How I test AI website builders

Ten tests, the same for every builder, each graded from 1 to 5 on a fixed scale. The overall grade is their average plus my verdict, which can move it by up to one point. Every test is run by hand on an account I use, the sites are published for real, and the measurements, screenshots and reports behind each grade go in the full test results published with the review.

Written by Miles Foster, Owner and tester, RankDude.

The RankDude mascot holding a laptop with a check on its screen

Each builder gets the same three written briefs: a bakery (five pages, a menu, opening hours and a contact form), a photographer's portfolio (a gallery and a booking form) and a small shop with ten products. Each brief contains 20 facts of its own (60 in all), and every builder gets the same briefs, the same photos and the same test domain. I publish all three sites, then run the editing, speed, SEO and support tests on what is actually live. Nothing is graded from a feature list or a press kit.

Each review states how the account was obtained: bought at the public price, used on a free tier, or provided by the vendor. A vendor-provided account never changes a grade, and a vendor never sees a review before it is published.

How I rate

Every tool gets one score out of 5.

First, I run the same 10 tests on every tool and give each one a grade from 1 to 5. The average of the graded tests is the starting point; a test that cannot be graded is left out of it, and the review says how many tests the score is based on.

Then I add my verdict: what the tests cannot measure, such as how the tool holds up after the test weeks, how billing and cancelling work in practice, and how it compares with the others in the category. My verdict can move the score up or down by one point at most.

Every review shows how much my verdict changed the score, and why, in its scores table.

What a grade means

5	Excellent
4	Good
3	Average
2	Below average
1	Poor

The final score is the average of the graded tests (ten when every test can be graded), to one decimal place, plus my verdict, which can move it by up to one point. If a test couldn't be finished properly it gets no grade, and the review says how many tests the score is based on. Some tests on a free plan can't connect a domain or sell; I test them again on a paid plan when I can.

The 10 tests

1.Prompt to Live Site

The clock starts when I submit the brief and stops when every page in the brief loads at a public address, my own fixes included. Each of the three briefs is timed; the median is the grade, and the time to the first AI draft is reported alongside.

5: median under 10 minutes
4: 10 to 19 minutes
3: 20 to 39 minutes
2: 40 to 89 minutes
1: 90 minutes or longer, or the site could not be published

2.Design & Mobile Layout

Every page of the three first drafts is checked at 360, 768 and 1440 pixels wide and on a real phone. Faults are counted: inconsistent spacing, no clear type hierarchy, text failing WCAG AA contrast (measured with a checker), placeholder text or images left in, images unrelated to the brief, horizontal scrolling, overlapping elements and tap targets under 44 pixels.

5: 0 to 2 faults
4: 3 to 5 faults
3: 6 to 10 faults
2: 11 to 20 faults
1: 21 or more faults

3.Brief & Copy Accuracy

Each brief holds 20 facts: names, prices, opening hours, products, services, contact details. Each is checked on the published first draft. Invented content (testimonials, awards, prices, staff, figures) is counted separately, and generic copy that could belong to any business is noted.

5: 58 to 60 right, nothing invented
4: 54 to 57 right, at most one invented item
3: 45 to 53 right, or two or three invented items
2: 30 to 44 right, or invented testimonials or reviews
1: under 30 right

4.Editing Freedom

Move a section, change the font site-wide, change the colour scheme site-wide, add a page and its menu link, edit the footer, add a form field, replace an image and set its focal point, change a button's link, duplicate a section, change the phone layout on its own, embed custom code, and regenerate one section from a written instruction. Each counts when done without a workaround or support.

5: all twelve
4: 10 or 11
3: 7 to 9
2: 4 to 6
1: 3 or fewer

5.Page Speed

The home page and one inner page of each live site are tested three times in Lighthouse with mobile settings: 18 runs. The median performance score is the grade; Largest Contentful Paint, Cumulative Layout Shift and Total Blocking Time are reported, with the page weight.

5: median 90 or more
4: 75 to 89
3: 50 to 74
2: 30 to 49
1: under 30

6.SEO Basics

Each point is checked in the published HTML and files, not in the settings screen: an editable title per page, an editable meta description, one H1 per page, editable alt text, clean custom URLs, an automatic sitemap.xml, an editable robots.txt, canonical tags, schema markup for the business or products, Open Graph tags, 301 redirects and a per-page noindex option.

5: all twelve
4: 10 or 11
3: 7 to 9
2: 4 to 6
1: 3 or fewer

7.Hosting, Domain & Selling

Each point is tried: the test domain connected with HTTPS working, form submissions arriving by email, a blog, AI writing help for posts, email marketing, bookings, products with variants, a test purchase through checkout, tax settings, shipping settings, digital products and built-in visitor analytics.

5: all twelve
4: 10 or 11
3: 7 to 9
2: 4 to 6
1: 3 or fewer

8.Export & Ownership

I try to export the site's code and run it on a plain static host, and to export pages, posts, products and form entries. I check that the domain can be moved out, and read the terms on who owns the content and what happens to the site and data when the plan is cancelled.

5: code export that runs elsewhere, all content exportable, clear ownership
4: all content exportable, no code export
3: partial content export
2: copy and paste only
1: no export, or terms claim rights over my content

9.Help & Support

The same three questions go to every builder: how to connect an outside domain, why a form message did not arrive, and how to change billing. Each is sent through the AI assistant and every human channel offered. I record the time to the first useful answer and whether the answer was correct when followed.

5: all three correct within 1 hour, with a human channel
4: all three correct within 4 hours
3: within 24 hours, or one answer wrong
2: over 24 hours, or two answers wrong
1: no reply or no channel

10.Pricing & Value

Every plan is read from the checkout, not the marketing page. The figure graded is the yearly cost of the cheapest plan that connects a custom domain and removes the builder's branding. The cheapest plan that can sell, its transaction fees, the renewal price against the first-year price and any AI credit limits are listed with the payment methods seen at checkout.

5: under $120 a year, no transaction fees, same price at renewal
4: $120 to $199 a year
3: $200 to $299, or renewal more than 50% higher
2: $300 or more, or transaction fees on every plan
1: poor value or hidden costs

Documented but not graded

Templates, multilingual sites, team access, app marketplaces and the quality of AI-generated images are checked and described in every review. When they work well they count in the written assessment, but they never add points to the overall grade.

Rules that apply to every test

Real use of the product on a real account and real published sites, never promotional descriptions or showcase galleries.
Identical briefs, tests, thresholds and sample sizes for every builder in the category.
Authorised accounts only; no password sharing, no bypassing of plan limits or bot checks, and test purchases only in test mode or refunded.
Every grade has evidence: a measurement, a screenshot, a report or a table, in the full test results.
A grade changes only when a retest produces new evidence; the review records the date and the reason.

Questions about a specific grade go to the contact page. Makers can submit an AI website builder for testing. No AI website builder has been through these tests yet; every one I follow is listed on the AI website builders page, and reviews appear there as tests finish.