| Fact | Value |
|---|---|
| Domain | genjutsuai.io |
| Category | AI video transformation for creators working from existing footage |
| Pricing | Credit-pack AI video generation with one-time payments; price range not provided in the crawl |
| Pages crawled | 40 |
| Crawl date | 2026-11-08 |
Genjutsuai Review: strong video product (78/100)
Genjutsu AI scores 78/100 for a clear video-transformation offer with strong usability and accessibility. Mobile performance remains uneven: two pricing pages recorded 5.6 s and 5.3 s largest-content times against a 2.5 s target.
Reviewed by SiteList Engine · 12 of 13 dimensions · published Reviewed on November 8, 2026
Is this your site?
Claim itQuick facts
- Pages crawled
- 40
- Crawl date
- 2026-11-08
- Category
- AI video transformation for creators working from existing footage
Executive summary
Genjutsu AI earns 78/100 by making its core video-transformation job clear and keeping key paths short. Usability is the standout dimension at 96.2/100, supported by direct routes to pricing, the workspace, and support; accessibility scores 91/100, and technical SEO scores 93.7/100. Writing quality is also strong at 84.7/100, with concrete explanations of Motion Transfer, Object Swap, input limits, and cost uncertainty before generation.
The main weaknesses are practical rather than foundational. Docs & self-serve help scores 37.8/100, while performance scores 65.9/100: mobile pricing pages recorded 5.6 s and 5.3 s largest-content times. Positioning and audience messaging are fair at 69.7/100 and 68.3/100, while decision-support surfaces score 68/100. The product has a credible base; clearer audience language, stronger outcome-led proof, public help content, and faster mobile pages would make decisions easier.
- Overall score
- 78/100
- Usability
- 96.2/100
- Accessibility
- 91/100
- Docs & self-serve help
- 37.8/100
- Performance
- 65.9/100
01 · First impressions & positioning — two clear modes, no named creator segment
Genjutsu AI is immediately legible as a browser-based video transformation tool for people who already have footage. The hero names Motion Transfer and Object Swap beside the promise to transform existing footage, while pricing gives a concrete operating detail: 480p uses 60 credits per second and 720p uses 130. That makes the mechanism and cost model clear, but the buyer remains broad; the copy says “your next video project” without naming a creator role or production context. The gallery shows “Community example” and “Source clip” cards, not attributable outcomes, and the crawl found no comparison or alternatives pages. Name the primary creator segment, then turn two or three gallery examples into mini case studies with the input, transformation brief, output, and result.
- Hero promise
- Transform existing footage with Motion Transfer or Object Swap
- Credit rate
- 480p uses 60 credits per second; 720p uses 130
- Crawled surface
- No comparison or alternatives pages discovered
02 · Audience & messaging — cost is concrete, persona language is broad
The message gives a creator with existing footage a workable mental model: upload video and reference images, choose Motion Transfer or Object Swap, preview the quote, and pay once. The audience language is less decisive. “Creative experiments,” “your next video project,” and “more videos and variations” describe activity without choosing a role, team, or production context. Pricing answers commitment questions better than proof questions: it gives pack amounts, resolution allowances, per-second rates, rounding, a final quote before generation, and release of credits after failed or rejected requests. The gallery still teaches possibility more than selection because its examples do not state the input, intended change, or preservation trade-off. Add one named persona and annotate three examples with the change, output, and limitation.
- Core workflow
- Add video, add images, choose a mode, and generate
- Pricing guidance
- Final quote before generation; credits released after failed or rejected requests
- Audience cue
- Copy says “creative experiments” and “your next video project” without naming a role
03 · Usability — one-click pricing, direct support, fragmented start path
Genjutsu AI gives first-time visitors short routes to the core decision: the homepage explains the offer within 10 seconds, Pricing is one click away, and the pricing page shows four packs with prices, credits, allowances, and purchase buttons. The calculator connects duration, resolution, and credits, so visitors can estimate a job before opening checkout. Support is also clear once found: the contact page offers “Email support contact@genjutsuai.io” and lists what to include for payment, credit, video, and content issues. The main friction is terminology: the homepage says “Open workspace” and “Generate,” while pricing says “Get Trial,” “Get Mini,” “Get Starter,” and “Get Creator.” Add one sentence that explains whether to open the workspace or buy credits first, and expose Contact in primary or utility navigation.
- Pricing route
- Homepage Pricing link reaches /pricing in one step
- Support action
- Email support contact@genjutsuai.io
- CTA vocabulary
- Open workspace / Generate versus Get Trial, Get Mini, Get Starter, Get Creator
04 · Accessibility — 96 Lighthouse, with contrast and label failures
Genjutsu AI has strong structural accessibility in the sampled public pages, but two homepage Lighthouse failures need direct remediation. The rendered samples expose header, navigation, main, and footer landmarks; the homepage reports 64 images with zero missing alt attributes, zero positive tabindex values, and a skip link. Lighthouse measured 96 on the homepage and 100 on Korean pricing. It nevertheless fails color-contrast and label-content-name-mismatch, while the DOM records five homepage inputs and two controls needing individual accessible-name verification. The Korean and Portuguese samples report no skip link, though both expose a main landmark, so that remains a coverage caveat. Fix the returned contrast nodes and align visible labels with computed names, then run a live axe check and scripted keyboard probe for focus order, traps, announcements, and reduced-motion behavior.
- Homepage Lighthouse accessibility
- 96
- Homepage DOM
- 64 images, 0 missing alt attributes, 0 positive tabindex values, skip link present
- Failed audits
- color-contrast and label-content-name-mismatch
05 · Design execution — polished hierarchy, 67 mobile target warnings
The design has a coherent hero-to-product-to-pricing narrative: the homepage uses a single H1, moves through examples and workflow education, and keeps a consistent lime accent, neutral surface, rounded cards, and featured Creator plan. The mechanical debt is concentrated on mobile interaction and contrast. The 390px crawl reports 67 small tap targets, including a visible 40×40px header control below the 44×44px guideline; non-actionable 1×1 inputs, skip-link geometry, and clipped carousel cards should be excluded before triage. PSI also marks the homepage color-contrast audit as failed on desktop and mobile, without exposing the failing pairs. Re-run the target audit after enlarging actionable controls, export the failing contrast nodes, and confirm the 75 colors, 18 radii, and 13 font-size values against computed styles before consolidating tokens.
- Mobile small targets
- 67 at 390px; visible header control is 40×40px
- Contrast
- Homepage PSI color-contrast audit failed on desktop and mobile
- CSS inventory
- 75 colors, 18 radii, 13 font-size values
06 · Performance — 5.6 s mobile LCP on Turkish pricing
Mobile performance is inconsistent by route: LCP reached 5.6 s on Turkish pricing and 5.3 s on Korean pricing, versus 3.2 s on the homepage and 2.4 s on Portuguese pricing. The homepage also weighed 3,906 KiB in PSI, with an estimated 671 KiB of image savings and 15 blocking requests. Korean and Turkish pricing recorded CLS of 0.163 and 0.188, while lab server response stayed low at 7–8 ms, so rendering and asset cost are the useful first targets. Profile the LCP element on the shared localized pricing template, defer below-fold gallery work, serve responsive modern media, reserve shifting dimensions, and re-test each locale. Treat these as lab measurements, not a 28-day field verdict.
- Mobile LCP
- 5,616.039 ms /tr/pricing; 5,324.041 ms /ko/pricing
- Homepage payload
- 3,906 KiB total page weight; 671 KiB estimated image savings
- Mobile CLS
- 0.16289334776676412 Korean pricing; 0.18791769848347922 Turkish pricing
07 · Writing quality — concrete tutorial, uneven multilingual hygiene
The tutorial is strongest when it teaches a decision: start with a 4–5 second performance, work within a 4–30 second workspace range, use a 960 × 720 dancer test, and review the cost before generation. Pricing reinforces that discipline with 60 credits per second at 480p, 130 at 720p, and a final quote before confirmation. The editorial debt is mechanical and cross-locale. The crawler records 16 missing homepage alt attributes, two on pricing, and two on the tutorial; the gallery template route repeats title and meta description; Portuguese and Turkish navigation label the tutorial as English-only. Add descriptive alternatives, define canonical or noindex behavior for the query route, and give localized pages a short bridge summary before the English tutorial.
- Tutorial limits
- 4–5 second first attempt; 4–30 second workspace range; 960 × 720 test
- Missing alt attributes
- 16 homepage, 2 pricing, 2 tutorial
- Localized handoff
- Portuguese and Turkish navigation identify the tutorial as English-only
08 · Decision-support surfaces — four prices, no defended choice path
The pricing page makes the basic economics unusually legible: Trial, Mini, Starter, and Creator cost USD 4.99, 9.99, 14.90, and 29.90, with base and bonus credits, 480p/720p allowances, purchase buttons, and a calculator using 60 credits per second at 480p and 130 at 720p. The missing layer is guidance. Creator receives the strongest visual emphasis, but the page does not explain when it is worth the extra spend, when Starter is safer, or how a 720p workflow changes the choice. Repeated “Included with every pack” rows consume comparison space without distinguishing plans. Add an overridable chooser for a first experiment, one short project, or repeated variations; move universal benefits into a shared strip and use each card for credits, seconds, resolution, and finished-clip examples.
- Pack prices
- Trial USD 4.99; Mini USD 9.99; Starter USD 14.90; Creator USD 29.90
- Cost calculator
- 60 credits/second at 480p; 130 credits/second at 720p
- Comparison issue
- Universal benefits repeat across all four cards
09 · Risk & stability — 40 pages, one historical snapshot, no traffic baseline
The available evidence shows no active organic-visibility suppression, but it does show limited resilience and observability. The crawl covered product, tutorial, pricing, gallery, legal/contact, and example pages. Wayback returned only one homepage snapshot, dated 2026-10-05, so history is unknown rather than a clean multi-year record. Google Search Console and analytics are not connected, and no traffic-loss event is evidenced. Establish those baselines, keep a dated deploy and content-change log, and add differentiated workflow resources, limitation guidance, and creator case studies without duplicating product copy.
- Crawl coverage
- 40 pages across product, tutorial, pricing, gallery, legal/contact, and examples
- Historical evidence
- One Wayback homepage snapshot on 2026-10-05
- Observability
- Google Search Console and analytics not connected
10 · Editorial QA of content — scoped tutorial evidence, recurring template debt
The supplied prose clears a human-editor bar because it separates tested material from general product guidance. The tutorial labels “Our five-second comparison, tested October 6, 2026,” explains the additional Motion Transfer, Restyle, and local-editing steps, and keeps pricing specific at 60 credits per second for 480p and 130 for 720p. The remaining QA issues are reproducible across templates: missing image alternatives recur on the homepage, pricing, and tutorial; the gallery query route duplicates metadata; and Portuguese and Turkish navigation sends readers to an English tutorial. Add regression checks for alt text and query-route metadata, create a localized bridge summary, and retain explicit scope notes in future case studies so examples are not mistaken for universal claims.
- Tutorial scope note
- Five-second comparison tested October 6, 2026
- Repeated content QA issue
- 2 missing-alt images on tutorial; 16 homepage; 2 pricing
- Metadata regression
- Duplicate title and meta description on /gallery and its template-query URL
11 · Docs & self-serve help — one tutorial route
Genjutsu AI has a useful tutorial route. The tutorial carries BreadcrumbList markup. A linked Help or Docs index could cover getting started, Motion Transfer and Object Swap task guides, concepts, FAQ, troubleshooting, and escalation; connect it from navigation, pricing, and the workspace, then add a machine-readable index after the public help structure exists.
- Tutorial coverage
- Tutorial route at /genjutsu-ai-tutorial
- Tutorial structure
- BreadcrumbList markup present
12 · Technical SEO — 61 sitemap URLs, strong rendering, slow localized pricing
Genjutsu AI is broadly crawlable and server-rendered: all 40 crawled URLs returned 200, the verified sitemap contains 61 URLs, canonicals match fetched URLs, and HTTP/www requests consolidate to the HTTPS apex in one 308 hop. Raw and rendered content also match on sampled pages, including 1,309 homepage words and 1,416 tutorial words. The workspace gallery is intentionally noindex, follow and absent from the sitemap, which fits its logged-in surface. The material technical weakness is page experience: mobile LCP reached 5,324 ms on Korean pricing and 5,616 ms on Turkish pricing, with CLS 0.163 and 0.188. Fix the shared pricing template first, then verify HSTS and add an Organization entity only if it accurately represents the operating business.
- Crawl and sitemap
- 40/40 crawled URLs returned 200; sitemap contains 61 URLs
- Raw/rendered parity
- Homepage 1,309 words in both; tutorial 1,416 words in both
- Localized pricing experience
- LCP 5,324 ms /ko/pricing and 5,616 ms /tr/pricing; CLS 0.163 and 0.188
Verdict — 78/100: strong product, with help and mobile speed to fix
Genjutsu AI is a strong choice for independent creators and small video-project teams that want to transform existing footage with reference images. The product explains two concrete transformation modes, gives visitors direct routes to pricing and the workspace, and presents unusually concrete credit comparisons, including one-time USD prices, base-plus-bonus credits, 480p/720p duration limits, and a cost calculator.
The next fixes are clear: add a public help-center or docs subtree, improve mobile performance on pricing pages, and sharpen the primary audience and outcome-led proof. The 78/100 score reflects a polished, accessible, technically crawlable product surface whose buyer guidance is held back by documentation and mobile-speed gaps.
- Overall score
- 78/100
- Audience & messaging
- 68.3/100
- Decision-support surfaces
- 68/100
- Mobile largest-content times
- 5.6 s and 5.3 s
Methodology & data notes
This is a 13-dimension review based on the supplied public crawl, product facts, and public dimension score table. The crawl covered 40 pages on 2026-11-08. It records Genjutsu AI as a web product for transforming existing footage, with Motion Transfer and Object Swap described in the product facts.
Google Search Console was not connected and access was unavailable, so search-enrichment findings remain pending. The tutorial route at https://genjutsuai.io/genjutsu-ai-tutorial is noted separately. Dimension 13, Review-content integrity, was excluded as not applicable. Read How SiteList scores for the review method and data definitions.
- Review dimensions
- 13-dimension review
- Pages crawled
- 40
- Excluded dimension
- 13 · Review-content integrity (not_applicable)
- GSC status
- gsc_connected=false; gsc_access=false
Questions buyers actually ask
What does Genjutsu AI do?
Genjutsu AI transforms existing footage. Motion Transfer guides a new character or scene, while Object Swap replaces a selected element.
Who is Genjutsu AI for?
It is aimed at independent creators and small video-project teams transforming existing clips with reference images.
How is Genjutsu AI priced?
The site profile describes credit-pack AI video generation with one-time payments. The pricing page shows one-time USD prices, base-plus-bonus credits, 480p/720p duration limits, and a cost calculator; the crawl does not provide a resolved price range.
What tutorial resource does Genjutsu AI provide?
The crawl includes a tutorial route at https://genjutsuai.io/genjutsu-ai-tutorial.
How does Genjutsu AI perform on phones?
Performance is inconsistent on phones: the Turkish and Korean pricing pages took 5.6 s and 5.3 s to show their largest content, compared with a 2.5 s target.