Skip to content
All field notes
Measurement
Published
Oct 25, 2026
Updated
Nov 1, 2026
Reading time
14 min read

VAST Impression Fires but ViewableImpression Does Not: A Debugging Guide

Debug missing VAST ViewableImpression callbacks with a practical workflow covering player support, OMID setup, XML, network requests, and receiver evidence.

AP
AdProof Research
Independent measurement desk
A video player produces an impression receipt while a separate viewability check leads to an unresolved empty receipt.
A received impression does not establish a viewability outcome. The amber receipt represents an unresolved observation, not proof of a failed player. Conceptual illustration.
In this guide (12 sections)

The ad plays, the VAST impression tracker fires, and the viewability tracker stays silent. Replacing the URL may not help: the player might not implement the event, the visibility conditions might never be met, or a separate verification integration might be missing.

The fastest useful investigation follows one playback through five boundaries: the XML received, the features supported, the signals observed, the request sent, and the receiving system's processing. This guide shows how to collect evidence at each boundary without forcing a viewability event that did not occur.

A developer described this exact symptom in a Stack Overflow question about VAST ViewableImpression callbacks. The question is a useful starting point, but a tag alone cannot identify what a particular player did at runtime.

First, identify which event is actually missing#

“Viewability tracking” can refer to publisher-determined VAST callbacks, a verification vendor using Open Measurement, or a platform's own measurement. Those mechanisms should not be treated as the same integration.

In VAST 4.3, section 3.5, ViewableImpression contains publisher-determined tracking URLs for Viewable, NotViewable, and ViewUndetermined outcomes. Player support is optional. Viewable corresponds to meeting viewability criteria; NotViewable means an executed ad never met them; ViewUndetermined means the player could not determine the outcome. The element belongs under InLine or Wrapper, rather than under the linear creative's TrackingEvents.

This is different from VAST verification resources used by an Open Measurement integration. IAB's CTV programmatic guidance describes the separate roles of VAST AdVerifications, OMID for measurement, and SIMID for interactive use cases. The presence of one mechanism does not prove the other will execute. If the OMID vendor script runs but its report stays empty, trace the verification session through provider reporting rather than substituting a publisher viewability callback.

Write the expected result before testing:

Expected observationWho is expected to produce it?Where will you verify it?
Ordinary impression callbackPlayer or supported delivery integrationOutgoing request and receiver receipt
Publisher-determined viewability callbackPlayer implementation supporting that VAST featureEvent decision and outgoing outcome URL
Verification measurementSupported verification integration and vendor resourceInitialization, measurement signals, and vendor receipt
Final reported viewable impressionReceiving measurement systemAccepted event in the correctly scoped report

If your team cannot identify the event owner, resolve that uncertainty first. Two teams waiting for each other to fire a callback can create a silent measurement gap.

Related guide

Separate impression and viewability events.

Use the checklist to record player support, visibility conditions and observed requests.

Use the tracking QA checklist

Opens a practical guide. No email required.

Use a small decision tree before editing the tag#

Work in an authorized test placement with the affected player and integration. A generic desktop preview can establish that a creative is playable, but it cannot establish that a different mobile app or CTV player supports its measurement.

  1. Was the expected viewability element present in the final response? If not, investigate tag generation and wrapper handling.
  2. Does this player/version support the intended mechanism? If not, fix the integration plan rather than rewriting a tracker URL.
  3. Did measurement initialize and receive valid runtime signals? If not, investigate player integration, verification resources, and environment restrictions.
  4. Was the relevant outcome reached? If not, distinguish measured non-viewable, undetermined, and unsupported cases.
  5. Was a request attempted? If yes, trace transport and receiver processing.
  6. Was it accepted but absent from reporting? Check scope, identifiers, processing delay, and filtration.

Keep each conclusion tied to evidence. “Nothing appeared in my dashboard” does not establish that nothing was sent.

Step 1: capture the final VAST and every wrapper#

Save the actual responses from the failing test session. Record request time, player version, device or browser, creative identifier, and any tag parameters relevant to the case. Protect signed playback URLs and other sensitive tokens when sharing logs.

Inspect the response after wrappers have been resolved, not just the first URL returned to the buyer. Compare the original viewability URLs with the player's parsed tracking model if the player exposes it. The aim is to find whether the configuration reached the component responsible for acting on it.

Check these details:

  • The XML parses successfully and matches the declared supported VAST version.
  • The correct response is returned for this test session, rather than a cached or different creative.
  • Tracking URLs remain intact through wrappers and substitutions.
  • The player retains the expected viewability element in its internal ad model.
  • The tag specifies the intended measurement mechanism rather than only an ordinary impression or completion tracker.

A structural XML fragment

The following fragment shows the relationship between ordinary impression and publisher-determined viewability URLs. It is an original illustrative fragment, not a complete playable VAST response. It omits required ad metadata and creative media. The .invalid hosts are deliberately nonfunctional placeholders.

<InLine>
  <!-- Required metadata and creative elements omitted. -->
  <Impression id="qa-impression"><![CDATA[
    https://tracker.example.invalid/impression?test=qa-01
  ]]></Impression>
  <ViewableImpression id="qa-viewability">
    <Viewable><![CDATA[
      https://tracker.example.invalid/viewable?test=qa-01
    ]]></Viewable>
    <NotViewable><![CDATA[
      https://tracker.example.invalid/not-viewable?test=qa-01
    ]]></NotViewable>
    <ViewUndetermined><![CDATA[
      https://tracker.example.invalid/undetermined?test=qa-01
    ]]></ViewUndetermined>
  </ViewableImpression>
  <!-- Required creative elements omitted. -->
</InLine>

Adding this fragment does not add a measurement engine to a player. Use your ad server's supported tag-generation workflow, validate the complete XML against the applicable specification, and confirm the runtime behavior separately.

On CTV, also test VAST Icons and the disclosure interface separately from verification resources. Preserve the actual response and released player result; icon display and interaction evidence do not establish ad impressions or viewability.

Do not put arbitrary viewability strings into a Tracking event attribute and assume the player will interpret them. A valid callback URL in the wrong location can remain unused while ordinary tracking continues.

Step 2: obtain feature-level support evidence#

“Supports VAST 4” is not enough information. Ask whether the exact SDK version and player mode support the specific feature you expect, on the actual device environment.

Some vendors document this explicitly. Equativ's SDK compatibility reference, for example, lists ViewableImpression separately from general VAST support. Use the corresponding documentation for your own player rather than transferring support claims from another product.

Request a precise answer:

Player and ad SDK:
Version and device/browser:
Client-side or SSAI/hybrid mode:
Publisher-determined VAST ViewableImpression supported?
OM SDK verification supported and enabled?
Supported VAST versions and verification-resource location:
Any format, device, access-mode, or wrapper limitations:
Reference documentation and a working test case:

If the feature is unsupported, make that a visible limitation in reporting and planning. A silent zero in a viewable-impression column can be misread as measured non-viewability. Your reporting should distinguish no measurement from a measured outcome.

For an integration upgrade, ask for a staged test covering the player version, affected inventory, and receiving measurement provider. Avoid deploying a library change solely on the strength of a successful VAST schema validation.

Step 3: verify the measurement lifecycle#

A player can play media while its measurement integration is absent or misconfigured. Check the runtime lifecycle rather than inferring it from playback.

For a concrete example, Google's IMA Android Open Measurement guide explains that supported integrations parse verification resources and send measurement data through OMID. It also documents the verification-resource placement for relevant VAST versions and obstruction handling. Follow the current guide for the SDK you use; copying a setup from another environment can produce a playable ad without the intended verification.

Collect answers to these questions:

  • Did the verification resource load, and was it the expected vendor resource?
  • Was a measurement session created for the correct ad view?
  • Did the session start at the intended point in ad execution?
  • Did playback and visibility state changes reach the measurement integration?
  • Were the view dimensions and lifecycle events associated with the correct creative?
  • Did the session end before it could report the expected outcome?

In a web player, record the effective verification access mode. Google's IMA HTML5 documentation describes different access levels and notes that providers may not support every mode. In its documented LIMITED mode, verification scripts cannot directly confirm the publisher domain. This is a support consideration, not a reason to grant unrestricted access automatically.

Keep OMID verification and publisher-determined VAST callbacks separate in the incident record. A successful OMID session does not, by itself, prove that the player also implements ViewableImpression outcome URLs.

Step 4: test visible, non-visible, and undetermined cases#

Establish expected behavior with the player vendor before running tests. Do not assume every outcome is emitted immediately, especially where the player must observe the rest of an ad before making a determination.

Use a reproducible test matrix:

Authorized QA scenarioWhat to recordWhat it helps separate
Ad in a supported, visibly in-view player for sufficient timeVisibility observations, playback time, callback decisionBaseline measurement and event handling
Ad leaves view before meeting the selected thresholdVisibility timeline and final outcomeProper non-viewability versus missing measurement
Unsupported or restricted measurement contextDocumented limitation and resulting stateUnknown outcome versus measured non-viewable
Overlay appears over the playerObstruction details and measurement signalsIntegration behavior around visibility changes
Ad interrupted or session endsLifecycle timing and available outcomeEarly exit versus transport loss
Same creative on a second supported player versionConfiguration and exact differencesA version-specific regression

A completion event should not be substituted for a visibility measurement. A video can progress while its visibility conditions differ. For minimum definitions and rate calculations, use the viewability report reconciliation guide.

Likewise, do not code “fire the viewable tracker two seconds after start” as a generic fix. Elapsed playback time alone does not establish the required visible area and continuous duration. Such a change can create a convincing event stream that misrepresents the measurement.

Step 5: trace request attempts to receiver receipts#

Once a supported implementation reports the expected outcome, inspect the actual outgoing request. A configuration screenshot proves the URL exists in setup; it does not prove it was requested.

Use the platform's supported network diagnostics and vendor logs. On web, preserve the relevant network trace and console errors for the test. On native apps and CTV devices, use the approved debugging tools available to that integration. Respect device and app security controls when collecting traffic.

EvidenceWhat it establishesNext check
Outcome exists in player diagnostics, no request attemptThe classification and emission stages differCallback registration and event handler
Request attempted, connection failsTransport did not complete normallyDNS, TLS, network policy, timeout, or endpoint health
Request redirectedThe original endpoint handed the request onwardFinal destination, preserved query, and receiver receipt
HTTP success, no accepted eventA response was returnedValidation, identity, mapping, and filtration
Accepted event exists, report omits itThe receiving path processed a recordScope, report definition, event time, and data maturity

Inspect final macro values, query encoding, scheme, and host. An unsubstituted macro or malformed identifier can produce a request that reaches an endpoint but cannot be assigned to the expected campaign.

For a field that remains unresolved or arrives malformed, inspect the actual VAST macro substitution and encoding boundaries. Separate the saved template, emitted URI and receiver interpretation before changing the tracking tag.

Check browser errors for mixed content, blocked resources, and policy restrictions where applicable. Do not assume every cross-origin beacon failure is a CORS issue; the relevant browser requirements depend on the request mechanism and resource type. Work from the actual error and the vendor's supported implementation.

Ask the receiver for a test reference, ingestion timestamp, and accepted or excluded outcome. HTTP status alone does not establish inclusion in the final viewable-impression metric.

SSAI and CTV require an additional ownership check#

In a stitched stream, the component firing ordinary impression beacons may not be the component capable of observing on-screen visibility. Map both responsibilities.

AWS MediaTailor's server-side tracking timing documentation explains that its beacons depend on ad segment requests and can be affected by prefetching. That is evidence about its beacon trigger, not proof of client visibility. Verify other vendors' behavior independently.

CTV measurement also depends on supported client integrations. IAB Tech Lab's OM SDK compliance program lists CTV integration environments among its supported certification platforms. That does not mean every app on those devices implements the relevant feature.

Record device family, app version, player or ad SDK, delivery mode, measurement provider, and who emits each event. A desktop preview of the same tag cannot close a device-specific gap.

For streaming count disagreements, the CTV impression discrepancy workflow explains how to compare sender events, receiver receipts, and final accepted counts. The SSAI verification guide covers the wider delivery path.

Package a useful escalation#

A vendor can act much faster on a reproducible case than on a campaign total. Send one concise incident record, with sensitive data removed:

Symptom: ordinary impression received; expected viewability outcome absent
Expected mechanism: VAST outcome URL / OMID verification / platform metric
Environment: device, app/browser, player, SDK, exact versions
Delivery mode: client-side / SSAI / hybrid
Test session and timestamp with time zone:

Evidence:
- Original tag, wrapper responses, and final parsed configuration
- Feature-support reference for this environment
- Measurement initialization and visibility/playback observations
- Expected outcome and when it should be emitted
- Outgoing request, network result, and receiver reference
- Accepted/excluded result and report filters

Confirmed boundary where observation stops:
Unresolved questions and requested owner:
Proposed change and post-change acceptance checks:

After a fix, repeat the same scenario and the negative cases. A repair should restore honest classification, not force every outcome into the viewable category. Then compare a completed production interval for the affected inventory to verify that the issue was broader than, and resolved beyond, the single QA session.

If playback never starts, follow the VAST playback error diagnosis first. A wrapper or media-loading failure is a different stage from a missing viewability event.

Frequently asked questions#

Does a VAST 4.x version declaration guarantee ViewableImpression support?

No. Ask for support for the exact feature, player version, and environment. XML compatibility and runtime measurement are separate checks.

Does ViewableImpression require OMID?

Do not treat them as the same mechanism. The VAST element carries publisher-determined outcome URLs; Open Measurement supports verification measurement through an integrated SDK and vendor resources. Your player or vendor may connect these components, but that behavior needs documentation.

Can I trigger the viewable URL manually to test it?

A controlled endpoint test can check transport and receiver handling when authorized. It does not prove that the ad was viewable or that the player generated the event. Label test traffic and keep it out of production performance conclusions.

Why is ViewUndetermined missing too?

First confirm support for that outcome and the vendor's emission behavior. An unsupported feature, missing session, lost configuration, or transport problem can prevent any viewability outcome. Silence alone cannot distinguish them.

Should I use VPAID to fix this?

Do not add a legacy executable wrapper as a default repair. Identify the supported measurement path for the inventory. Current industry architecture separates measurement and interactivity, and support varies by environment and player.

For a mobile rich-media creative, separate MRAID host interactions from the measurement path. Establish which SDK supplies the interface and verify the intended native behavior; a file named mraid.js is not a substitute for a supported measurement integration or a missing VAST outcome callback.

Preserve the distinction between playback and proof#

A good incident outcome identifies the exact missing boundary: configuration, support, measurement, emission, transport, or reporting. That finding tells the responsible team what to change and what evidence will establish success.

Review AdProof's measurement approach and measurement workflows, or discuss the evidence needed for your delivery path. Bring the test trace and support matrix so the conversation can focus on the unresolved boundary.

Sources and research notes#

Researched on 25 October 2026. The XML fragment, diagnostic workflow, test matrix, and escalation template are original illustrative material. The XML is intentionally incomplete and must not be deployed as a standalone VAST response. Product capabilities depend on the exact integration and can change.

Your next step

Separate impression and viewability events.

Use the checklist to record player support, visibility conditions and observed requests.

Use the tracking QA checklist

Opens a practical guide. No email required.

Back to all field notes