Skip to content
Faisal Mahmud TusherDigital marketing, SEO & analytics
with Faisal Tusher

Media Planning & Buying › Media Measurement

Post-Campaign Analysis for Media Plans

How to run a post-campaign analysis: delivery against plan, reach and frequency achieved, cost efficiency against the buy, creative and placement learnings.
Guideintermediate9 min readUpdated

What a post-campaign analysis is for

Every media plan rests on assumptions: the size of the audience, the cost per thousand each medium will charge, the reach each line will add, the overlap between them, the frequency the message needs. Most of those assumptions are educated guesses the first time. The post-campaign analysis is where they meet what actually happened.

It has three jobs. The first is accountability: did the buy deliver what was paid for, and if not, has the shortfall been recovered? The second is evaluation against the media objectives: did the plan reach the audience it promised, at the frequency it intended? The third, and the most valuable over time, is learning: what should change in the next plan's mix, costs, schedule or creative? The campaign wrap-up checklist makes sure the groundwork is done first: every line switched off, final numbers reconciled and the learnings filed where the next brief will find them.

The analysis is not an effectiveness study. It can say that the media reached a large share of the target audience at the planned frequency and cost; it cannot by itself say that the media caused sales to rise. That question needs methods built for causal inference, which the media measurement hub introduces as routes. Keeping the two separate stops a good delivery report being presented as proof of impact, and stops a weak sales month being blamed on media that ran exactly as planned. The IPA's effectiveness work has long encouraged advertisers to distinguish what the media delivered from what the campaign achieved, and to evaluate both.

For broadcast, podcast and print lines, the guide to measuring TV, radio, podcasts and print covers the evidence for what the campaign achieved, from codes and vanity addresses to branded search lift and regional tests.

Delivery against plan

The starting point is a line-by-line comparison of what the frozen plan said would run with what ran. Every line in the schedule gets a row.

MeasurePlannedDeliveredWhat a gap usually means
Weight (impressions, ratings, spots, sites)From the plan and insertion orderFrom reconciled delivery reportsUnder-delivery, bumped spots, sold-out inventory, pacing problems
Dates and flightingPlanned flight datesActual on-air or live datesLate creative, late booking, technical delays
Placement and contextAgreed programmes, sites, sectionsPlacements reported and verifiedSubstitution, run-of-site where specific placements were bought
Audience deliveryTarget audience share of deliveryMeasured or verified audience shareBroad targeting, weak data segments, panel variation
Make-goods and added valueAgreed in the insertion orderActually deliveredMake-goods outside the useful window, bonus inventory not supplied
SpendPlanned and contracted costInvoiced and reconciled costOver-billing, unagreed fees, currency or tax changes

The analysis should use reconciled delivery figures, not the first report the seller sent. Where the buyer has already agreed make-goods or credits, the table shows both the original shortfall and how it was resolved. A pattern of gaps by seller or medium is often more useful than any single number, because it shapes the next round of negotiation. Where the team kept weekly delivery checks and a change log during the campaign, as the tutorial on managing a media plan while it runs sets out, the table can separate shortfalls the sellers caused from changes the team decided. Gathered across a season's campaigns, those seller-by-seller gaps become the evidence for reviewing your media owners before the next bookings.

Reach and frequency achieved

The next question is whether the plan reached the audience it promised, at the frequency it intended. Reach and Frequency achieved come from different sources by medium: industry audience measurement for television, radio, out-of-home and print, platform reports for digital, and modelled estimates for the combined plan.

Three cautions apply. First, cross-media reach cannot be found by adding each medium's reach together, because the same people are reached by several media; the cross-platform reach and frequency guide covers deduplication methods and their limits. Second, the average frequency matters less than the distribution: a campaign can hit its average while showing the message many times to a small group and once to everyone else. Third, reach achieved should be measured against the target audience, not the total audience delivered.

The reach and frequency calculator is useful for checking whether a reported reach figure is plausible given the impressions delivered and the audience size. If the plan fell short of its reach goal, the analysis should say why: too few media, too much weight on one platform's audience, a flight that was too short, or a planning assumption about overlap that proved optimistic.

It also helps to compare the achieved distribution with the effective frequency range the plan set. If the plan aimed for most of the audience to see the message several times, the analysis should show what share actually did, what share saw it only once and what share saw it far more often than intended. Heavy over-exposure among a small group is a sign that frequency capping, the mix of media or the targeting of one line needs to change; widespread single exposure suggests the plan spread its weight too thinly or ran too briefly for the message to register.

Cost efficiency against the buy

Efficiency is judged against what was agreed, not against a generic benchmark. For each line, compare the planned cost per thousand of the target audience (CPM) with the cost actually achieved, using reconciled spend and verified target-audience delivery. The cost per thousand calculator does the arithmetic.

A line can look efficient on its native metric and inefficient on a target basis, or the reverse. A cheap display line that delivered most of its impressions outside the target audience, or below viewability standards, was not cheap. A television line that cost more than planned but delivered more of the target audience may have been good value. The analysis should show both the native rate and the target-audience rate so readers can see which drove the result.

For performance lines, outcome rates such as cost per lead or cost per sale belong in the analysis too, judged against the KPI each line was set. They should sit alongside the delivery measures, not replace them, and they should not be used to rank broad-reach lines against performance lines.

Creative, placement and timing learnings

Some of the most useful findings are not about cost or delivery but about what worked within the media. The analysis should look for patterns across:

Creative. Which formats and lengths delivered better completion, attention or response; whether creative wear-out appeared later in the flight; whether versions adapted for a medium outperformed versions simply resized.

Placement. Which programmes, sites, positions or contexts delivered the target audience most efficiently; which placements under-delivered or carried quality concerns; whether premium placements justified their cost.

Timing. Which dayparts, days and weeks delivered best; whether the flighting pattern held the intended weight through the key period; whether competitor activity changed the context.

Learnings should be stated as hypotheses for the next plan rather than conclusions, because a single campaign rarely isolates one variable. Where a learning matters enough to act on at scale, the next plan can build in a test, and the incrementality testing guide covers how to design one that measures real effect.

Illustrative example — not a real result

A streaming service reviews its autumn campaign

Shimmerdrift, an invented streaming service, reviews a six-week campaign with a £380,000 budget. Delivery: television delivered 94% of booked ratings, with make-goods agreed for the remainder; online video over-delivered by 8% at no extra cost. Reach: the plan aimed for 60% of adults aged 18 to 44 at three or more exposures and achieved an estimated 54%, because online video overlapped with television more than assumed. Efficiency: the target-audience cost per thousand for out-of-home came in at £6.40 against £5.80 planned, while podcast sponsorship came in at £9.10 against £12.00 planned. Learning: the next plan moves weight from online video into podcasts and audio to add new reach.

Writing the analysis and the next plan

The written analysis should be short enough to be read and specific enough to be used. A workable structure has five parts: a summary of the objectives and whether they were met; the delivery table with gaps explained; reach and frequency achieved against the goal; cost efficiency by line; and the learnings framed as changes for the next plan. Effectiveness results, if any, appear as a separate section attributed to the analytics work that produced them.

The monthly marketing report template offers a format for the recurring reporting that sits around individual campaign reviews. The post-campaign analysis is the deeper document, written once per campaign. Where the review must also cover root causes, process problems and agreed changes beyond the media, the campaign post-mortem template takes this analysis as one of its inputs.

Write for two readers. The budget holder wants to know whether the money was well spent and what will change; the next planner wants the detail behind every revised assumption. A one-page summary at the front serves the first, and the line-level tables and notes behind it serve the second. Share the analysis with the buyers and, where relationships allow, with the main media owners, because the learnings about placement and delivery are often the most useful input to the next negotiation.

The final section matters most. Each learning should map to a specific change in the next plan's assumptions: a revised cost per thousand for a medium, a different overlap allowance, a shorter or longer flight, a placement to drop or add, a creative format to test. If the next plan does not cite the last analysis, the analysis has not done its job. Adding each line's cost and delivery to a media cost history kept from your own buys keeps those revised assumptions available to every later plan, not only the next one.

Common mistakes

  • No frozen baseline. Without the signed plan saved as it was, there is nothing to compare against.
  • Summing reach across media. Overlap makes the total smaller than the sum; use a deduplicated estimate.
  • Using unreconciled data. First delivery reports often change once make-goods and credits are agreed.
  • Treating delivery as effect. A plan that delivered perfectly may still not have changed behaviour, and the analysis should not imply otherwise.
  • Judging all lines on outcome rates. Broad-reach lines will look poor on cost per sale; that was not their job.
  • Stating learnings as conclusions. One campaign rarely isolates a cause; frame learnings as hypotheses to test.
  • Filing the analysis. If its findings do not change the next plan's assumptions, it was wasted effort.

Frequently asked questions

When should a post-campaign analysis be done?

Soon enough that the team still remembers the decisions and the next plan has not been written, but late enough for final delivery data, reconciled invoices and any audience measurement to arrive. For offline media the final audience data can lag the campaign, so many teams publish a first read on delivery quickly and a fuller analysis once the data settles.

Is a post-campaign analysis the same as proving the campaign worked?

No. It shows whether the media was delivered as planned, to the audience planned, at the cost agreed, and what can be learned about the media choices. Whether the media caused a change in sales or behaviour is a separate question answered by lift tests or mix modelling, which the analysis can reference but does not replace.

Who should write the post-campaign analysis?

The planner usually leads, with the buyer supplying delivery, cost and reconciliation data and the analytics team supplying any effectiveness results. The planner is best placed to judge the figures against the plan's intent, and the analysis is most useful when written by the person who will build the next plan.

Terms used on this page

Tools to try

Frameworks

Where this page is used

Also related

Sources

  1. Institute of Practitioners in Advertising — IPA

Related