Run a Campaign Postmortem That Produces a Specific Next Decision

Do not turn the review into a victory speech

A campaign can generate many clicks while failing to produce useful enquiries, or meet its sales target despite confusing operations. A postmortem should examine what happened, why the team believes it happened, and what evidence is missing. It should not assign credit based on visibility or blame a person for every unexpected result.

Freeze the evidence set

Before discussing conclusions, record the campaign period, intended audience, offer, spend, assets, channels, tracking definition, and known interruptions. Preserve the version of the report used for the meeting. If the data later changes because of delayed imports or corrections, document the revision rather than quietly replacing the original figures.

Separate outcome from execution

Outcome questions concern qualified enquiries, bookings, sales, or another declared goal. Execution questions concern whether the correct offer ran, approvals were timely, landing pages worked, and staff could handle responses. A weak result with flawless execution suggests a different next step from a good result achieved through an unsustainable workload. Keep these dimensions separate.

Choose a narrow learning question

Avoid ending with “improve marketing.” Choose one question that the next campaign can test, such as whether a clearer eligibility line reduces rejected enquiries or whether a different response handoff reduces missed callbacks. State what observation would support or weaken the hypothesis. Where the sample is small, treat the result as a useful signal rather than proof of a universal effect.

Example: a local open house

A real estate office runs a weekend open house campaign. Registrations are high, but many attendees arrive outside the stated time window because one social graphic omitted the schedule. The postmortem separates audience interest from information quality and assigns one corrective test: use a shared event-detail block across every asset and measure the number of schedule-related calls next time.

Record action ownership

Finish with a short list of decisions, each with an owner, due date, evidence of completion, and the campaign where it will be checked. Archive the rest as observations rather than turning every comment into a task. A good postmortem reduces repeated mistakes and improves future choices; it does not guarantee that the next campaign will perform better.

Working template

Review note: objective | evidence set | outcome | execution issue | confounding event | one next test | action owner.

Sources and further reading

• When reporting results, disclose missing tracking and concurrent events instead of treating a post-campaign observation as proof of causation.