hosting
GET /v1/sites/{siteId}/vitals
Get a site's homepage Core Web Vitals.
Authentication
Send an API key as a bearer token. The endpoint checks sites.view; a key without it is refused with 403, not 404.
This endpoint takes no organisation id. Your key already identifies the organisation it belongs to, and the response is scoped to it.
Try it
Replace anything in angle brackets with your own values, and the key placeholder with a key from your dashboard.
curl -X GET https://api.zinndigital.com/v1/sites/{siteId}/vitals \
-H "Authorization: Bearer zdk_live_…"Signed in? The API console in your dashboard fills in your real organisation id and your own key, and runs the request against the live API so you can see the actual response. Open this endpoint in the API console
Details
The site's newest Core Web Vitals reading — LCP, CLS and INP for its homepage, with the "good / needs improvement / poor" band for each and for the reading as a whole. Backs the Performance tab's vitals card and the free SEO Basics tile, both of which are sold as included on every plan and every site. RLS-scoped to a site the caller can view (sites.view); an out-of-scope or unknown id is a 404, exactly like getSite. Homepage only, by design. Every surface that sells this sells homepage Core Web Vitals. Measuring every URL of every site is a different product. Collection is asynchronous. Readings are taken off the request path by a scheduled collector and read back here — a live Lighthouse run would be a 20-second uncached external call on a tab that mounts for every site (CLAUDE.md §2.16). Until a site has been measured, available is false, every metric is null and every band is unknown. source must be rendered. A lab reading is a synthetic Lighthouse run; a field reading is what real Chrome users experienced (Chrome UX Report) and is what Google ranks on. Field data does not exist for sites below CrUX's traffic threshold — which is most low-traffic sites — so lab is the coverage tier. Presenting a lab score as though real users produced it is a false claim about what visitors experienced, so clients must label which one they are showing.
Parameters
| Name | Type | Required | What it is |
|---|---|---|---|
siteId (path) | Uuid | Yes | Site ID (UUIDv7). |
Response
| Name | Type | Required | What it is |
|---|---|---|---|
available | boolean | Yes | Whether any vitals reading exists for this site yet. |
url | string | Yes | The exact URL measured. Empty when unavailable. |
source | string<field, lab, > | Yes | field = Chrome UX Report (real users, what Google ranks on); lab = a synthetic Lighthouse run. Clients MUST label which is shown. Empty string when unavailable, following… |
field_available | boolean | Yes | Whether real-user (field) data exists for this site at all, independent of which reading is returned — so the UI can say "no real-user data yet" while still showing the lab number. |
lcp_ms | object | Yes | Largest Contentful Paint, p75, milliseconds. Good ≤ 2500, poor > 4000. |
cls_milli | object | Yes | Cumulative Layout Shift, p75, in thousandths (0.083 → 83). An integer so no band comparison is ever made against a float at the 0.1/0.25 boundaries; clients render… |
inp_ms | object | Yes | Interaction to Next Paint, p75, milliseconds. Good ≤ 200, poor > 500. INP replaced First Input Delay as a Core Web Vital in March 2024; FID is retired and is deliberately not… |
ttfb_ms | object | Yes | Time to First Byte — a diagnostic, not a Core Web Vital. |
fcp_ms | object | Yes | First Contentful Paint — a diagnostic, not a Core Web Vital. |
performance_score | object | Yes | Lighthouse performance score, 0–100. Always null on a field reading — the Chrome UX Report computes no score, and inventing one would present a number Google never produced. |
accessibility_score | object | Yes | Lighthouse accessibility score, 0–100 (docs/85 §2.1). Lighthouse computes all four of its categories from a single page load, so this costs about a second on a 10–30 s run —… |
best_practices_score | object | Yes | Lighthouse best-practices score, 0–100 — mixed content, deprecated APIs, console errors. null on a field reading. |
seo_score | object | Yes | Lighthouse SEO score, 0–100. null on a field reading. |
stack_packs | string[] | Yes | The technology stacks Lighthouse detected on this run (wordpress, woocommerce, …). Ids only. Lighthouse also ships stack-specific remediation text; that is untranslated… |
band | VitalsBand | Yes | A Core Web Vitals verdict against Google's published p75 thresholds. The overall band is the worst of the three measured metrics, because that is how Google assesses a page:… |
lcp_band | VitalsBand | Yes | A Core Web Vitals verdict against Google's published p75 thresholds. The overall band is the worst of the three measured metrics, because that is how Google assesses a page:… |
cls_band | VitalsBand | Yes | A Core Web Vitals verdict against Google's published p75 thresholds. The overall band is the worst of the three measured metrics, because that is how Google assesses a page:… |
inp_band | VitalsBand | Yes | A Core Web Vitals verdict against Google's published p75 thresholds. The overall band is the worst of the three measured metrics, because that is how Google assesses a page:… |
measured_at | object | Yes | — |
opportunities | SiteVitalsOpportunity[] | Yes | What is actually making this page slow, largest saving first — the "why" behind the numbers above. ⚠️ An empty array does not mean "nothing to fix". It means this reading… |
screenshots | SiteVitalsScreenshots | Yes | What the page looked like while it loaded — the filmstrip and the finished render. available is reported separately from the parent envelope's flag because a perfectly good… |
Errors this endpoint can return
401 · 403 · 404 · 429