Stunning Review: clear platform, slow delivery (72/100)

Field	Value
Domain	stunning.so
Category	AI-powered no-code web app builder for businesses
Pricing	Unknown; price range not provided
Business model	SaaS subscription
Pages crawled	38
Crawl date	2026-09-29

Stunning earns 72/100 because its core product story and technical foundation are strong, while delivery speed and decision guidance hold back the experience. First impressions and positioning scored 87/100, usability 85/100, design execution 87/100, and technical SEO 92/100. The site clearly targets businesses that want to automate operations without coding knowledge or setup complexity.

The main weakness is performance: mobile LCP reached 11.9 seconds, despite a 301ms TTFB. Accessibility scored 62/100, writing quality 62/100, and editorial QA 55/100, reflecting contrast, structure, repetition, and localization issues. Decision-support surfaces scored 45/100 because four pricing tiers are presented without guidance for matching a plan to a user or use case. Prioritize hero delivery, accessibility fixes, clearer plan selection, and editorial cleanup.

Stunning positions itself as an end-to-end no-code platform: “Stunning builds it, then runs it.” The claim is specific, and the review scores positioning 87/100, but the site does not clearly distinguish the product from Bubble or Webflow. It also does not specify whether its primary buyer is a small business or a larger enterprise. Keep the end-to-end promise, add plain category language such as “no-code platform for business systems,” and make the primary audience explicit. The brand name appears in the H1 while the domain is stunning.so, a low-severity recognition issue worth monitoring.

Stunning speaks to businesses that want to automate operations without coding knowledge or setup complexity. That audience direction is useful, but the site does not specify company size or industry, and it leaves important questions unanswered, including what happens when something goes wrong and why a buyer should trust the platform. The vocabulary includes database, admin view, and API without plain-language explanation. Add audience-specific context, explain those terms as they appear, and address support, security, and refund expectations where the product facts support them. The available Trustpilot signal is 4.4/5 from 207 reviews, but it does not replace specific customer outcomes.

Stunning is visually engaging, yet first-time visitors must work too hard to find pricing or identify the primary next step. The Pricing link is present in the main navigation, but the path runs through product pages, while “Start building free” leads to the product builder rather than pricing. Multiple calls to action, including “Start building free” and “Get started for free,” weaken the visual hierarchy. Put pricing at a direct top-level path, link to it from key hero areas, and select one primary conversion action. On mobile, group FAQ anchors such as #billing and #credits into clearer sections or accordions.

Accessibility scores 62/100 because the sampled site has structural, labeling, alternative-text, skip-link, and contrast failures. The DOM shows no main landmark, 27 inputs without accessible names, and 1 image missing alt text; no skip link was found across the sampled pages. These defects make the primary flow harder to navigate with assistive technology. Add a main landmark and skip link, label each input with a connected label or accessible name, provide accurate alternatives for informative images, and preserve empty alt text for decorative images. Recheck focus styles and contrast after these structural fixes.

Design execution is strong at 87/100, with a clear hierarchy and consistent responsive treatment. The measured weaknesses are concentrated in contrast and mobile controls. Body text computed as #6B7280 on white at a 3.4:1 ratio, below the 4.5:1 requirement for body text; testimonial text at #9CA3AF reaches only 2.8:1. Some “Read More” controls measure 40px by 28px, and mobile inputs use 12px text. Darken the affected text, increase interactive controls to at least 44px, and raise mobile input text to 16px. Keep the existing spacing and visual structure.

Mobile delivery is the clearest performance weakness: LCP reaches 11.9 seconds while TTFB is 301ms. The LCP element is a 1280×720 video poster, served at 482KB as /videos/hero-film-en-poster.webp, with no preload or fetchpriority=high. The homepage also has 41 blocking requests, including 883KB from Google Tag Manager and 634KB from Facebook, plus a total payload of 11.5MB. Preload the LCP poster, defer non-critical third-party scripts, split or remove unused JavaScript, and add explicit image dimensions. Measure each change against LCP and blocking-request counts.

Stunning has useful product detail, but the sampled copy is weakened by broad positioning, oversized paragraphs, repeated headings, and visible localization keys. Builder copy names the screens, the database behind them, and the checkout, plus integrations including Moyasar, Paymob, OTO, Bosta, and Wafeq. The homepage promises that a description becomes your website, app or system, but the lead should choose one outcome. Extraction records 1,704 words in one homepage paragraph and 2,507 words in one blog paragraph. Lead with one concrete outcome, split the long blocks into descriptive sections, keep one h1 per page, and replace keys such as faqPage.title with reader-facing English.

The pricing page shows four tiers—ECO, HOBBY, PRO, and TEAMS—but gives buyers no guidance on which plan fits a use case or audience. Generic “everything in X” claims blur the differences, while the comparison axes emphasize vendor features such as custom domains and email support instead of capacity, integrations, or total cost. Annual billing is also unclear, and the one-time top-up option is not explained clearly. Add audience-fit recommendations only where plan facts support them, replace inherited-feature shorthand with unique differences, surface monthly versus annual cost, and make the mobile comparison usable.

Stunning’s crawl and indexation signals are stable. robots.txt permits all paths, the sitemap contains 1,382 URLs, sampled pages return 200, and HTTP redirects cleanly to HTTPS. Noindex is scoped to auth-gated paths including /signup, /login/, /dashboard/*, and /licenses. Raw and rendered parity is 100%, so the sampled content is not dependent on client-side injection for indexing. Wayback snapshots remain consistent from January 2023 through June 2026. Risk work is now verification work: establish field performance data and validate JSON-LD on pricing and builder templates, since those artifacts were not included.

Editorial QA scores 55/100 because the FAQ structured data contains Question nodes whose acceptedAnswer objects have no answer text, while the rendered page exposes unresolved localization keys such as faqPage.categories.billing.title. That creates a mismatch between structured data and reader-facing content. The sample also flags 7 h1 elements on /pricing, 3 on /product/builder, and an 86-character integrations title; the blog title is 72 characters. Replace visible localization keys with reader-facing English, populate visible and structured answers together, reduce each page to one h1, and trim long titles. Broader quality conclusions are limited to the 38 of 40 pages crawled.

The technical foundation scores 92/100. robots.txt allows all paths, the sitemap contains 1,382 URLs, and noindex is correctly limited to auth-gated surfaces. Raw-versus-rendered counts show 0% JS-only content across sampled templates; for example, /pricing records 1,607 words in both forms. HTTPS is enforced through HTTP-to-301-to-HTTPS, TLS is valid until November 2026, and standard security headers are present. /llms.txt returns 200 with 3,971 bytes. Maintain the current crawl and rendering setup, then validate JSON-LD on /pricing and /product/builder and connect field data to complete the technical picture.

Stunning presents a no-code platform for businesses that want to build web applications, databases, and business systems from descriptions. Its strongest evidence is the combination of 87/100 positioning, 87/100 design execution, 85/100 usability, and a strong technical foundation.

Three fixes would materially improve the product entry experience: reduce the 11.9-second mobile LCP, repair accessibility issues including missing landmarks and unlabeled inputs, and add plan-selection guidance above the four-tier pricing grid. The copy also needs editorial cleanup where long paragraphs, repeated structures, and localization issues weaken clarity. This is best suited to businesses evaluating no-code operations automation; buyers should validate performance and plan fit before committing.

This 13-dimension review combines a 38-page crawl from 2026-09-29 with the public dimension score table. The review covers positioning, messaging, usability, accessibility, design execution, performance, writing quality, decision-support surfaces, risk and stability, editorial QA, and technical SEO; scores are weighted into the overall 72/100 result. Search Console was not connected, which limits traffic-change diagnosis. Read How SiteList scores for the scoring method.