VideoObject Schema (Healthcare) — the markup that qualifies a video for three races.
VideoObject Schema (Healthcare) is the JSON-LD block Google requires on every embedded YouTube video before it enters the three ranking races (YouTube search + Google web + AI Overview). Required fields: uploadDate, duration, thumbnailUrl, contentUrl, name, description.
How it works.
A healthcare video without VideoObject is a video Google can play but cannot cite. The JSON-LD block sits on the embed page and tells Google six things: what the video is called, what it is about, when it was uploaded, how long it runs, where its thumbnail lives, and where its content stream lives. Those six fields turn a passive iframe into a citable knowledge asset. When AI Overview needs a visual answer for "how does root canal treatment work" — the AIO extractor scans the corpus for pages that pair a spoken-answer article with a properly-marked video. Pages that carry a valid VideoObject win the video citation; pages that do not, do not. On YouTube itself, native ranking signals dominate. On Google web and AIO, VideoObject is the entry ticket.
- name & description. Match the on-YouTube title and first paragraph verbatim, but rewritten to pass NMC and ASCI compliance filters.
- uploadDate. ISO 8601 date matching the actual YouTube publish date — drift breaks rich-result eligibility.
- duration. ISO 8601 duration string (PT3M42S). Wrong format is the single most common invalidation.
- thumbnailUrl. A stable image URL — ICG uses server-side maxresdefault fetches, not stale i.ytimg.com URLs.
- contentUrl & embedUrl. contentUrl points to the video source; embedUrl to the iframe URL. Both must be public and live.
- transcript & hasPart. Optional but decisive — a transcript pulls the entire spoken track into the AI Overview extraction pool.
Why it matters for
Indian healthcare.
VideoObject metadata is machine-readable and citable — which means every field is a compliance surface. Description text carrying "best" or "leading" fails ASCI. A description confirming diagnosis or promising a specific clinical outcome fails the NMC Code of Ethics. For fertility videos, success-rate language in the description violates the ART Act 2021 the moment AIO cites it. For dental practices, treatment-outcome claims in the description trip DCI ethics review. Transcripts inherit the same rules — a transcript that reads back patient names or clinical detail fails DPDP Act 2023 Section 8. ICG's VideoObject deployment on YODA runs every field through these filters before the JSON-LD ships. A schema pass is also a compliance pass.
How ICG uses this.
VideoObject is auto-generated by YODA the moment a video is embedded onto a clinic money page or insight article. The extraction pipeline pulls YouTube's official metadata (title, description, publish date, duration), runs a compliance-rewrite pass against NMC and ASCI, generates the thumbnail server-side, and merges the finished JSON-LD into the parent page's schema graph. Because every field is normalised, the same video embed can live on a specialty money page (yielding a Google web citation), inside a 6+ H2 article (yielding an AI Overview citation), and on a city landing page (yielding a local-video pack citation) — three races entered from a single deploy. The pairing with Speakable Schema means the article's H1 definition and FAQ answers become the spoken payload for the same query the video wins the visual answer to.
| Field | Required format | Healthcare filter |
|---|---|---|
| name | Verbatim YouTube title | ASCI superlatives stripped |
| description | 150-500 char plain text | NMC + ASCI + ART Act pass |
| uploadDate | ISO 8601 (YYYY-MM-DD) | Reconciled with YouTube API |
| duration | ISO 8601 (PT#M#S) | Auto-fetched |
| thumbnailUrl | Stable HTTPS image URL | Server-side hosted (not i.ytimg) |
| transcript | Full spoken track as text | DPDP Section 8 · PC-PNDT scan |
Frequently asked questions.
Do I still need VideoObject if the video is a YouTube embed?
Yes. YouTube supplies its own metadata for the YouTube search race, but the embed on your clinic website is a separate surface. Without VideoObject on the embed, Google web and AI Overview cannot associate the video with the parent page. The video and the page compete separately instead of compounding — and the AIO citation goes to the page that structured the pairing.
What are the required VideoObject fields for healthcare?
Six mandatory: name, description, uploadDate, duration, thumbnailUrl, contentUrl. Recommended: publisher, transcript, hasPart with clip timings. For healthcare, description must pass NMC and ASCI filters — no cure-rate claims, no unsubstantiable superlatives. Transcript pairing lifts the entire spoken track into the AI Overview extraction pool.
Does VideoObject help YouTube rankings on YouTube itself?
Not directly — YouTube search runs on YouTube-native signals. VideoObject helps the two other races: Google web (video-rich snippets in blue-link results) and AI Overview (video citation in the AIO carousel). Winning all three requires the schema on the embed page and a well-optimised video description + tags on YouTube itself. YODA runs both sides.
What breaks a healthcare VideoObject in production?
The most common breaks: uploadDate mismatched with actual YouTube publish date, duration in wrong ISO 8601 format, thumbnailUrl pointing to a stale i.ytimg.com URL, description carrying an ASCI-flagged superlative, and contentUrl pointing to a private or unlisted video. Any of these disqualifies the video from Google web and AIO citation while still playing normally to the visitor.
Read next.
Audit your VideoObject deploys
across the corpus.
ICG runs a free 30-min Brand & Growth Diagnostic that scans your video embeds, flags missing or malformed VideoObject fields, and shows exactly which videos are ineligible for AI Overview citation under current Google policy.
