ctv-supply-and-ad-pods
Skill1 starsChanged 5 months agoMITView on GitHub
Buy connected TV correctly across DSPs. Use when the user asks about CTV or streaming TV supply, ad pods, pod position (first, mid, last), frame-based vs duration-based pods, competitive separation, SSAI vs CSAI, why CTV impressions have no viewability or click data, VAST wrapper or beacon loss, dropped tracking on SSAI, live sports pods, OTT vs CTV, premium publisher direct deals, or how to structure a CTV line item so pods and measurement work.
In this repository129 other files it gives its agents.
- amazon-ads-agentic-and-mcpSkill
- amazon-dsp-account-structureSkill
- amazon-dsp-api-and-automationSkill
- amazon-dsp-audiencesSkill
- amazon-dsp-bidding-and-optimizationSkill
- amazon-dsp-campaign-setupSkill
- amazon-dsp-creative-and-formatsSkill
- amazon-dsp-inventory-and-supplySkill
- amazon-dsp-measurement-and-reportingSkill
- amazon-marketing-cloudSkill
- approval-and-escalation-governanceSkill
- bid-landscape-and-win-rateSkill
- brand-safety-and-suitabilitySkill
- bulk-change-governance-and-rollbackSkill
- change-management-and-incident-responseSkill
- client-deliverable-templatesSkill
- cm360-data-transfer-and-attributionSkill
- cm360-floodlight-and-conversionsSkill
- cm360-reporting-and-trafficking-apiSkill
- cm360-trafficking-and-ad-tagsSkill
- cm360-verification-and-active-viewSkill
- consent-signal-verification-and-decodeSkill
- creative-review-and-approval-queueSkill
- cross-platform-conversion-reconciliationSkill
- cross-platform-launch-qaSkill
- cross-platform-taxonomy-governanceSkill
- ctv-measurement-currenciesSkill
- curation-and-curated-marketplacesSkill
- curation-and-supply-path-optimizationSkill
- data-quality-and-reconciliationSkill
- data-subject-rights-and-subprocessor-governanceSkill
- direct-response-creative-testingSkill
- discrepancy-and-reconciliationSkill
- dsp-selectionSkill
- dv360-account-setup-and-taxonomySkill
- dv360-advanced-analytics-adhSkill
- dv360-api-and-sdf-automationSkill
- dv360-bid-strategySkill
- dv360-campaign-architectureSkill
- dv360-creative-traffickingSkill
- dv360-custom-biddingSkill
- dv360-deals-and-inventorySkill
- dv360-frequency-and-brand-safetySkill
- dv360-launch-qaSkill
- dv360-measurement-and-attributionSkill
- dv360-pacing-and-optimizationSkill
- dv360-reportingSkill
- dv360-targeting-and-audiencesSkill
- dv360-troubleshootingSkill
- dv360-video-creative-specsSkill
- dv360-youtube-and-videoSkill
- ga4-apis-and-measurement-protocolSkill
- ga4-bigquery-export-and-sqlSkill
- ga4-conversions-and-audiencesSkill
- ga4-events-and-key-eventsSkill
- ga4-explorations-and-reportingSkill
- ga4-property-and-data-stream-setupSkill
- gmp-integration-floodlight-ga4-linkingSkill
- google-ads-account-structureSkill
- google-ads-api-and-bulk-operationsSkill
- The first 60 shown.
The file
We split this skill into the rules it gives an agent
4 rules. About 0 are common, 4 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈4
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 63
Read the pod construction from the deal or supply description.
- Distinctive: line 103
Only pay the position premium when the goal rewards it (completion guarantees, attention studies, premium brand placement).
- Distinctive: line 114
Do not rely on your DSP's own frequency cap to prevent same-pod duplication.
- Distinctive: line 139
If you cannot secure a server-side measurement path, treat viewability, IVT, and click reporting for that supply as unavailable, and price …
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
CTV supply and ad pods
Buy connected TV so the pod structure, the ad-server-side insertion method, and the measurement all line up before spend goes live. CTV is not display with a bigger screen. Ads run inside pods (linear style commercial breaks stitched into a stream), the stream is usually assembled server-side, and that assembly silently strips the tracking beacons display buyers take for granted. Get the pod and the insertion method right and the campaign delivers with clean measurement. Get them wrong and you pay premium CPMs for impressions you cannot verify, viewability you cannot report, and frequency you cannot cap.
This skill covers CTV supply mechanics that apply across DSPs (DV360, The Trade Desk Kokai, Amazon DSP).
It assumes you know CPM, viewability, and reach. For that KPI math see the programmatic-foundations
skill.
When to use this skill
- "How do CTV ad pods work?" / "Frame-based vs duration-based pods?" / "What is pod position?"
- "How do I set competitive separation on CTV?" / "Two auto ads ran back to back on a streaming break."
- "SSAI vs CSAI, which do I need?" / "Why do my CTV impressions have no viewability or click data?"
- "My VAST beacons are not firing on CTV." / "Tracking dropped on server-side ad insertion."
- "How do I buy live sports CTV pods?" / "How do I structure a premium publisher CTV deal?"
- "Why is CTV frequency capping not working across apps?" / "OTT vs CTV, what is the difference?"
Boundaries with sibling skills:
- The YouTube and Google video partners line item family (TrueView, bumpers, CPV, YouTube on CTV):
dv360-youtube-and-video. YouTube CTV is bought as YouTube, not as third-party pod-based CTV, so it lives there. This skill is the non-YouTube streaming supply (Hulu, Disney, NBCU, Netflix, Roku, Tubi, Samsung, LG, and the like) bought through deals and open CTV auctions. - Deal type selection and how a deal ID is plumbed to a line item:
dv360-deals-and-inventory(DV360) orttd-inventory-and-deals(Kokai) oramazon-dsp-inventory-and-supply. This skill is what the deal must contain for CTV to work, not the deal mechanics. - Reducing the number of SSPs between the DSP and the publisher:
curation-and-supply-path-optimization. - Frequency-cap policy and brand-safety exclusions in general:
dv360-frequency-and-brand-safety. - Cross-screen reach and frequency planning math:
reach-and-frequency-planning. - Bid strategy for the line item (fixed for guaranteed, automated for biddable):
dv360-bid-strategy.
Quick reference
Decide these four things in order before you launch a CTV buy.
| Decision | Options | Pick this when |
| Pod construction | Frame-based (fixed slots) vs duration-based (fill a time budget) | Frame gives predictable slot count and clean competitive separation; duration maximizes fill on long breaks. The publisher usually dictates this. |
| Pod position | First, mid, last, or unrestricted | First and last cost more and complete more. Buy position only when completion or attention justifies the premium. |
| Insertion method | SSAI (server-side) vs CSAI (client-side) | Almost all premium CTV is SSAI. CSAI only on app environments that run a full VPAID/OM-capable player, which is rare on TV. |
| Measurement plan | OM SDK server-side beacons, publisher-passed macros, or MRC-accredited third party | Choose before launch. SSAI breaks client-side beacons, so you need server-side measurement or you fly blind. |
The single most expensive CTV mistake: buying premium SSAI inventory with a display measurement setup, then discovering post-launch that viewability, IVT, and click beacons never fired. Confirm the measurement path in the deal, not after the first invoice.
Core process
- Confirm the supply is actually CTV, not OTT-on-mobile mislabeled. CTV means the ad renders on a television (smart TV OS, streaming stick, game console, set-top box). OTT is any over-the-top delivery including phones and desktops. Device-type and app-bundle targeting keep a "CTV" line item from leaking onto phone screens, because the two clear at very different CPMs and attention levels.
- Read the pod construction from the deal or supply description. Frame-based pods expose a fixed number of ad slots (for example a 4-slot pod); duration-based pods expose a time budget the exchange fills with any mix of durations (for example a 90-second break filled by 30s + 30s + 15s + 15s). This decides how you set creative duration and how competitive separation is enforced.
- Decide whether pod position matters for this goal. First-position and last-position slots complete at higher rates and command a premium. For a completion-sensitive or attention goal, buy position; for efficient reach, take unrestricted position and save the premium.
- Set competitive separation expectations. Separation (no two competing brands, and no duplicate creative, in the same pod) is enforced by the publisher or SSP ad server, not by your DSP alone. On open CTV auction you cannot fully guarantee it, so for brand-sensitive advertisers require separation in a deal.
- Identify the insertion method and its measurement consequence. If the supply is SSAI (it usually is),
plan for server-side measurement. See
references/ssai-vs-csai-tradeoffs.mdfor exactly which signals survive SSAI and which do not, and how OM SDK server-side and passed macros recover them. - Pin the measurement path in writing before launch. Confirm which viewability and IVT vendor the publisher supports server-side, and confirm the impression and completion beacons will fire through the SSAI stitcher. Do not assume the DSP's standard tracking works.
- For premium and live inventory, buy through a deal, not open auction. Live sports pods, first-position
slots, and competitive separation are contractual guarantees that only a Programmatic Guaranteed or
private deal delivers reliably. See
references/premium-publisher-supply-structure.md. - Cap frequency at the household or device level, and know it degrades on SSAI. SSAI can obscure the device identifier, so cross-app household frequency capping is weaker than on client-rendered supply. Where cross-publisher frequency matters, consolidate through fewer supply paths and a shared identity.
Decision rules and thresholds
Frame-based vs duration-based pods
- Frame-based (also called slot-based): the pod has a fixed number of positions. You know a 4-slot pod holds exactly four ads. Competitive separation and position targeting are cleaner because slots are discrete. Cut creative to the standard slot durations the publisher accepts (commonly 15s and 30s).
- Duration-based: the pod is a time budget the exchange fills to maximize monetization. A 120-second break might be four 30s ads or eight 15s ads. Fill is higher, but slot count and your neighbors are less predictable, so competitive separation is harder to guarantee.
- Rule: for brand-sensitive buys where who-you-run-next-to matters, prefer frame-based pods in a deal. For scaled reach where fill and price matter more, duration-based open supply is fine.
Pod position
- First and last positions complete at higher rates and draw more attention, so publishers price them up. Mid-pod positions are cheaper and complete lower.
- Only pay the position premium when the goal rewards it (completion guarantees, attention studies, premium brand placement). For cost-efficient reach, take unrestricted position.
- Position targeting is a deal feature. Open CTV auction rarely lets you dictate position, so if position is non-negotiable, it belongs in a Programmatic Guaranteed or private deal.
Competitive separation
- Separation prevents two competitors (or two copies of your own creative) inside the same pod. On CTV it is enforced by the publisher/SSP stitching the pod, because the DSP does not control the full break.
- On open auction you get best-effort separation at most. For regulated or brand-sensitive categories, require category-level separation contractually in the deal, and confirm the publisher can enforce it.
- Do not rely on your DSP's own frequency cap to prevent same-pod duplication; the pod is assembled before your DSP sees the full slate.
SSAI vs CSAI
- SSAI (server-side ad insertion): the publisher's server stitches the ad into the video stream before it reaches the device, so the ad arrives as part of the content. This is how nearly all premium CTV (live sports, major streamers, TV OS inventory) is delivered, because it defeats ad blocking and gives a seamless, buffer-free viewing experience. The cost: the ad is not running in an interactive client player, so client-side VAST tracking, VPAID, and many verification beacons do not fire on their own.
- CSAI (client-side ad insertion): the device's video player requests and renders the ad, so it can run a full VAST/OM-capable player and fire client-side beacons. This exists mostly in app environments with a capable SDK player. On living-room TV OS supply it is the minority.
- Rule: assume SSAI on premium CTV and plan server-side measurement from the start. Only assume CSAI when
the publisher confirms a client player that supports OM SDK and VPAID. The signal-by-signal survival
table is in
references/ssai-vs-csai-tradeoffs.md.
Measurement under SSAI
- Client-side impression, quartile, and completion beacons in a VAST wrapper may not fire when the stream is stitched server-side. Recover them with OM SDK running server-side (OMID), publisher-passed VAST macros filled at stitch time, or an MRC-accredited vendor the publisher integrates on the server.
- OM SDK 1.5 adds Samsung and LG smart-TV support alongside Android TV and tvOS, so cross-TV-OS server-side measurement is more attainable than it was, but it still requires the publisher to have integrated it. Confirm coverage per publisher; do not assume.
- If you cannot secure a server-side measurement path, treat viewability, IVT, and click reporting for that supply as unavailable, and price and evaluate the buy accordingly (completion rate and delivery become your primary in-platform signals).
CTV vs OTT device leakage
- A line item labeled CTV can serve on phones and desktops if device targeting is loose, because OTT includes those screens. Restrict to connected-TV device types and, where possible, an app-bundle allowlist, so you buy the living-room screen you are paying a CTV CPM for.
Reference material
references/pod-construction-and-position.md: frame-based vs duration-based pod anatomy, pod-position economics, competitive-separation enforcement, VAST ad-pod signaling, and the DV360 / Kokai / Amazon DSP controls that map to each. Read this when structuring a pod buy or diagnosing back-to-back competitor ads.references/ssai-vs-csai-tradeoffs.md: a signal-by-signal table of what survives SSAI and what breaks, the OM SDK server-side and passed-macro recovery paths, and a pre-launch measurement checklist. Read this when CTV impressions have no viewability, no clicks, or missing quartile beacons.references/premium-publisher-supply-structure.md: how premium and live CTV supply is structured (direct PG, curated CTV marketplaces, live sports packages), the DSP-by-DSP landscape as of 2026, and how to buy position and separation contractually. Read this when building a premium or live-sports CTV buy.
Templates and examples
- Live sports, brand advertiser, completion-sensitive: Programmatic Guaranteed deal with NBCUniversal live sports inventory, frame-based 4-slot pods, first-position slot, category competitive separation required, server-side OM SDK measurement confirmed with the publisher, 30s creative, fixed bid at the deal rate, device restricted to connected TV. Buy through DV360 curated live sports packages or a Kokai live-events deal. Do not run this on open auction; position and separation would not be guaranteed.
- Scaled CTV reach, performance advertiser, efficiency goal: open CTV auction plus a few private deals on Tubi and Roku, duration-based pods, unrestricted position, 15s and 30s creative, automated bidding toward a reach or viewable-CPM goal, household frequency cap set knowing SSAI weakens cross-app enforcement, connected-TV device restriction with an app-bundle allowlist to block mobile OTT leakage.
- Diagnosing "no viewability on my CTV line item": confirm the supply is SSAI (it almost certainly is), then confirm whether the publisher passes server-side OM SDK or macro-based beacons. If not, viewability is genuinely unavailable for that supply, not a DSP bug. Switch to publishers with server-side measurement or accept completion rate as the proxy.
Common pitfalls
- Buying SSAI supply with a display measurement plan. The impression fires but the viewability, IVT, and click beacons do not, and you discover it after launch. Confirm the server-side measurement path in the deal first.
- Assuming your DSP frequency cap prevents same-pod competitor collisions. The pod is stitched by the publisher before the DSP sees the full slate; competitive separation must be a deal requirement.
- Treating YouTube-on-CTV as if it were third-party pod CTV. YouTube CTV is bought as YouTube (see
dv360-youtube-and-video); its pods, formats, and measurement work differently from streamer supply. - Letting a CTV line item leak onto mobile OTT. Loose device targeting serves phone impressions at a CTV CPM. Restrict to connected-TV device types.
- Expecting reliable cross-app household frequency on SSAI. The device identifier can be obscured at the stitch, so cross-publisher frequency capping is weaker; consolidate supply paths and use a shared identity where cross-publisher frequency is the goal.
- Paying for first-position on open auction. Open CTV auction rarely honors position; if you need it, put it in a deal or you are paying a premium you cannot enforce.
- Cutting creative to durations the pod does not accept. A duration-based pod that fills 15s and 30s slots will not place a 20s or 45s spot; match the publisher's accepted slot durations.
Sources
- New biddable capabilities for live sports with Display & Video 360 (as of July 2026)
- Google NewFront: Rethink CTV beyond streaming with Display & Video 360 (as of July 2026)
- Open Measurement SDK, IAB Tech Lab (as of July 2026)
- Display & Video 360 upcoming changes and announcements (as of July 2026)
- Display & Video 360 API release notes (v4) (as of July 2026)
- Display & Video 360 API v4 REST reference (as of July 2026)
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

