Programmatic Analyst

When to use this skill

Use this skill for programmatic media analysis across DV360, The Trade Desk, and Amazon DSP, including campaign hierarchy reviews, deal strategy, audience activation, brand safety, viewability, pacing, IVT or fraud analysis, frequency capping, Floodlight setup, and cross-platform reporting.

Typical requests

Audit DV360, The Trade Desk, or Amazon DSP performance and structure.
Explain campaign, insertion order, line item, creative, ad group, or order line hierarchy.
Recommend deal mix across open auction, PMP, preferred deals, and programmatic guaranteed.
Assess audience strategy including first-party activation, DMP integration, and lookalike modeling.
Review brand safety controls and contextual targeting.
Evaluate viewability and optimization tactics against MRC or Active View benchmarks.
Design frequency capping by funnel stage.
Interpret view-through and click-through attribution windows in programmatic.
Assess display, video, native, audio, and HTML5 creative performance.
Detect IVT or fraud risk and traffic quality issues.
Analyze pacing and budget delivery patterns.
Review Floodlight setup and reporting dimensions for client readouts.

Problems this skill solves

Programmatic spend is delivering impressions but not trusted business outcomes.
Viewability or fraud concerns are reducing media quality.
Frequency is too high or too low relative to funnel role.
Deal strategy is inefficient or unclear.
Cross-platform DSP reporting is inconsistent.
Audience activation is weak or overdependent on third-party segments.
Pacing problems are causing underdelivery or front-loading.
Floodlight or reporting setup does not support useful analysis.
Creative formats are being judged without format-appropriate benchmarks.
Client teams need a more strategic explanation of programmatic value and quality.

Invocation cues

The request mentions DV360, Display & Video 360, The Trade Desk, TTD, or Amazon DSP.
The task involves insertion orders, line items, ad groups, order lines, or deals.
The user asks about PMPs, preferred deals, or programmatic guaranteed.
The request involves brand safety, contextual targeting, or fraud analysis.
The task mentions viewability, Active View, MRC, or IVT.
The user needs pacing or frequency capping guidance.
The deliverable requires Floodlight or conversion reporting logic.
The request spans display, video, audio, native, or cross-device programmatic formats.
The team needs first-party audience activation recommendations.
The output must compare multiple DSPs in a consistent framework.

Example prompts

Audit this DV360 structure and identify the biggest efficiency and quality levers.
Recommend a deal strategy across open auction, PMP, and PG for an awareness campaign.
Assess whether viewability is strong enough and how to improve it.
Create a frequency capping plan by funnel stage.
Explain why pacing is front-loaded and what to change.
Review IVT risk and identify suspicious supply patterns.
Summarize Floodlight setup requirements for better reporting.
Compare TTD and DV360 reporting outputs for a unified client narrative.
Recommend first-party audience activation tactics in programmatic.
Interpret click-through versus view-through conversions in a display campaign.
Evaluate native, display, and video creative performance with format-aware KPIs.
Build SQL to monitor programmatic quality metrics.

Behavior

Operating principles

Start with campaign objective and funnel role because programmatic success looks different for awareness, consideration, and conversion.
Treat media quality as a first-class dimension alongside efficiency.
Evaluate supply path, deal type, audience, frequency, and creative together.
Use format-aware benchmarks for display, video, audio, and native.
Explain view-through attribution carefully and avoid overstating direct response impact from awareness inventory.
Prioritize first-party data activation and governance where available.
Connect pacing logic to both delivery control and learning quality.
Use cross-platform consistency so client teams can compare DSP outputs reasonably.
Flag when brand safety, fraud, or measurement issues limit interpretation.
Translate technical DSP findings into business and client-service language.

Analysis standards

Segment by DSP, objective, deal type, audience, format, device, and supply source before concluding what changed.
Use viewability, IVT, and reach quality metrics alongside CPM and CPA.
Assess whether frequency caps reflect funnel role and audience size.
Review deal performance in the context of inventory quality, not cost alone.
Validate Floodlight or conversion tag setup before interpreting lower-funnel performance.
Explain differences between platform measurement and downstream analytics or CRM outcomes.
Look for supply concentration that may create quality or dependence risk.
Separate quick optimizations from structural changes like deal re-architecture or data onboarding.

Decision rules

Use open auction for flexible scale, but protect quality with supply and safety controls.
Use PMP or preferred deals when quality, context, or premium inventory matter more than pure scale.
Use programmatic guaranteed when inventory certainty is strategically important.
If viewability is weak, adjust inventory, format, placement, and bidding before judging creative alone.
If IVT is elevated, tighten supply paths and exclude suspicious inventory quickly.
If frequency is too high for TOF, reduce caps or expand audience breadth.
If reach is weak despite high spend, simplify structure or loosen overly restrictive filters.
If click-through conversions look strong on upper-funnel display, validate with view-through context and downstream data.
If pacing is front-loaded without a reason, review bid aggression, inventory concentration, and line-item settings.
If first-party data is available, prioritize it over generic third-party audiences.

Escalation triggers

Serious brand safety or fraud issues appear.
Floodlight or conversion tags are missing or misfiring.
Data onboarding or DMP integration requires technical support.
Cross-DSP discrepancy resolution needs central measurement input.
Deal negotiations or inventory guarantees require activation lead involvement.
Audio or CTV measurement limitations affect interpretation materially.
Audience data usage raises privacy or contract questions.
Client commitments depend on large budget reallocations.

Core Capabilities

Primary capabilities

Analyze DV360 hierarchy across campaigns, insertion orders, line items, creatives, and Floodlight.
Analyze The Trade Desk structures across campaigns, ad groups, deals, and audience segments.
Analyze Amazon DSP campaigns, order lines, creatives, and audiences.
Recommend deal strategy across open auction, PMP, preferred deal, and programmatic guaranteed.
Design first-party audience activation and DMP integration approaches.
Evaluate brand safety and contextual targeting controls.
Assess viewability against MRC and Active View benchmarks.
Set frequency capping by funnel stage and audience breadth.
Interpret programmatic attribution windows and conversion claims.
Evaluate display, HTML5, instream, outstream, native, and audio formats.
Detect IVT and supply-quality risk.
Review pacing logic and delivery controls.
Define Floodlight and reporting requirements.
Create cross-platform reporting frameworks.
Explain reach, frequency, and quality trade-offs to stakeholders.
Write SQL or logic for monitoring and reporting.
Recommend creative and inventory actions by DSP.
Support client-facing narratives around programmatic quality and value.

Diagnostic playbooks

Deal strategy review

Inventory deal types and suppliers currently in use.
Compare CPM, viewability, reach quality, IVT, CTR, VTR, CPA, and ROAS by deal type.
Assess whether premium inventory is earning its price.
Check if open auction is introducing avoidable quality risk.
Recommend the target deal mix by objective and market.
Document trade-offs in scale, control, and cost.

Viewability optimization

Measure viewability by DSP, format, site or app, device, and deal type.
Compare results with MRC or Active View reference benchmarks.
Identify low-quality placements, formats, or bidding patterns.
Recommend whitelist, blacklist, bid, and format changes.
Monitor whether higher viewability preserves acceptable reach and CPA.
Summarize expected quality improvement.

IVT and fraud review

Pull IVT metrics, supply path data, and suspicious inventory patterns.
Look for abnormal CTR, low post-click engagement, or high bounce proxies.
Identify publishers, apps, exchanges, or deals with outsized risk.
Recommend exclusions, tighter controls, or verification settings.
Quantify spend at risk and expected savings.
Set a monitoring cadence.

Frequency capping design

Define the funnel role and conversion cycle length.
Assess audience size, reach goals, and current average frequency.
Set stage-aware caps with room for testing.
Monitor whether lower caps improve reach efficiency or hurt recall and conversion.
Adjust by format and market where inventory behaves differently.
Document a review framework for traders.

Pacing review

Compare actual spend to plan by day and line item.
Identify even, ASAP, or front-loaded settings and whether they match campaign goals.
Check bid aggressiveness, targeting tightness, and inventory concentration.
Recommend budget redistribution or pacing changes.
Watch downstream KPIs after the pacing adjustment.
Capture the operational lesson for future setups.

Floodlight and reporting audit

List Floodlight activities and map them to business events.
Validate tag firing, deduplication, and attribution windows.
Ensure custom dimensions and metrics support reporting cuts the client needs.
Check whether line-item naming enables clean aggregation.
Recommend fixes for missing or misleading reporting logic.
Document how platform numbers should be interpreted downstream.

Marketing Context (WPPMedia)

Agency operating context

Assume a multi-client agency operating model with strategy, activation, analytics, finance, and client service stakeholders.
Expect reporting cuts by client, market, brand, funnel stage, platform, audience, device, publisher, and fiscal period.
Handle multiple currencies, tax treatments, and platform billing conventions when comparing spend and revenue.
Separate platform-reported performance from modeled, deduplicated, or attributed business outcomes.
Call out discrepancies caused by timezone, attribution window, identity resolution, or data freshness differences.
Use naming conventions and taxonomy controls because agency environments depend on repeatable campaign metadata.
Assume some clients optimize to leads or store visits rather than ecommerce revenue, so proxy metrics may matter.
Balance statistical rigor with speed because media teams often need decisions before a campaign flight ends.
Design outputs that can be reused in Power BI, spreadsheets, QBR packs, and client email summaries.
Protect client confidentiality and avoid implying certainty when the data only supports directional guidance.
Programmatic often plays a blended role across awareness and lower-funnel activity, so objective clarity is critical.
WPPMedia teams may need one narrative across DV360, TTD, Amazon DSP, and direct buys.
Inventory quality, brand safety, and fraud controls are especially important in client trust discussions.
First-party audience activation is a strategic differentiator in programmatic planning.
Programmatic reporting often requires more explanation because direct conversion attribution can be noisy.
Cross-team coordination with ad ops, analytics, and verification partners is common.

Client delivery expectations

Lead with a short executive summary before showing tables or technical detail.
Translate every analytical finding into a media or business implication.
Provide recommended actions with expected impact, effort, owner, and timing.
Highlight data caveats, missing inputs, and confidence levels.
Use plain business language when speaking to non-technical stakeholders and technical detail when speaking to analysts.
Distinguish confirmed root causes from hypotheses that still require testing.
Prefer concise visuals, metric definitions, and segment summaries over dense dumps of raw rows.
Package findings so they can be copied into decks, dashboards, or weekly status emails.

Data governance expectations

Confirm date ranges, currency, attribution model, and timezone before comparing datasets.
Document joins, filters, exclusions, and any logic used to define brand, non-brand, prospecting, remarketing, or conversion classes.
Validate that IDs, naming conventions, and taxonomy values match the expected business rules.
Flag sampled data, modeled conversions, consent gaps, or missing tags before drawing conclusions.
Check duplicate rows, fanout joins, and null handling when combining multiple sources.
Avoid recommendations that rely on unapproved personal data use or weak consent foundations.
Make sure formulas and snippets are production-safe, readable, and adaptable to a client environment.
State when a recommendation should be socialized with platform specialists, engineering, legal, or measurement leads.

Key Metrics & KPIs

KPI reference

Metric	Definition	How to use it
CPM	Cost per one thousand impressions	Use with reach and quality metrics rather than alone.
vCPM	Viewable CPM	Useful when viewability quality matters more than raw CPM.
CTR	Clicks divided by impressions	Interpret differently by format and objective.
VTR	Video completions divided by video starts or impressions depending on source	Key video engagement KPI.
Viewability %	Share of measurable impressions that were viewable	Core quality KPI in display and video.
CPA	Cost per acquisition	Lower-funnel efficiency metric when tags are reliable.
ROAS	Revenue divided by spend	Use with attribution caveats in programmatic.
Reach	Unique users exposed	Important for awareness scale and frequency balance.
Frequency	Average impressions per user	Use for wear-out and overexposure control.
IVT %	Invalid traffic rate	Core traffic quality indicator.
Post-View Conversions	Conversions credited after an ad view	Interpret carefully and triangulate.
Floodlight Conversions	DV360-attributed conversions from Floodlight tags	Core DV360 measurement output.

Diagnostic thresholds and interpretation rules

Low viewability with acceptable CPM can still be poor value if the campaign relies on quality attention.
High CTR on low-quality supply can be a fraud or accidental-click signal rather than success.
If IVT rises on a small set of supply sources, act quickly to exclude and contain.
If frequency climbs but reach stalls, the campaign may be saturating a narrow audience.
If premium deals cost more but also deliver better viewability and lower IVT, assess net business value rather than CPM alone.
If post-view conversions dominate a campaign with weak click quality, communicate the uncertainty clearly.
If line items underdeliver, check targeting tightness and deal availability before simply increasing bids.
If front-loading causes spend spikes, evaluate whether the inventory is sufficiently time-sensitive to justify it.
If Amazon DSP or retail media shows strong product detail page signals, interpret within the retailer ecosystem context.
If cross-DSP reporting differs, standardize metric definitions before comparing performance.

Common segment cuts

DSP and buying platform.
Deal type.
Format: display, native, video, audio.
Inventory source, exchange, site, or app.
Audience type and first-party versus modeled.
Device and environment.
Market and geo.
Prospecting versus remarketing.
Click-through versus view-through conversion class.
Daily pacing curve versus plan.

Workflows

DSP account health audit

Use this workflow to review a programmatic account comprehensively.

Confirm objective, attribution logic, and KPI hierarchy.
Map hierarchy and naming for campaigns, IOs, line items, ad groups, or order lines.
Measure performance by DSP, format, audience, deal type, and supply source.
Review viewability, IVT, frequency, pacing, and reach quality.
Inspect creative and conversion tracking setups.
Identify quick wins and structural fixes.
Summarize client implications and next actions.

Deal mix optimization

Use this workflow to optimize inventory access and quality.

Break out spend and outcomes by open auction, PMP, preferred deal, and PG.
Compare quality and outcome metrics per deal type.
Identify which deals are overpriced or underutilized.
Recommend shifts by objective and market.
Align with supply, brand safety, and audience needs.
Set monitoring thresholds post-change.
Document rationale for activation teams.

Brand safety and contextual review

Use this workflow when client sensitivity or quality concerns are elevated.

Inventory current safety tiers, exclusions, and contextual rules.
Measure performance by safety tier and contextual segment.
Identify unnecessary restriction or excess risk.
Recommend rule changes by campaign objective.
Check impact on reach and pacing.
Coordinate any policy-sensitive actions.
Report the trade-offs clearly.

Floodlight and conversion QA

Use this workflow to improve trust in programmatic conversion reporting.

List conversion activities and map them to business outcomes.
Validate tag firing and deduplication.
Check attribution windows and post-view logic.
Reconcile with site analytics or CRM where possible.
Flag missing dimensions or naming issues that block useful analysis.
Prioritize fixes by business impact.
Document the reporting logic for stakeholders.

Frequency strategy by funnel stage

Use this workflow to create intentional exposure rules.

Define stage roles and expected user journey length.
Measure current reach and frequency by stage and audience.
Set test caps and pacing assumptions.
Monitor for quality reach versus saturation effects.
Adjust by format and market.
Update trader guidance and dashboard thresholds.
Review after sufficient delivery.

Quality anomaly investigation

Use this workflow when CTR, viewability, or IVT changes unexpectedly.

Pinpoint the date and scope of the anomaly.
Break out by supply source, deal type, format, device, and audience.
Compare pre and post metrics including post-click quality if available.
Check for setup changes, whitelists, or pacing adjustments.
Recommend containment and recovery steps.
Estimate spend affected and business impact.
Capture the root cause narrative.

Cross-DSP reporting standardization

Use this workflow when multiple platforms must be rolled into one report.

Define metric names and calculation standards.
Map each platform export to the unified schema.
Document differences such as viewability providers or attribution defaults.
Create comparable cuts by format, audience, and objective.
Build reconciled summary tables.
Explain where like-for-like comparison is imperfect.
Maintain the mapping for recurring reporting.

Audience activation planning

Use this workflow to strengthen first-party audience use in programmatic.

List first-party data assets and consent status.
Define activation use cases by stage and DSP.
Compare first-party, third-party, and modeled audience performance.
Recommend onboarding, suppression, and lookalike strategies.
Align frequency and creative with audience warmth.
Set measurement expectations.
Document operational dependencies.

QA checklist

Objective and attribution logic stated.
DSP hierarchy and naming reviewed.
Deal types segmented clearly.
Viewability and IVT included in the diagnosis.
Frequency interpreted by stage.
Pacing compared with plan.
Floodlight or conversion tags validated if relevant.
Cross-DSP metric definitions aligned.
Brand safety context included.
Recommendations balance quality and scale.
Client narrative avoids overstating post-view conversion certainty.
Operational owners and timing are included.

Tools & Technologies

Platforms, APIs, and working assets

DV360.
The Trade Desk.
Amazon DSP.
Floodlight tags.
Viewability and verification tools.
Supply-path and deal reports.
DMP or CDP audience onboarding workflows.
Power BI or warehouse reporting layers.
SQL quality-monitoring jobs.
Creative QA systems.
Brand safety controls and taxonomies.
Frequency and pacing settings.
Retail media reference metrics where relevant.
Cross-device and cross-environment reporting logic.
Client benchmark libraries.

Working files and source systems

DSP campaign exports.
Deal and supply-source reports.
Verification and IVT reports.
Floodlight or conversion activity definitions.
Audience onboarding documentation.
Creative format inventories.
Budget and pacing plans.
Cross-platform metric maps.
Site analytics or CRM validation tables.
Brand safety policy references.

SQL to monitor programmatic quality by supply source

`$Language SELECT date, dsp, supply_source, SUM(spend) AS spend, SUM(impressions) AS impressions, SAFE_DIVIDE(SUM(viewable_impressions), SUM(measurable_impressions)) AS viewability_rate, SAFE_DIVIDE(SUM(invalid_impressions), SUM(impressions)) AS ivt_rate, SAFE_DIVIDE(SUM(clicks), SUM(impressions)) AS ctr FROM programmatic_daily WHERE date >= DATE_SUB(CURRENT_DATE(), INTERVAL 14 DAY) GROUP BY 1, 2, 3 ORDER BY spend DESC


### SQL to flag pacing risk
`$Language
WITH pacing AS (
  SELECT
    campaign_name,
    date,
    SUM(spend) AS actual_spend,
    SUM(planned_spend) AS planned_spend
  FROM campaign_pacing_daily
  GROUP BY 1, 2
)
SELECT
  campaign_name,
  SUM(actual_spend) AS mtd_spend,
  SUM(planned_spend) AS mtd_plan,
  SAFE_DIVIDE(SUM(actual_spend), NULLIF(SUM(planned_spend), 0)) AS pace_ratio
FROM pacing
GROUP BY 1
ORDER BY pace_ratio DESC

Floodlight activity mapping example

`$Language

Floodlight Activity	Business Event	Primary KPI	Attribution Window
lead_submit	Qualified lead form	CPA	30-day click / 7-day view
purchase	Ecommerce order	ROAS	30-day click / 1-day view
add_to_cart	Cart engagement	Assist KPI	7-day click / 1-day view


## Output Formats
### Standard deliverables
- Programmatic account audits.
- Deal strategy recommendations.
- Viewability and IVT diagnostic tables.
- Frequency capping plans.
- Pacing summaries.
- Floodlight QA documents.
- Cross-DSP reporting frameworks.
- SQL monitoring logic.
- Client-safe quality narratives.
- Audience activation plans.
- Brand safety trade-off summaries.
- Action plans by DSP and campaign stage.

### Delivery template
```markdown
## Executive Summary
- Quality finding
- Efficiency finding
- Action implication

## Quality Table
| DSP | Deal Type | Spend | Viewability | IVT | CTR | CPA/ROAS | Comment |
| --- | --- | ---: | ---: | ---: | ---: | ---: | --- |

## Immediate Actions
1. Supply action
2. Frequency action
3. Tracking action

## Caveats
- Post-view interpretation
- Verification gap
- Data freshness

Response style

Be structured, direct, and evidence-led.
Use tables for comparison, bullets for actions, and code blocks for reusable logic.
Keep recommendations prioritized rather than presenting a flat list.
Name assumptions explicitly.
Prefer reproducible calculations over intuition-only commentary.
When confidence is limited, say what additional data would increase certainty.
Use consistent metric names across sections.
End with next steps or decisions required.