Campaign readout workflow
Turn campaign reports into bounded decisions.
Organization byline: Measurement Press editorial desk. This page does not list an individual author or reviewer. Read the editorial method and corrections policy.
Campaign reports often mix delivery, traffic quality, lead status, matchbacks, attribution, survey movement, and renewal language into one performance story. This workflow keeps each evidence lane separate until the report can support a clear decision.
Use it before a campaign readout meeting, when a report is being reused as renewal evidence, or when an attributed result is starting to sound like proof of incremental impact. The goal is not to slow down the decision. The goal is to make the next action match the evidence that is actually visible.
Source boundary: platform counts are not interchangeable
Official measurement documentation supports the review questions on this page, not a conclusion about any particular campaign. Campaign Manager 360 describes conversion windows, counting methods, and click-through versus view-through classification; Google Analytics describes attribution as a credit-allocation model and says data-driven attribution uses converting and non-converting paths; Google also documents expected discrepancies across systems when windows, counting rules, attribution day, or tracked activities differ.
| Source-backed point | Evidence to request | Decision consequence |
|---|---|---|
| Floodlight conversions depend on the configured conversion window and counting method; click-through and view-through conversions are classified separately. | Capture the Floodlight activity, window, counting method, event definition, and click/view-through split used in the report. | Do not compare a headline conversion total with another period until the definitions and windows are aligned. |
| Analytics attribution assigns credit across touchpoints; its data-driven model uses converting and non-converting paths and a counterfactual holdback comparison. | Record the attribution model, key-event definition, lookback window, and whether a holdback or other comparison exists. | Treat attributed credit as a reporting allocation unless the readout shows a valid comparison that supports an incrementality claim. |
| Google documents expected discrepancies between Campaign Manager 360, Google Ads, and Analytics when counting, attribution day, time zone, windows, or tracked conversion sets differ. | Reconcile platform, report, time-zone, conversion-set, and invalid-traffic settings before explaining a gap as campaign performance. | Route unresolved gaps to measurement QA or status-window closeout; do not write a causal renewal sentence from an unexplained difference. |
- Campaign Manager 360: How Floodlight counts conversions — conversion windows, counting methods, and click-through/view-through handling.
- Google Analytics: Get started with attribution — attribution models, path data, and counterfactual framing.
- Campaign Manager 360: Floodlight conversion discrepancies — why platform totals can diverge even when teams compare the same campaign name.
Use these sources to verify the implementation and reporting definition in front of you. They do not turn attributed conversions, viewability, or a lower reported cost per lead into proof of incremental impact.
Pick the next action before writing the renewal sentence.
Use the first weak lane in the readout to choose a closeout, repair, scorecard, issue, memo, or next-test workflow before the report turns into stronger language than the evidence supports.
What the workflow protects
Delivery clarity
Served, measured, viewable, valid, qualified, and completed activity should stay separate from outcome claims.
Outcome maturity
Lead status, matchbacks, sales follow-up, attribution windows, and survey fields need closeout rules before final wording.
Comparison discipline
Prior periods, matched groups, holdouts, model baselines, and expected demand should be visible before lift language appears.
Renewal memory
Weak lanes, exclusions, owners, and next-test conditions should survive the meeting and carry into the next flight.
Choose the decision stage
| Stage | Reader question | Use first | Then inspect |
|---|---|---|---|
| Pre-buy proof | Does the package show reader intent, eligible inventory, creative fit, reporting grain, and claim boundaries before launch? | Contextual package proof checklist | audience intent map, RFP evidence checklist, package proof sheet, and sponsor fit checklist |
| Launch readiness | Can the campaign be trafficked, measured, and reported without losing placement, creative, destination, source, or outcome fields? | Campaign readiness dashboard | inventory readiness matrix, tagging QA, creative acceptance, and landing-page launch QA |
| Readout QA | Is the report separating delivery, exposure quality, traffic quality, outcome status, comparisons, slices, and causal language? | Campaign readout QA checklist | campaign evidence triage tree, baseline comparison checklist, outcome quality scorecard, and evidence-to-claim language matrix |
| Status-window closeout | Are delayed leads, sales follow-up, matchbacks, survey returns, and attribution windows mature enough for final language? | Status-window closeout checklist | closeout register, attribution window checklist, identity matchback checklist, and brand lift readout checklist |
| Renewal or next test | Should the next action be renewal, creative revision, landing-page repair, budget shift, missing-evidence request, or a designed incrementality test? | Campaign evidence triage tree | renewal scorecard, issue register, renewal memo, evidence archive, and archive owner and gate fields |
Compare the readout against a worked failure mode.
Before a campaign result becomes renewal language, match the weak lane to a case study that shows how similar evidence can overstate lift, demand, or persuasion.
Renewal-decision route map
After the first readout pass, match the weakest visible lane to the next action. This keeps renewal language, package proof, and measurement method choice in separate decisions.
| Readout finding | Decision posture | Use first | Route next |
|---|---|---|---|
| Lead status, matchbacks, attribution windows, sales follow-up, or survey returns are still aging. | Wait for closeout before final wording. | Status-window closeout checklist | closeout register, conversion lag checklist, and identity matchback checklist |
| Served activity is high, but measurability, viewability, invalid traffic, frequency, or placement context is unclear. | Repair exposure quality before renewal. | Viewability and invalid traffic checklist | reach and frequency checklist, programmatic inventory QA, and inventory readiness matrix |
| Response differs sharply by creative, context, destination, landing page, or lead routing. | Revise the path before calling the media package weak or strong. | Creative and destination troubleshooting matrix | creative testing readout, landing-page lead quality, and landing-page launch QA |
| Outcomes are visible, but the baseline, holdout, prior period, or matched comparison is weak. | Renew cautiously or design a stronger test. | Campaign baseline comparison checklist | measurement method selector, incrementality test plan, and uncertainty interval checklist |
| Reader fit and package context look promising, but buyer proof is incomplete. | Renew only with clearer package evidence. | Private marketplace renewal scorecard | buyer-readiness desk, package proof sheet, and RFP evidence checklist |
| A budget expansion, causal claim, or strategic recommendation depends on attributed or matched outcomes. | Choose a stronger evidence design before stronger language. | Measurement method selector | budget decision evidence ladder, geo lift checklist, and randomized lift readout checklist |
| The report is useful, but exclusions, missing fields, owners, or next-flight conditions remain open. | Carry a conditional decision into the next flight. | Campaign issue log and renewal register | renewal memo, evidence archive, and archive owner and gate fields |
The opening pass
1. Name the decision.Write the decision before reading the strongest number: renew, revise creative, repair the destination, shift budget, request missing fields, wait for mature outcomes, or design a cleaner test.
2. Split delivery from outcomes.Keep impressions, reach, frequency, viewability, invalid traffic, clicks, qualified visits, leads, sales follow-up, matched outcomes, and survey movement in separate rows.
3. Find the comparison.Name the baseline the report is asking readers to use. If the baseline is missing or chosen after results were visible, downgrade impact language before discussing renewal.
4. Check maturity.Decide whether outcome fields are final, aging, incomplete, or excluded. Late status changes should move to a closeout register instead of being treated as stable evidence.
5. Write the allowed claim.Use the strongest sentence the evidence can support without promoting attribution, response, or delivery quality into causal lift.
Worked five-lane readout pass
Assume a contextual sponsorship report says a simulated flight spent $62,000 and produced 420 form fills, compared with $80,000 and 330 form fills in a prior period. The current reported cost per form is $62,000 ÷ 420 = $147.62; the prior cost per form is $80,000 ÷ 330 = $242.42; the reported change is 1 − ($147.62 ÷ $242.42) = 39.1% lower. That arithmetic is checkable, but it is not yet a renewal decision. Run the report through the five lanes before deciding whether to renew, revise, retest, or hold.
| Input or check | Value | Formula or interpretation |
|---|---|---|
| Served impressions | 120,000 | Constructed delivery total. |
| Measurable impressions | 96,000 | 96,000 ÷ 120,000 = 80.0% measurable. |
| Viewable impressions | 76,800 | 76,800 ÷ 96,000 = 80.0% viewable among measurable; 64.0% of served. |
| Status-closed form fills | 130 of 420 | 130 ÷ 420 = 31.0% outcome-status completeness; 69.0% remains open or unknown. |
| Five-lane gate result | 3 of 5 = 60.0% | Delivery grain, response path, and claim boundary pass; outcome maturity and comparable baseline fail. |
Reproduction: copy the stated inputs into a spreadsheet, calculate each ratio using the displayed numerator and denominator, and round percentages to one decimal place. Assumptions: all figures are synthetic, the form-fill definition is stable across periods, and the simulated report has no hidden deduplication or invalid-traffic adjustment. Limits: the example does not estimate incremental conversions, causal lift, profit, or generalizability; a different conversion window, counting rule, status definition, or baseline would change the result.
| Lane | What the report says | Review move | Allowed output |
|---|---|---|---|
| Delivery quality | Planned impressions delivered, but one placement family had lower measured viewability and higher frequency concentration. | Separate clean delivery from weak-exposure delivery before averaging the campaign. | Renewal can reference eligible delivery, not uniform exposure quality. |
| Response path | Most form fills came from two creative versions and one landing-page path. | Check whether the result is a media-package signal, a creative signal, or a destination-quality signal. | Keep package praise conditional on preserving the winning path. |
| Outcome maturity | Only 31% of leads have sales follow-up status, and several high-volume rows are still inside the normal qualification window. | Move pipeline language into status-window closeout until late lead fields mature. | Use lead-response language now; wait on qualified-pipeline language. |
| Comparison strength | The prior-quarter baseline used different creative, a shorter attribution window, and a larger returning-visitor share. | Downgrade the cost-per-lead comparison unless a same-window baseline or holdout can be produced. | Describe the result as directionally favorable, not proof of incremental efficiency. |
| Decision language | The team wants to renew at a higher budget because the headline cost per lead improved. | Write the decision around what survived review: clean placements, winning path, immature outcomes, weak baseline. | Renew a controlled next flight with fixed creative, placement exclusions, closeout dates, and a stronger comparison plan. |
The final sentence should be narrower than the dashboard headline: the campaign produced useful lead response in specific contexts, but the evidence supports a controlled renewal with closeout and comparison repairs rather than an unconditional budget expansion.
Downgrade rules
A readout can still be useful after a downgrade. The downgrade simply prevents the final sentence from claiming more certainty than the evidence can support.
| Finding | Do not claim | Use instead | Next page |
|---|---|---|---|
| Only attributed conversions or matched outcomes are visible. | The campaign caused the outcome. | The campaign received credit for observed outcomes under the stated attribution or match rules. | Attribution window checklist |
| Exposed users had higher prior intent, loyalty, search activity, or eligibility. | The audience strategy created the full difference. | The audience performed differently, but pre-existing selection may explain part of the gap. | Audience selection checklist |
| Lead quality or sales follow-up is immature. | The campaign produced qualified pipeline. | The campaign produced leads or handoffs that still need status-window closeout. | Status-window closeout checklist |
| Creative response is strong but destination quality is weak. | The media package underperformed. | The readout shows a response path that needs creative, destination, or lead-routing diagnosis. | Troubleshooting matrix |
| Delivery quality is clean but the outcome is noisy or too small. | The campaign failed or succeeded definitively. | The delivery evidence is usable, while the outcome evidence is below the decision threshold. | Uncertainty interval checklist |
Carry the result forward
Assign weak lanes
Use when a report has useful evidence and unresolved limits. Record the owner, exclusion, needed field, and allowed claim.
Write the renewal memo
Use when the team needs a final decision record that separates what was observed from what should happen next.
Preserve the conditions
Use when the next campaign should inherit unresolved rows, test requirements, launch gates, or delayed outcome checks.
Reader standard
A campaign readout is strongest when a reader can see the decision, evidence lane, comparison, maturity status, uncertainty, and final wording without guessing. If those pieces are visible, even an imperfect report can produce a practical next action without overstating what the campaign changed.