Investment Reviews
Part of Media investment reviews
Identifying spend that produced little useful evidence
Find media spend that failed to answer its intended question and distinguish weak results from missing evidence.
Spend produced little useful evidence when the campaign review cannot connect it to the question it was meant to answer. Identify the affected amount and the missing link. A poor measured result differs from activity whose delivery or outcome cannot be assessed.
Define the intended question
Recover the approved channel role. Was the buy meant to assess audience coverage, test a placement, generate accepted enquiries or estimate additional orders caused by a media change? Each question needs different records. Compare the evidence planned with what is available for the dates and audience that ran.
| Finding | What the review can say | What to investigate |
|---|---|---|
| Delivery and outcome are usable; result is weak | The observed result fell short under stated definitions | Audience, message, offer and commercial value |
| Delivery is known; outcome link is missing | Activity ran, but its business response is unknown | Event capture, qualification and reporting access |
| Outcome is known; placement detail is missing | The business result is recorded, but the bought context cannot be checked | Supplier export and booking identifiers |
| Both are incomplete | The spend did not answer the intended question | Whether any record can be recovered |
A missing field is not a zero result. Nor should a reporting gap conceal confirmed off-brief delivery.
Trace the gap to spend
Start with the line item, flight and actual charge, including directly required work where relevant. Identify the portion for which the intended question cannot be answered. If a report covers only some dates, separate those dates rather than calling the entire channel unmeasured. Record whether the gap arose in the booking, live setup, supplier report or business outcome path.
Check whether the evidence can still be obtained. A dated delivery export, accepted-enquiry file or corrected line-item mapping may resolve the question. Request definitions with the data, and compare reporting against booking and delivery records before treating an apparent omission as proof that a placement never ran.
If a people-level coverage figure is unavailable, preserve known impressions and spend, but mark people-level coverage unavailable. Do not relabel devices or impressions as people.
Judge the decision cost of the gap
Ask which decision the missing evidence prevents. A learning flight may have been poorly specified if its planned scale and records could never answer its question. A campaign delivered to the wrong market may provide ample data about an irrelevant audience. Conversely, weak short-term sales evidence may be expected for a channel whose approved job was broad introduction; assess the evidence agreed for that job.
For a service campaign, suppose spend and submitted forms are available but the business cannot distinguish accepted enquiries from duplicates for the flight. The review can describe delivery and submissions. It cannot report a reliable cost per accepted enquiry. Repair the record before using that measure to justify more spend, or narrow the claim made about the channel.
Finish with an evidence-gap register: amount affected, intended question, missing record, recovery owner, deadline and decision blocked. Use it to improve the next booking or measurement requirement.
