How we choose
The same eight criteria, applied to every tool on this site, whether or not it pays us a commission. Most of the tools we cover pay us nothing and are listed anyway — including Zapier, whose partner program is closed to publishers, so it could never pay us however we wrote about it. Which tools do pay us is named in full on our affiliate disclosure, and marked on every link.
The eight criteria
1. Ease of use. How long it takes a first-time user to publish something real — not how many features exist. See the note below: this is the one criterion we have not yet completed for any tool.
2. Pricing and value. What the entry plan actually includes, and what triggers the jump to the next tier. We look for the gap between the advertised price and the real one.
3. Features that matter. Whether it does the core job well, rather than how long the feature list is.
4. Integrations. What it connects to natively, and where you'd need a paid connector to fill a gap.
5. Scalability. What happens at ten times your current list size, in cost and in performance.
6. Beginner friendliness. Quality of onboarding, templates and documentation for someone with no technical background.
7. Support. Channels available, which plans get them, and typical response expectations.
8. Free plan or trial. Whether you can test it properly before paying, and what the free tier genuinely covers.
How ratings are produced, and why there aren't any yet
A rating means we set up an account and built something real with it — usually a landing page, a form and a follow-up email — and recorded where it was slow, confusing or limited. Ratings come from that process, not from the vendor's feature list.
We have not completed that testing for any tool on this site, so there is currently no ease-of-use rating anywhere here. The comparison table used to carry a column of “not yet rated” cells; across every tool that was noise rather than honesty, so the column was removed and this note replaces it. Anything that would amount to a difficulty judgment of ours — including estimates of how long setup takes — is held to the same standard and left out.
One distinction worth naming. Where users at large visibly disagree about a tool, we may report that disagreement and say so plainly — for example, that reviews are divided on how intuitive a particular interface is. That is a fact about how a tool has been received, and it is often the most useful thing we can tell you before you try it yourself. It is not us issuing a rating, and we won't settle the argument for you until we've done the work. Where you see that phrasing, read it as “opinion is split, budget time to find out which side you're on”, not as a score.
We would rather publish nothing than a number we can't defend. When the testing is done, the ratings go in with the date they were produced.
What a commission does and doesn't change
Nothing above changes when a tool starts paying us. In practice that has meant leaving criticism in place after the money arrived, and we can point at the specific paragraphs: the credit-burn warning on Make, the contact-pricing warning on GetResponse, and the error-handling and concurrency warnings on Pabbly Connect are word for word what they were when none of those companies paid us a cent. The code comments on those pages say so, aimed at whoever edits them next.
It also means leaving recommendations pointing away from us. Our Kit vs GetResponse comparison still sends most readers to Kit, which pays us nothing. Our Zapier alternatives page tells you to stay on Zapier if it connects an app the others don't, and to self-host n8n if you want automation that is genuinely free.
And where we can't claim independence, we say so instead of implying it. In automation, Make and Pabbly Connect both pay us while n8n and Zapier pay us nothing — more commercial interest than we have in any other category. Our Pabbly vs Make comparison is the first page here where every option earns us a commission — so rather than pretend otherwise, it opens by telling you that, builds its case only on billing mechanics the vendors publish themselves, and points you at a third tool that pays us nothing and beats both for long workflows.
If you ever find a page here where the advice moved after the commission did, that is a failure of this method and we'd want to know — use the contact page.
What we will state confidently
Structure, not experience. Pricing, plan limits, fee models, what a free tier actually includes, what a plan caps — these are checkable against the vendor's own documentation, and they are what most buying decisions actually turn on.
Where sources genuinely conflict we say so rather than pick one. Teachable and Thinkific both restructured recently and reported figures disagree, so those two show ranges and send you to the vendor's pricing page. That is deliberate. The same applies where a vendor contradicts itself: Make advertises two different application counts on a single pricing page, and Pabbly states its annual discount two ways, so we tell you that rather than choosing whichever number suits the argument.
Pricing accuracy
Every price on this site carries a checked date. Software companies change plan limits, rename tiers and adjust pricing without warning — in the last year alone Make renamed its billing unit, Kit roughly doubled its entry price, Kajabi cut its entry contact allowance by 75%, and both Teachable and Thinkific removed their free plans.
And the harder problem: a price can stay put while the thing it buys changes. In August 2026 alone, Zapier altered how tasks are counted without moving a single published figure, n8n removed the workflow limits on its cloud plans while its prices held, and Pabbly restructured its subscription tiers outright. A re-check that only compares numbers would have passed all three. We re-check quarterly and re-read the billing terms rather than just the figures, but confirm current terms on the provider's own site before you buy.
What disqualifies a tool
We don't list tools we can't describe honestly, and we don't open a category until there are at least three tools in it — two isn't a comparison. Where a category has only one contender, we say what it fits rather than calling it best.