Manifest position 819 · CrUX rank bucket 1000
https://www.pinterest.com
Coverage complete
Coverage-complete audit of four representative public archetypes. Authenticated feed, pin creation, checkout and account settings were not covered because no test account or source was supplied.
Attempts
1 / 3
Judged checks
58 / 58
Blocked
0
Not run
0
This report has no published overall score. Blocked and not-run checks are not passes. A coverage-complete report means every check has a judged outcome; it does not mean every check passed.
All 58 atomic check outcomes
| Principle / check | Status | Confidence | Evidence or reason |
|---|---|---|---|
respect-user-preferencesrespects-color-scheme | issues | high | The dark-mode screenshot is visually identical to the light capture; DOM computed style reports a white html background and color-scheme: normal. F01 medium: The public experience does not adapt to dark colour preference. |
respect-user-preferencesrespects-reduced-motion | pass | medium | The emulated reduce condition matched and CSS disables transitions/animations while smooth carousel motion is limited to no-preference. |
respect-user-preferencesrespects-contrast | pass | medium | The high-contrast screenshot remained legible and CSS contains prefers-contrast adaptations for sign-in controls. |
implement-natural-interactionsview-transitions | issues | high | The CSS feature probe found reduced-motion, container-query, scroll-state, scroll-marker and anchor-positioning rules, but no view-transition rules on the sampled public shell. F02 low: Route and major state changes do not expose View Transition styling. |
implement-natural-interactionsscroll-driven-animations | pass | medium | CSS evidence uses animation-timeline:view(x) for carousel markers rather than a sampled scroll event animation. |
implement-natural-interactionsphysical-gestures | pass | medium | Carousel CSS uses native horizontal overflow, overscroll-behavior and scroll controls. |
provide-guided-navigationscroll-state-aware-chrome | pass | medium | CSS evidence uses container-type:scroll-state and target-current scroll markers for carousel position feedback. |
provide-guided-navigationanchored-positioning | pass | medium | CSS evidence uses anchor-name, position-anchor and anchor() positioning for carousel controls. |
provide-guided-navigationdirects-attention | pass | medium | Carousel markers and target-current styling provide position feedback; headings clearly segment landing sections. |
maximize-content-reduce-noiseno-intrusive-interruptions | issues | high | Desktop captures of the home and shopping pages show a centered modal plus full-page scrim before any interaction. Reject is available and visually comparable, but the content remains blocked. F03 medium: A consent modal obscures the primary page content on initial load. |
maximize-content-reduce-noisesemantic-dismissible-primitives | issues | high | DOM evidence reports no <dialog> while a fixed [role=dialog] fills the viewport; the sampled page also had no popover or details primitives. F04 low: Modal surfaces use custom role=dialog containers instead of the native dialog primitive. |
maximize-content-reduce-noisereduced-chrome | pass | medium | Outside required consent/auth surfaces, screenshots devote most viewport area to imagery, headings and primary actions with restrained borders. |
adapt-to-the-form-factorresponsive-no-horizontal-scroll | issues | high | Layout evidence measured scrollWidth 929px against a 360px visual viewport, 569px horizontal overflow; the mobile screenshot clips the hero heading and secondary CTA. F05 critical: The homepage overflows and clips core content at a 360px viewport. |
adapt-to-the-form-factorcomponent-level-responsiveness | pass | medium | CSS evidence contains multiple container-query breakpoints and cqw-based carousel sizing. |
adapt-to-the-form-factorinput-modality-aware | pass | medium | Login controls are at least 48px tall and the focused email field had a visible 2px blue outline. |
support-core-task-successclear-purpose-and-primary-action | pass | medium | The home hero states “Create the life you love on Pinterest” with a prominent join action; Ideas and Shopping have specific headings. |
support-core-task-successprimary-flow-completion | issues | high | The Ideas page displays useful categories behind a full modal login wall on first load, preventing the representative inspiration-browsing flow without account action. F06 high: Public discovery journeys are interrupted by an immediate account wall. |
support-core-task-successclear-system-state-and-recovery | issues | high | The login probe found email/password required=false, aria-invalid=false, and no role=alert or aria-live error after activating submit with empty values. Focus styling itself was visible. F16 high: Empty login submission provides no observable error or required-field state. |
be-fast-and-stablegood-core-web-vitals | issues | high | Lighthouse measured LCP 14,865ms, CLS 0.797 and TBT 509.5ms (performance score 0.19). A separate trace measured LCP 3,098ms and TBT 111.79ms, confirming run variability but still showing blocking work. F07 critical: Cold-load performance and visual stability are far outside good Core Web Vitals ranges. |
be-fast-and-stablevisual-stability | issues | high | Lighthouse measured LCP 14,865ms, CLS 0.797 and TBT 509.5ms (performance score 0.19). A separate trace measured LCP 3,098ms and TBT 111.79ms, confirming run variability but still showing blocking work. F07 critical: Cold-load performance and visual stability are far outside good Core Web Vitals ranges. |
be-fast-and-stableefficient-main-thread | issues | high | Lighthouse estimated 946,268 bytes of unused JavaScript and 27,543 bytes of unused CSS; trace recorded two long tasks up to 117.82ms. F08 high: Large unused JavaScript and long main-thread tasks burden startup. |
be-fast-and-stableefficient-resource-delivery | issues | high | Lighthouse measured LCP 14,865ms, CLS 0.797 and TBT 509.5ms (performance score 0.19). A separate trace measured LCP 3,098ms and TBT 111.79ms, confirming run variability but still showing blocking work. Images evidence found 27/35 oversized, 35/35 legacy JPG, 34 below-fold images not lazy-loaded, 11 missing srcset, and all 35 without explicit dimensions. F07 critical: Cold-load performance and visual stability are far outside good Core Web Vitals ranges. F14 high: Landing-page images are oversized, legacy-formatted and insufficiently deferred. |
be-fast-and-stabletrim-unused-and-duplicate-code | issues | high | Lighthouse estimated 946,268 bytes of unused JavaScript and 27,543 bytes of unused CSS; trace recorded two long tasks up to 117.82ms. F08 high: Large unused JavaScript and long main-thread tasks burden startup. |
be-inclusivenames-roles-labels | issues | high | The images primitive found 32 of 35 images without alt text on the landing experience. These visuals communicate inspiration categories and are not all safely decorative. F09 high: Most sampled images lack alternative text. |
be-inclusivesufficient-contrast | pass | medium | Lighthouse accessibility scored 0.96 with no color-contrast failures; the contrast screenshot remained readable. |
be-inclusivestructure-and-focus | pass | medium | Lighthouse accessibility scored 0.96 and the active login input showed a visible focus outline; skip-to-content exists. |
be-inclusivelegible-text | pass | medium | Desktop screenshots show readable line lengths, strong hierarchy and no sampled text overlap. |
be-inclusivezoom-reflow-targets-and-media | issues | high | Layout evidence measured scrollWidth 929px against a 360px visual viewport, 569px horizontal overflow; the mobile screenshot clips the hero heading and secondary CTA. F05 critical: The homepage overflows and clips core content at a 360px viewport. |
follow-best-practicesno-console-errors | issues | high | Lighthouse captured a Google FedCM NetworkError, two first-party 404 responses and an identity-provider sign-in error. F10 medium: The page logs runtime and network errors during a normal load. |
follow-best-practicessound-document-and-assets | issues | high | Images evidence found 27/35 oversized, 35/35 legacy JPG, 34 below-fold images not lazy-loaded, 11 missing srcset, and all 35 without explicit dimensions. F14 high: Landing-page images are oversized, legacy-formatted and insufficiently deferred. |
follow-best-practicesbrowser-platform-hygiene | pass | medium | Lighthouse best-practices scored 0.96; no permission prompt appeared on load and the viewport/doctype/charset checks passed. |
be-discoverabletitle-and-description | pass | medium | DOM recon found title “Pinterest” and a descriptive meta description; Ideas and Shopping have archetype-specific titles. |
be-discoverablecrawlable-and-mobile-friendly | issues | high | Discoverability evidence found only 1% rendered-word coverage in raw HTML, no raw h1, and isJsShell=true; the crawler screenshot says Pinterest requires JavaScript. Ideas and Shopping fare better at 48% and 62%. F11 high: The canonical homepage is effectively a JavaScript shell for non-JS crawlers. |
be-discoverablecanonical-and-indexing-signals | pass | medium | The homepage returned 200 outside the rate-limited HAR, exposes a canonical URL, and robots.txt is available with crawler rules. |
be-discoverablestructured-and-shareable-metadata | pass | medium | DOM recon found Open Graph title, description, image, URL and Twitter card metadata matching the landing page. |
be-private-and-securesecure-transport-and-headers | issues | high | Cookies evidence flags _b and _routing_id without Secure and all four cookies as SameSite=None; headers include CSP/HSTS/nosniff/frame protection but omit Permissions-Policy and retain unsafe-inline allowances. F12 high: Cookie and browser-policy settings leave avoidable security gaps. |
be-private-and-securedata-minimisation-and-third-parties | issues | high | Trackers evidence observed 10 third-party origins and known Pinterest tracking hosts; most requests were to pinimg plus Google/recaptcha dependencies before account use. F13 medium: The unauthenticated landing load has a broad third-party and tracking footprint. |
be-private-and-securein-context-permissions-and-modern-auth | issues | high | The login screenshot and DOM show email/password, Google and QR-code choices; no passkey/WebAuthn option was present on the representative authentication surface. F17 medium: The sampled sign-in flow offers passwords and federated/QR options but no visible passkey path. |
be-private-and-securedefensive-browser-policies | issues | high | Cookies evidence flags _b and _routing_id without Secure and all four cookies as SameSite=None; headers include CSP/HSTS/nosniff/frame protection but omit Permissions-Policy and retain unsafe-inline allowances. F12 high: Cookie and browser-policy settings leave avoidable security gaps. |
be-resilientprogressive-enhancement | issues | high | Discoverability evidence found only 1% rendered-word coverage in raw HTML, no raw h1, and isJsShell=true; the crawler screenshot says Pinterest requires JavaScript. Ideas and Shopping fare better at 48% and 62%. F11 high: The canonical homepage is effectively a JavaScript shell for non-JS crawlers. |
be-resilientresilient-runtime-behaviour | pass | medium | Consent and login overlays remained fully contained at the sampled desktop viewport and provided explicit actions. |
be-resilientoffline-and-installable | not-applicable | high | The sampled unauthenticated public content archetypes were judged as web pages rather than the installed authenticated app; offline installability was not required for these paths. |
be-resilientnetwork-and-http-failure-states | pass | medium | Despite optional identity and resource errors captured by Lighthouse, the sampled public shell remained rendered and actionable rather than blank. |
be-internationalisedlang-dir-and-logical-properties | pass | medium | The localized UK response declares html lang=en; sampled CSS uses extensive margin/padding inline/block logical properties. |
be-internationalisedlocale-aware-data | not-applicable | high | No dates, numbers, currencies, durations or calendar data appeared in the sampled public archetypes, so locale formatting was not applicable to visible content. |
be-internationalisedtime-zone-correctness | not-applicable | high | No event time, recurrence or time-zone-sensitive data appeared in the sampled public archetypes. |
be-trustworthyno-dark-patterns | pass | medium | The cookie modal presents Reject all, customise and Accept all at comparable prominence; login clearly links terms and free signup. |
be-trustworthyhumane-error-handling | issues | high | The login probe found email/password required=false, aria-invalid=false, and no role=alert or aria-live error after activating submit with empty values. Focus styling itself was visible. F16 high: Empty login submission provides no observable error or required-field state. |
be-trustworthytrustworthy-input-assistance | pass | medium | Login fields expose email/current-password autocomplete and the signup probe exposed email/new-password/bday tokens. |
be-trustworthysafe-commercial-and-account-flows | pass | medium | The sampled sign-in surface states its purpose, exposes recovery and alternate methods, and makes signup/terms visible. |
be-sustainableoptimised-assets | issues | high | Images evidence found 27/35 oversized, 35/35 legacy JPG, 34 below-fold images not lazy-loaded, 11 missing srcset, and all 35 without explicit dimensions. F14 high: Landing-page images are oversized, legacy-formatted and insufficiently deferred. |
be-sustainableno-wasteful-work | issues | high | Lighthouse estimated 946,268 bytes of unused JavaScript and 27,543 bytes of unused CSS; trace recorded two long tasks up to 117.82ms. F08 high: Large unused JavaScript and long main-thread tasks burden startup. |
be-sustainablethird-party-and-media-budget | issues | high | Trackers evidence observed 10 third-party origins and known Pinterest tracking hosts; most requests were to pinimg plus Google/recaptcha dependencies before account use. Images evidence found 27/35 oversized, 35/35 legacy JPG, 34 below-fold images not lazy-loaded, 11 missing srcset, and all 35 without explicit dimensions. F13 medium: The unauthenticated landing load has a broad third-party and tracking footprint. F14 high: Landing-page images are oversized, legacy-formatted and insufficiently deferred. |
be-agent-readystructured-agent-capabilities | not-applicable | high | This emerging capability was judged out of scope without declared agent-facing intent; absence is an opportunity rather than a defect under the contextual principle. |
be-agent-readyon-device-inference | not-applicable | high | No user task in the sampled public paths required on-device generation or summarisation; the contextual emerging check was judged not applicable. |
be-memory-efficientno-leak-under-repeated-interaction | issues | medium | After ten bottom/top scroll cycles, heap self size grew from 60.1MB to 66.8MB (+11.3%), node count from 897,011 to 1,058,064 (+18.0%), objects from 93,574 to 170,245, and closures from 48,729 to 62,773. This may include intentional lazy-loaded content, so confidence is medium. F15 medium: Repeated feed-style scrolling retains substantial additional heap. |
be-memory-efficientbounded-footprint | issues | high | The baseline heap summary contains 897,011 nodes, 3,902,046 edges and 60.1MB self size before an authenticated feed is opened. F18 medium: The initial heap/DOM footprint is heavy for a public landing page. |
be-memory-efficientno-detached-dom-or-unbounded-listeners | pass | low | Neither heap summary listed Detached* constructors among the largest retained populations; this is low-confidence because listener instrumentation was unavailable. |
Provenance
Canonical report: results/atomic/reports/0819-www_pinterest_com.json
Report SHA-256: 5bba4889850ba4177842d0e2344cb404dccb433db78f2e564e3e91034751c18f
Local retained report: runs/2026-07-17T17-27-24-856Z/atomic-reports/www_pinterest_com/2026-07-23T05-38-33-985Z/report.json
Local evidence root: runs/2026-07-17T17-27-24-856Z/atomic-reports/www_pinterest_com/2026-07-23T05-38-33-985Z
Catalog SHA-256: sha256:78ccfdb2d483f4c57d9dafed80fd86c6265585a56457c8dcfddc254b80fb44d7
Raw screenshots, HARs, traces, heaps, and other browser artifacts are retained at the local evidence root and intentionally are not committed. Artifact paths in the canonical report are relative to that root.