrong one. If we cannot describe the case where we would tell you to buy the other thing, we do not understand the products well enough to publish yet.

How we are funded

Affiliate commissions. When you buy through some of our links, the vendor pays us and you pay exactly the same price you would have paid anyway.

This is a real conflict of interest, and we are going to be more direct about it than is comfortable, because the alternative is performing a neutrality we do not have.

As of today, every winner on this site is a product that pays us.Three verdicts, three commissionable picks. We are telling you this rather than hoping you do not notice, because it is the single most reasonable thing to be suspicious about, and a site that hid it would deserve the suspicion.

Two facts are the only real test of whether that matters. In Kinsta vs Cloudways we picked Cloudways — the cheaper product, and the one that pays us less; Kinsta is the largest commission available to this site and it lost. And we have already reversed a published verdict against our own interest, because the arithmetic said we were wrong. Watch for whether that keeps happening. It is a better test than any promise we could make here.

Our answer is not “trust us”. It is the evidence section at the bottom of every page. The Cloudways verdict rests on one checkable fact — that Cloudways charges $11/mo for the same 2GB server DigitalOcean sells for $12 — and if that fact is wrong, the verdict collapses in public, in front of you. That is the point of publishing our reasoning and our sources instead of a score out of ten: a bought verdict has to visibly stop following from its own evidence, and you can see it happen.

Three commitments we will hold to. The winner is decided before the commercial arrangement is looked at. No vendor sees a verdict before publication, and no vendor can pay for one. And when a product we earn nothing from is the right answer, it will win and we will say plainly on the page that we make nothing from recommending it — the site is built to display exactly that, and if it never once happens, you should stop believing us.

When we are wrong

Prices change, plans get restructured, and vendors reverse policies. Every page carries the date each claim was verified. If a source has moved on and we have not, the verdict is stale — and a stale verdict on a page like this is a failure, not a technicality. If you catch one,tell us and we will fix it, update the date, and say what changed.