SEO & Search

Video Sitemaps for B2B Sites: Keep Page, Player, and File URLs Distinct

A fictional software publisher has a tutorial called “Build a supplier approval queue.” Its editor records four addresses before preparing a video sitemap entry.

Resource Fictional address
Watch page https://example.com/watch/supplier-queue/
Thumbnail https://example.com/media/supplier-queue.jpg
Embedded player https://example.com/player/supplier-queue
Media file https://example.com/media/supplier-queue.mp4

The addresses are invented. They represent one tutorial, with one specific video and one thumbnail. None has been requested or verified as a working resource.

The watch page is where the viewer arrives. The player presents the recording. The media file contains the video. The thumbnail represents it before playback. Keeping those roles separate is the first publishing check.

Google's video URL guide documents these distinctions. The worksheet below applies them to a publisher's review. Its process and fictional decisions are editorial proposals, checked against Google Search Central documentation on October 28, 2026.

Establish the publishing record before writing XML

Give the recording an internal identity that the editor and developer can both recognize. In this example, use supplier-queue-v1. This is a proposed inventory field, not a Google sitemap tag.

Record the approved visible title and description beside that identity. Add the watch page where the video is intended to be the main activity. Keep the player and media fields separate even when the publishing system provides both.

Ask who owns each address. A publisher can control its tutorial page while a video host controls the player. A media team may supply the thumbnail without controlling either delivery system. Assign unresolved questions to the person who can verify them.

Avoid building the record from a search result or an old spreadsheet alone. Open the current published page. Compare its actual references with the media record. A familiar slug is a clue to inspect, not proof that two resources represent the same recording.

Keep the B2B video marketing plan alongside the publishing record when deciding which tutorials deserve distribution. This article covers the identifiers and metadata used for discovery. It does not evaluate YouTube performance.

Map the record to the documented fields

Google's video sitemap reference defines required video tags and the media-or-player alternative. Use that reference when reviewing the exported file.

Field Meaning in this example Review condition
loc The tutorial watch page Names the containing page
video:video The video's metadata container Describes the specific embedded recording
video:thumbnail_loc The tutorial's thumbnail image Required image reference
video:title The recording's title Required; recommended to match the visible video title
video:description The recording's description Required; agrees with the visible description
video:content_loc The actual media file Supply this or video:player_loc; use a known file address
video:player_loc A player for this specific recording Alternative when the media file address is unavailable

The reference recommends the media file address when possible. It allows the specific player address as the alternative. Both must differ from the parent page's loc. A general channel page is not the specific player in this mapping.

The required description has a maximum of 2,048 characters. It must match the on-page video's description in meaning, without requiring identical wording. Google recommends matching the title displayed where the video is embedded. Escape HTML entities or use a CDATA block, as the reference directs.

Keep optional fields outside the first review unless the publisher has verified values for them. A plausible duration, rating, or publication date is still invented metadata. An empty field in the internal worksheet is preferable to a fabricated public value.

Read one illustrative entry

This entry uses the fictional page and media file. It chooses video:content_loc, so a separate video:player_loc is unnecessary for this example.

<?xml version="1.0" encoding="UTF-8"?>
<urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9"
        xmlns:video="http://www.google.com/schemas/sitemap-video/1.1">
  <url>
    <loc>https://example.com/watch/supplier-queue/</loc>
    <video:video>
      <video:thumbnail_loc>https://example.com/media/supplier-queue.jpg</video:thumbnail_loc>
      <video:title>Build a supplier approval queue</video:title>
      <video:description>A fictional screen demonstration of queue setup and review.</video:description>
      <video:content_loc>https://example.com/media/supplier-queue.mp4</video:content_loc>
    </video:video>
  </url>
</urlset>

Read the entry from the outside inward. The url block names the page. The video:video block describes the recording embedded there. The thumbnail points to an image. The final field points to the video's content bytes.

This separation lets a reviewer ask a precise question about each resource. A correct page location cannot compensate for a thumbnail showing the wrong recording. A correct title cannot turn a player interface into a media file.

For a host that supplies only an embeddable player, replace the video:content_loc element with video:player_loc. Its value would be the verified specific player address. Google's reference and hosting examples document that alternative, including third-party iframe players.

Do not derive a file address from a third-party video identifier. Ask the provider what it exposes. Record the limitation when the actual media location remains unavailable.

Check identity before checking syntax

A parser can accept XML whose fields describe different recordings. The publisher still needs a content review.

For the fictional tutorial, compare the visible title, thumbnail, player selection, and description against supplier-queue-v1. Imagine the media library contains a second recording about supplier rejection. A similarly named image from that recording would be a content mismatch even if its URL worked.

Have the editor describe the first scene and the task shown in the recording. Compare that observation with the thumbnail and description. Keep the comparison short enough that another reviewer can repeat it.

Now verify the page location. A service page may also embed the tutorial, but it serves a different reader task. The example's loc identifies the intended watch page. It does not identify the MP4, the player interface, or the media library.

Google's video sitemap best practices exclude unrelated videos and small addenda. Its video guidance separately addresses watch-page purpose. Confirm that purpose before treating a technically complete entry as ready.

Investigate four hypothetical defects

The thumbnail requires an editor session. The fictional page loads for the public, but the image was copied from a private preview. Record the actual response condition. Ask the media owner for an approved public thumbnail. Keep restricted assets restricted until publication is authorized.

The media URL expires. The developer confirms that a signed preview address stops working after its documented lifetime. Ask which stable published address is supported. Do not remove a token and assume the shortened address works. Record the provider and configuration behind the answer.

The title describes an older recording. The page now demonstrates queue review, while the sitemap still names queue creation. Ask the editor which recording belongs on the page. Update the approved publishing record before replacing the exported metadata.

The parent location names the wrong page. An exporter used the MP4 address as loc and the watch page as content_loc. Trace both values to their source fields. Correct the field mapping in the publisher's normal review process. Editing one output entry alone leaves the source mistake unresolved.

These scenarios are teaching examples. They are not findings about a live site. Google's video guidance supports the access, stable asset, and resource-role checks. The proposed owner assignments and repair sequence belong to this worksheet.

Test access at each resource boundary

Google's video sitemap requirements require referenced files to be accessible to Googlebot. A valid exported address does not establish that access.

Record page, thumbnail, and media or player checks separately. Capture the date, session condition, response, and resource observed. An ordinary public request can identify a local delivery problem. It does not prove what Googlebot received.

If a provider controls delivery, record which question needs its answer. “Host is configured” says little about the specific thumbnail or recording. Ask for evidence tied to the address in the sitemap.

For an expiring resource, preserve the known lifetime and the follow-up observation. Avoid copying live access tokens into a widely shared worksheet. The reviewer needs the delivery condition and authorized evidence, not another reusable credential.

Use the image sitemap and CDN audit when tracing broader asset-host ownership. Keep video field mapping in this record so an image-only inventory cannot obscure the player and media distinction.

Give the publisher a repeatable review checklist

Use one row per recording-page relationship. Add an evidence link or note for each answer. The checklist is a proposed publishing method, not an official eligibility score.

Review area Publisher question Completion record
Identity Do page, thumbnail, player, and media describe the approved recording? Media identity and editor confirmation
Page location Does loc name the intended containing watch page? Exact page address and purpose
Media alternative Is the known file or specific player mapped to its correct tag? Field choice and source evidence
Access Can each intended public resource be reached under the checked conditions? Separate dated observations and limitations
Metadata agreement Do title and description agree with the page and recording? Approved text and comparison result
XML Are namespace, required fields, and escaping correct? Parsed export and field review
Ownership Who resolves each unknown field or delivery question? Named owner and next action
Review date Which publication version does the record describe? Checked date and next review trigger

Run the review when the video, page, host, or thumbnail changes. Those are proposed triggers. Choose a periodic review interval that fits the publisher's release process rather than claiming Google requires this schedule.

Close the synthetic example without claiming discovery

For the supplier queue record, the editor confirms that all metadata describes supplier-queue-v1. The developer checks the field roles against the current export. The publisher records unresolved access evidence separately.

The illustrative XML contains one parent page and one video block. It includes the required thumbnail, title, description, and one permitted media-or-player choice. Its XML structure and resource roles can be reviewed independently of any deployment.

That is the scope of the example. The invented addresses prove no delivery behavior. A real release would require authorized checks against the publisher's actual resources and current configuration.

Google's sitemap overview describes sitemaps as discovery aids and states that inclusion guarantees neither crawling nor indexing. Keep that limitation in the completion note. A clean export is evidence about the export.

Use the technical SEO audit checklist for the wider site review. Leave this handoff with the approved resource map, the checked export, and the remaining questions.

Keep each address in its proper field.

Sources

Topic map: Technical SEO, crawling, and migrations

Articles in this cluster, then the 3 clusters that share the most topics with it.

This cluster: Technical SEO, crawling, and migrations

Neighbor: SEO by industry and location

Neighbor: Search measurement and reporting

Neighbor: Topical authority and keyword strategy