Quick snapshot, so we start from the same page

• How much annual programmatic spend does your team manage today? Options: Under $1M, $1M–$5M, $5M–$20M, $20M–$50M, Over $50M
• Tick the channels that make up the majority of that spend Options: Display, Video (in-stream), Connected TV, Audio / Streaming, Digital out of home, Native
• Who owns vendor selection and contracting authority for your programmatic stack? Options: Head of Programmatic/VPMedia, Procurement, Agency trading desk, Central marketing ops, Shared committee
• Walk me through your current supply path in a typical campaign, including the DSPs, exchanges, and any known reseller chains
• Which single metric would you require the pilot to prove to consider switching platforms? Options: CPA / CAC, Viewable CPM, Fraud-filtered impressions, Reach / unique households, Cost per Completed View, Other
• How quickly do you need directional results from the pilot to make a decision, 30 days, 60 days, 90 days, or longer? Options: 30 days, 60 days, 90 days, Longer than 90 days, Unsure

Where value is leaking from your stack

• Which single part of your current programmatic setup is most responsible for hidden fees or lost auctions? Options: Reseller chains/resellers, Multiple intermediary exchanges, Latency in bidding path, Fee-adding SSPs, Data resale fees, Unknown
• How many distinct intermediaries typically touch a dollar between your bid and the publisher in a given campaign? Options: None or direct, 1, 2, 3–4, 5 or more, Unsure
• When was the last time you obtained a full, queryable log of every impression, intermediary fee, and bid decision for a live campaign? Options: Within last 30 days, Within last 6 months, Within last 12 months, Longer than 12 months, Never
• How do hidden fees or latency appear in your performance reporting today, for example higher CPMs, frequent lost auctions, or skewed viewability?
• If an analysis exposed 15 percent in reseller fees on a core channel, what immediate action would you expect your team to take? Options: Pause spend on that path, Push for supply-path changes, Demand contractual rebates, Run a parallel test with transparent supply, Other

Brand safety and trust, straight to the point

• If your CMO saw one viral screenshot tomorrow of your creative next to objectionable content, what level of brand safety failure would be a deal breaker? Options: Any viral screenshot leads to pause, Multiple small incidents are tolerated, Only sustained, high-volume incidents are deal breakers, Depends on remediation speed
• Choose the brand safety controls you require in every buy Options: Pre-bid category controls, URL-level verification, Custom blocklists, Human review for flagged placements, Viewability thresholds, Third-party brand safety certification
• How many brand safety incidents has your team escalated in the last 12 months? Options: 0, 1–2, 3–5, 6–10, More than 10
• Describe the most recent incident, including how it was discovered and what remediation steps were taken
• What exact percentage of impressions filtered by fraud or brand-safety controls would trigger pausing a campaign? Options: 0.0% (any issue), 0.1%–0.5%, 0.5%–1%, 1%–3%, Over 3%

Targeting and measurement after cookies, the practical test

• If third-party cookies stopped tomorrow, which audience signal do you believe would cause the largest increase in CPA for your core campaigns? Options: Third-party cookie IDs, Pixel-based retargeting, Household graphs, Behavioral third-party segments, None of the above
• How much of your buyer audiences are backed by first-party data segments today, roughly as a percentage? Options: 0%–10%, 11%–25%, 26%–50%, 51%–75%, Over 75%
• Choose contextual targeting approaches you currently use or want to test Options: Content category matching, Semantic / topic modeling, Keyword-level targeting, Publisher-level contextual bundles, Sentiment or tone filters
• Who owns identity resolution in your org and do they have contractual permission to share or sync audiences for buying? Options: In-house identity team, permissions in place, In-house identity team, permissions pending, Third-party identity vendor, permissions in place, No clear owner or permissions
• Do you have data sharing and privacy approvals in place to allow audience syncing and log-level measurement across a new platform? Options: Yes, fully approved, Partial approvals in place, No approvals yet, Need to confirm with legal/privacy
• If we can only match your key target audiences in CTV at a certain rate, what minimum match rate would make you stop the pilot? Options: Less than 10%, 10%–25%, 26%–50%, 51%–75%, Over 75%

The other options you are actively weighing

• Name the alternative you are most likely to pick if our pilot does not show immediate cost or transparency gains Options: Stay with incumbent DSP, Shift more budget to walled gardens, Run a full internal build, Work with another independent DSP, Pause reallocations altogether
• List the alternatives you have evaluated or are still considering, including internal build and walled garden options
• What would have to be true about your current approach for you to stay with it instead of changing platforms? Options: Proven lower total fees, Full bid-level transparency, Equal or better CPA, Significant remediation of brand safety, Faster integrations
• Has anyone on your team proposed solving this problem without an outside vendor or partner? Options: Yes, internal build proposed, Yes, modify existing stack, No, external partner preferred, Discussion ongoing
• If the incumbent agreed to meet your transparency and fraud thresholds within contract, would you remain with them? Options: Yes, No, Maybe, depends on commercial terms, Need to evaluate case by case

What must be true operationally for the pilot to run

• Point to the single missing internal capability that would stop this project before it starts Options: Secure log endpoint or S3, API integration capacity, Data sharing approvals, Measurement partner access, Procurement or contract block
• Do you have engineering resources available to complete API integrations and log delivery within the pilot timeline? Options: Yes, dedicated engineers, Partial availability, need scheduling, No, would require additional hire or contractor, Unsure
• List the external systems that must be connected for targeting and measurement, for example ad servers, data warehouses, or measurement vendors
• How would you rate the maturity of your first-party data in terms of identity keys, schema cleanliness, and readiness to sync? Options: High maturity and ready, Moderate, needs mapping, Low, needs cleanup, No usable first-party data
• Identify the team that will own log ingestion and confirm whether a secure endpoint is available for daily or streaming export Options: Ad ops, Engineering, Data platform, Vendor to provide, Not yet identified
• Are there procurement, legal, or privacy approvals that could delay the start beyond 4 weeks? Options: Yes, likely delay, Possibly, depends on contract value, No, approvals already in place, Unknown
• Is there any absolute blocker that would prevent the pilot from starting within your desired timeline? Options: Yes, list blocker in next field, No blockers, Depends on negotiation

Designing the pilot and the decision rules

• If the pilot proves the CPA or viewability improvement you need, what would stop you from moving full budget next quarter? Options: Procurement or contracting timeline, Executive approval delays, Lack of trust in data, No operational capacity to scale, Nothing, we would move quickly
• Indicate the channels you want included in the pilot Options: Display, Video, CTV, Audio, DOOH, Mixed channel test
• Describe the targeting approaches you want to test, for example contextual-only, first-party audience syncs, or combined strategies
• Select the fraud filters and verification signals you require for the pilot Options: Pre-bid fraud filters, Post-bid filtering by log, Third-party verification, Viewability thresholds, IP / bot filters, PSA/IVT blocks
• Provide the acceptance criteria you will use to evaluate the pilot at 30 days, for example X% CPA improvement or Y% reduction in fraud-filtered impressions
• How would you like log-level placement data delivered, daily files, a streaming endpoint, or a queryable API? Options: Daily flat files (S3), Streaming endpoint (Kafka/HTTP), Queryable API, Secure FTP, Other
• If the pilot meets acceptance criteria, is the budget reallocation authority pre-approved and who would sign off? Options: Yes, pre-approved and buyer signs off, Yes, requires final PO from procurement, No, requires executive approval, Unsure
• What is your target start date for the pilot? Options: Within 2 weeks, Within 1 month, Within 2 months, 3 months or later, TBD

• Solution Experience Session

• Confirm the current state and its cost
• You confirm the measurement plan captures the supply-path, fraud, viewability, and CPA signals required to evaluate the pilot.
• Provide the historical CPA, viewability, and fraud-filtered impression counts for the comparable campaign on your current DSP to establish baselines.
• You confirm that the log outputs and queries shown will let you verify intermediary fees and impression sources for every impression.
• Map buyer scenarios to measurable signals
• Provide the single pilot scenario and channel mix you want to run as the $50k, 30-day test so we can finalize logging and acceptance criteria.
• You agree that the CTV deduplication approach will produce deduped household reach and reduce frequency waste as described.
• Share the schema fields required for your audit so the platform can provision the impression-level log with the exact columns you need.
• Prove supply-path transparency
• You agree on the specific pilot acceptance criteria and the next steps to operationalize the test campaign.
• Prove contextual targeting and measurement
• Run a sample forecast and deliver a CPA and reach projection for the proposed $50k pilot using the agreed targeting and supply-path constraints.
• Schedule the technical handoff meeting to exchange log endpoints, access credentials, and the data ingestion timeline.
• Prove CTV deduplication and reach validation
• Validate the shared plan
• Agree pilot scope and next operational steps

• Solution Experience Session
• Solution Experience Deck
• Solution Brief

• meeting
• slides
• document

• Provision DSP account and seat access
• Integrate header-bidding with CTV app inventory
• Configure private marketplace (PMP) deal seats
• Onboard buyer first-party data (server-to-server sync)
• Activate contextual targeting engine
• Enable real-time household deduplication across linear and streaming
• Deploy viewability and verification tags/SDKs
• Enable real-time fraud detection and filtering
• Ingest creative assets and serve ads
• Execute 30-day $50,000 test campaign
• Provide queryable log-level impression export
• Configure frequency capping and pacing controls
• Supply-path transparency reporting and fee breakdown export

Provision DSP account and seat access

• Do you require a new DSP account or will you use an existing DSP account ID for the pilot? Options: New account, Use existing account
• Which seat roles should be provisioned (for example admin, campaign manager, reporting only)? Options: Admin, Campaign manager, Reporting only, Creative upload
• Provide the DSP account ID, primary admin email, and billing code needed for provisioning.
• How many user seats and API clients should be created initially for campaign operations and automated reporting? Options: 1-5, 6-20, 21+
• Are there compliance or billing constraints tied to seats such as a purchase order number, cost center, or approved billing entity? Options: Yes, No
• Indicate which authentication method you will provide for seat API access (API key, OAuth2, SAML) and any existing credential rotation policy. Options: API key, OAuth2, SAML

Integrate header-bidding with CTV app inventory

• List the CTV app inventory endpoints you want integrated using app bundle IDs or publisher app IDs.
• Do you have PMP deal IDs or direct header-bid placements to map to each app endpoint for the pilot? Options: Yes — PMP/Deal IDs available, No — need to create deals
• Specify the OpenRTB and VAST versions the header-bid adapter must support for your app endpoints. Options: OpenRTB 2.5, OpenRTB 3.x, VAST 4.x
• Who on your team owns the app developer relationship and will approve SDK or header-bid code changes?
• Estimate the expected daily impression volume per app endpoint for the initial pilot run. Options: <10k, 10k-100k, >100k
• Are there app-level measurement constraints such as MRC viewability or restricted SKUs that must be enforced on integrated placements? Options: Yes — MRC required, No

Configure private marketplace (PMP) deal seats

• List the PMP deal IDs and the buyer seat IDs to be configured for the pilot along with the publisher-provided deal config file if available.
• What floor price or deal CPM ranges should be enforced per PMP deal during the pilot? Options: Fixed CPM, Floor price range, Open to negotiation
• Indicate allowed creative types per deal such as VAST video, HTML5 banner, or audio tag and any creative duration limits. Options: VAST video, HTML5 banner, Audio tag, Native
• Confirm whether deal-level reporting dimensions such as supply path string and intermediary fee breakdown are required for reconciliation. Options: Yes, No
• Specify the acceptance criteria that will confirm PMP deal seats are correctly configured (for example test bid wins, correct seat ID in bid responses).
• List any publisher-imposed creative or ad-serving restrictions to record against each deal (for example max duration, companion creatives, muted start).

Onboard buyer first-party data (server-to-server sync)

• Identify the first-party segments from your source CRM or CDP to be synced via server-to-server including segment names and expected record counts.
• What identifier mapping will you use for S2S sync such as hashed email, encrypted user ID, or household ID, and how will hashing be performed? Options: Hashed email, Encrypted user ID, Household ID, Other
• Provide the S2S endpoint URL, authentication method, and sample payload schema we should validate before the initial push.
• Which privacy regime governs these data transfers (for example GDPR, CCPA) and which consent artifacts or DPIAs will you supply? Options: GDPR, CCPA, Both, Other
• Indicate target segment size and minimum match rate required for the pilot segment to be considered usable. Options: <10k, 10k-100k, >100k
• Choose sync cadence for the pilot: real-time lookups, hourly batch, or daily batch. Options: Real-time lookups, Hourly batch, Daily batch

Activate contextual targeting engine

• Which contextual taxonomy will drive targeting for the pilot such as the IAB content taxonomy or a custom category list? Options: IAB taxonomy, Custom taxonomy, Keyword-based only
• What matching granularity do you require for contextual signals: page-level, section-level, or creative-level classification? Options: Page-level, Section-level, Creative-level
• Describe the brand-safety categories to include and exclude using taxonomy codes or explicit keyword lists.
• Which classifiers should be enabled: language detection, sentiment scoring, and brand-safety labeling? Options: Language detection, Sentiment scoring, Brand-safety labeling, All
• When measuring contextual targeting lift against your cookie-based baseline, which uplift window and signals should be used?
• Are there creative landing page constraints tied to contextual placements such as restricted content or age gating? Options: Yes, No

Enable real-time household deduplication across linear and streaming

• Which household identifier approach will you use for deduplication: household ID hashing, MAID-to-household mapping, or CTV device identifier? Options: Household ID hashing, MAID-to-household mapping, CTV device ID
• Provide the list of linear broadcast source IDs and streaming source identifiers we must map for deduplication.
• Specify the maximum deduplication latency acceptable in real-time bidding, for example under 100 milliseconds or under 500 milliseconds. Options: <100ms, <250ms, <500ms
• How should frequency counts be reconciled when an individual household is reached via both linear and streaming within a 24-hour window?
• Are there opted-out households or suppression lists to exclude from deduplication and delivery? Options: Yes, No
• Indicate the acceptance metric that will verify deduplication accuracy such as percent reduction in duplicate reach or absolute unique household count. Options: % duplicate reach reduction, Absolute unique households, Other

Deploy viewability and verification tags/SDKs

• Which measurement standard must the SDKs meet such as MRC viewability compliance or a custom measurement spec? Options: MRC viewability, Custom standard
• Provide the exact app or site SDK package names and versions where verification code must be installed.
• Attach the tag or SDK installation endpoints and list any API keys or credentials required for deployment testing.
• What viewability threshold do you require to count an impression for reporting, for example 50% in-view for 1 continuous second? Options: MRC standard 50% 1s, MRC 50% 2s, Custom threshold
• Do you require cross-format verification where the same impression must include viewability metrics across CTV, video, and display? Options: Yes, No
• Who will validate SDK installation on test devices and provide evidence such as screenshots or a sample verification log?

Enable real-time fraud detection and filtering

• Which fraud detection signals and thresholds should be applied at bid-time such as bot-score threshold, device spoofing flags, or IP blocklists? Options: Bot-score threshold, IP blocklist, Device spoofing detection, All
• Provide the list of supply sources, domains, or IP ranges you require excluded from bidding for the pilot.
• State which post-impression signals must be retained in logs for fraud analysis such as user agent, device fingerprint, and RTB debug fields. Options: User agent, Device fingerprint, RTB debug fields
• Specify the maximum allowable invalid traffic (IVT) rate for the pilot above which the campaign is considered failing. Options: <1%, <2%, <5%
• Do you require real-time alerts or webhook notifications when suspected fraudulent activity spikes during the 30-day pilot? Options: Yes, No
• Identify the approver on your team who will sign off on the fraud filter rule set and exceptions during the pilot.

Ingest creative assets and serve ads

• Which creative formats and VAST versions will you supply for the pilot such as VAST video, HTML5 banner, or audio tag? Options: VAST video, HTML5 banner, Audio tag, Native
• Provide the creative spec sheet including accepted sizes, max duration, max file weight, and required tracking pixels.
• Choose how creatives will be delivered for serving: direct upload to the DSP seat or via creative CDN tags (VAST tag URLs). Options: Direct upload, Creative CDN tags (VAST URLs)
• Specify creative QA acceptance criteria that must be met prior to live serving such as rendering correctly on target devices and firing click and impression trackers.
• Do you require dynamic creative optimization templates or creative sequencing for the pilot? Options: Yes, No
• Name the party responsible for hosting creatives, rotating variants, and providing fallback assets during the pilot.

Execute 30-day $50,000 test campaign

• Confirm the pilot budget and timing: $50,000 over 30 days starting on which calendar date?
• Select the channels to include in the $50,000 pilot such as display, video, CTV, and audio. Options: Display, Video, CTV, Audio
• Describe the target audiences and bid strategy you want used for the pilot such as target CPA, target ROAS, or reach-focused bidding. Options: Target CPA, Target ROAS, Reach-focused
• Which KPIs will determine success for the pilot (for example CPA, viewability rate, and fraud-filtered impressions)? Options: CPA, Viewability rate, Fraud-filtered impressions, Other
• What acceptance criteria will confirm the 30-day $50,000 pilot is successful; please provide numeric thresholds for CPA, minimum viewability percentage, and maximum IVT.
• Identify the campaign owner who will handle real-time optimizations, pacing adjustments, and hourly checks during the pilot.

• desired_state
• current_state
• stakeholders
• gaps
• success_criteria
• decision_readiness
• desired_state
• decision_readiness
• current_state
• success_criteria
• gaps
• stakeholders
• desired_state
• success_criteria
• stakeholders
• gaps
• current_state
• decision_readiness
• decision_readiness
• decision_readiness
• decision_readiness
• decision_readiness

• Subscription Order Form
• Master Services Agreement (MSA)
• Transparency & Data-Access Service Level Agreement (SLA)
• Pilot Acceptance and Scaling Rules
• Data Processing Agreement (DPA)
• Audit & Third-Party Verification Addendum
• Industry Compliance Addendum (conditional)

• Go-live Health Check
• First Measurement Review
• Acceptance Gate Decision
• Ongoing Performance and Issue Triage

• Update and circulate the operational runbook covering reporting cadence, log endpoints, and escalation paths before the next quarterly review.
• Provide full log-level placement exports covering the measurement window for independent verification.
• Implement agreed fraud-filter or targeting adjustments and record the change history for audit.
• Publish a corrective action tracker with deadlines and verification criteria in advance of the acceptance gate.
• Restate acceptance criteria and evidence requirements
• Produce a documented acceptance decision for each criterion recorded in the Pilot Campaign Evaluation stage and capture the named buyer signatory where required for managed engagements.
• Agree remediation steps with deadlines for any failed criteria and commit to verification evidence.
• Confirm the incumbent system is either decommissioned or formally retained read-only with data archived, and that fallback usage is closed.
• Publish the formal acceptance record that references the Pilot Campaign Evaluation criteria and include the named buyer signatory or documented buyer decision.
• Execute remediation work for failed criteria and provide verification evidence by the agreed dates.
• Provide proof of incumbent decommissioning or archived-read-only status and evidence that no active campaigns remain in the legacy system.
• Performance trends since acceptance
• Confirm fraud-filtered impressions and viewability rate remain aligned to the targets recorded in the Pilot Campaign Evaluation stage or document corrective steps if not.
• Resolve or reprioritize open issues and bugs with agreed resolution timelines.
• Agree the priority order for enhancement requests that materially affect measurement or delivery and confirm the next check-in cadence.
• Publish the quarterly performance packet with aggregated metrics and placement-level logs for independent review.
• Deliver a prioritized issue backlog with severity and target resolution dates for all open tickets.
• Reconfirm scope and success criteria location
• Confirm the deployment and integrations are producing complete log exports and reporting feeds for the test campaign.
• Establish a short list of high-priority defects with agreed remediation tasks and verification checkpoints.
• Confirm the location of numeric acceptance criteria in the Pilot Campaign Evaluation stage and the owners responsible for each criterion.
• Deliver a deployment health checklist and sample log export covering the initial days of the campaign.
• Produce reproduction steps and temporary mitigations for each high-priority blocker.
• Circulate user access and training gaps for quick remediation before the first measurement review.
• Present first 30-day campaign results
• Determine whether cost per acquisition and fraud-filtered impressions are trending toward the targets recorded in the Pilot Campaign Evaluation stage.
• Document root causes for any shortfalls and agree a corrective plan with deadlines leading to the acceptance gate.
• Confirm the acceptance gate date and list of required evidence and log exports for that meeting.
• Present final outcome data with log-level proof
• Investigate deviations and operational root causes
• Deployment and integration validation
• Data quality and audit
• Document pass or fail per criterion and capture signatory
• User access and onboarding status
• Open issues and bug triage
• Root-cause diagnosis for metric gaps
• Remediation plan for failed criteria
• Early adoption signals and usage checks
• Agree corrective actions and timelines
• Enhancement request backlog and prioritization
• Confirm acceptance gate schedule and deliverables
• Operational housekeeping
• Blockers and open defect triage
• Incumbent wind-down checkpoint
• Agree immediate remediation actions