A programmatic campaign audit is a structured review of how media is bought, delivered, measured and optimized. It should do more than confirm that impressions were served. A useful audit connects campaign settings and supply quality to business outcomes, identifies avoidable waste and creates a prioritized action plan.
This checklist covers 25 checks across five areas: objectives and setup, inventory and supply, audiences and delivery, measurement and optimization, and commercial impact. Use it before launch, during an active flight or after a campaign closes. The exact menu names and available controls will vary by demand-side platform, ad server, verification provider and market, so treat platform mechanics as implementation details rather than universal rules.
How to run a programmatic campaign audit
Start by defining the audit period, markets, buying channels, insertion orders, line items, creatives, conversion events and reporting sources in scope. Export the relevant platform, analytics, ad-server and verification data before changing settings. Otherwise, optimization changes can make it difficult to distinguish historical issues from current performance.
For each finding, record the evidence, business impact, owner, recommended action and priority. A simple classification works well:
- Critical: a tracking, compliance, billing or delivery issue that can materially invalidate results.
- High: a quality or efficiency problem requiring prompt action.
- Medium: an improvement opportunity that should be tested or scheduled.
- Monitor: an observation without enough evidence for immediate intervention.
Keep the audit distinct from broader programmatic advertising strategy. The strategy sets the role of the channel; the audit tests whether execution supports that role.
1. Objectives and campaign setup
1. Confirm the business objective
Verify that the campaign has one primary objective, such as qualified reach, completed video views, site visits, leads or revenue. Secondary metrics can provide context, but optimizing simultaneously for incompatible outcomes often produces unclear decisions.
2. Check the conversion definition
Document what counts as a conversion, when it is counted and whether it represents a meaningful business action. For lead-generation campaigns, distinguish a form submission from a qualified opportunity where the available data supports that distinction.
3. Reconcile budgets, dates and pacing
Compare the approved budget and flight dates with insertion orders, line items and platform settings. Check daily caps, lifetime budgets, pacing methods and time zones. Look for settings that can cause early underdelivery, late spending or unplanned concentration.
4. Review geography, language and eligibility
Confirm that location, language, device, environment and other eligibility settings match the brief. Pay particular attention to exclusions for markets where the business cannot sell, serve or legally communicate.
5. Validate creative-to-placement fit
Check dimensions, file types, landing pages, click destinations, video lengths, audio requirements and call-to-action consistency. A technically accepted creative may still be unsuitable for the inventory or user experience where it appears.
2. Inventory and supply quality
6. Map the supply path
Identify the main supply sources, exchanges, publishers, apps, private marketplaces and direct deals in use. The goal is not to maximize the number of sources. It is to understand where impressions originate and whether each path adds reach, quality, control or useful data.
7. Review domain and app transparency
Assess whether reporting identifies the sites and applications receiving spend with enough detail for decision-making. Investigate unknown, sparse or unexpectedly broad inventory rather than treating every reported impression as equivalent.
8. Check brand-safety and suitability controls
Review category exclusions, keyword controls, curated lists, pre-bid protections and post-bid monitoring where applicable. Confirm that controls reflect the advertiser’s risk tolerance. A broad exclusion may reduce risk but also remove legitimate reach, so document the trade-off.
9. Assess invalid-traffic signals
Compare delivery and verification reports for suspicious activity, non-human traffic indicators, unusual placement patterns and discrepancies. Do not assume that a single score explains the whole problem. Investigate concentration by source, device, format, geography and time.
10. Examine viewability and attention proxies carefully
Review viewability or other available exposure-quality measures by inventory source and format. These indicators are useful diagnostics, not substitutes for outcomes. A high viewability rate does not prove that an impression was valuable, noticed or incremental.
11. Look for MFA and low-value supply patterns
Where reporting allows, investigate made-for-advertising characteristics, excessive ad density, refresh behavior, thin content and high-frequency domains. Use a documented definition and consistent evidence; avoid labeling a publisher based solely on one weak signal.
12. Evaluate frequency and duplication across buying paths
Compare exposure across line items, devices, publishers and channels. If identity resolution is incomplete, state that limitation clearly. The audit should identify probable duplication without presenting modeled reach as exact.
3. Audience and delivery controls
13. Verify audience source, purpose and freshness
List every audience used, including first-party, contextual, modeled, partner and platform segments. Confirm why each segment is included, when it was refreshed and whether it remains appropriate for the objective. Remove segments that cannot be explained or evaluated.
14. Separate prospecting, retargeting and retention
Review whether these groups are isolated in reporting and budget controls. Mixing them can make blended CPA or return metrics look stronger than the incremental contribution of prospecting actually is.
15. Check exclusions and suppression logic
Confirm that converters, existing customers, employees, restricted users and recent site visitors are treated according to the campaign plan. Suppression windows should reflect the buying cycle rather than being copied from another campaign without review.
16. Analyze delivery by meaningful dimensions
Break out spend, conversions, cost, revenue and quality indicators by device, operating system, browser, format, exchange, publisher, geography, hour and audience when sample size permits. Look for persistent differences, not isolated outliers.
17. Review frequency management
Check campaign, line-item and channel-level frequency controls where available. Compare exposure patterns with the sales cycle and creative rotation. A limit that is too low can restrict consideration; a limit that is too high can create waste and fatigue.
4. Measurement, attribution and optimization
18. Reconcile platform, ad-server and analytics totals
Expect some differences because systems use different counting rules, time zones, attribution windows, filters and definitions. The audit should explain material discrepancies rather than force all sources to match.
19. Validate tags, events and landing-page tracking
Test impression, click, viewability, video-completion and conversion events in a controlled manner. Confirm that consent requirements, redirects, cross-domain behavior and server-side or browser-side limitations are documented where relevant.
20. Review attribution windows and models
Record the lookback and view-through settings used in each reporting system. Compare platform-reported conversions with analytics or CRM outcomes without treating any one model as objective truth. Attribution is a decision framework, not a direct observation of causality.
21. Check optimization signals against the real goal
Determine whether bidding is optimized toward the event that best represents value. If the true goal is qualified pipeline but the campaign optimizes to inexpensive form fills, the setup may reward volume at the expense of quality.
22. Assess test design
Document what changed, what remained constant, the audience or inventory split, the measurement window and the decision rule. Avoid declaring a winner when multiple variables changed or when results are too limited to support a reliable conclusion.
23. Review change history and learning periods
Map major bid, budget, audience, creative and supply changes against performance. Frequent edits can make results difficult to interpret and may interrupt delivery patterns. Establish an approval and annotation process for material changes.
5. ROI and commercial impact
24. Calculate costs consistently
Separate media spend from platform fees, verification, data, creative, agency and other applicable costs. State whether reported CPA, ROAS or return includes these components. A media-only metric and an all-in metric answer different questions.
25. Connect delivery to incremental business value
Compare campaign outcomes with qualified leads, sales, margin, retention or other agreed commercial measures. Where possible, use controlled experiments, geo tests, holdouts or credible pre/post comparisons to assess incrementality. If those methods are unavailable, label the result as attributed or modeled rather than incremental.
For a deeper treatment of measurement choices and common efficiency traps, see Programmatic Advertising ROI: Metrics, Attribution and Common Traps. If the audit spans multiple formats, also review omnichannel programmatic advertising so reach and frequency are not evaluated in isolated channel silos.
What to do after the audit
Turn findings into a short action plan rather than a long list of observations. First fix measurement, compliance and billing risks. Next address high-impact supply, audience and pacing issues. Then create controlled tests for uncertain recommendations.
- Prioritize by expected impact and confidence. A small change supported by strong evidence may outrank a larger but speculative opportunity.
- Assign an owner and deadline. Include the buying team, analytics owner, creative team, client stakeholder or supply partner as appropriate.
- Preserve a baseline. Save reports and settings before changes so future performance can be compared fairly.
- Define the success measure. Specify whether the goal is better quality, lower waste, stronger delivery, more qualified conversions or improved commercial return.
- Schedule a follow-up. Recheck critical items promptly and establish a recurring review cadence for active programs.
When a campaign audit should trigger a strategic review
An audit may reveal that the problem is not execution alone. Reconsider the channel mix when the available inventory cannot reach the intended audience, the conversion path is too weak to measure, the buying objective conflicts with the available optimization signal or the campaign relies on assumptions that cannot be tested.
Programmatic can support broad reach, addressable audiences and automated buying, but it is not automatically efficient because it is automated. Strong performance depends on clear objectives, defensible measurement, suitable supply and disciplined optimization. A recurring audit makes those conditions visible before small issues become expensive ones.
For broader planning across paid channels, use the paid media hub as the next reference point.
