Media budget at large companies: how to decide the next unit of investment
An allocation model that weighs marginal return, uncertainty, operational capacity and brand goals, not just average performance.
Allocating a media budget means choosing where the next tranche of investment has the highest expected value, given the business's constraints. The best average performance in the past does not guarantee the best return on the next increase. Saturation, product availability and uncertainty must all enter the decision.
At large companies, the budget is usually split across brands, regions, channels and objectives. Each team presents a different metric. Comparing those numbers directly can create a false contest: a reach campaign and a demand capture campaign do not answer the same question.
Organize the portfolio by function
Separate resources for brand building, commercial activation, channel upkeep and learning. This does not call for universal percentages. It calls for a rationale for each function, an evaluation horizon and review criteria. A balanced plan for one business can be wrong for another.
Define real constraints: service capacity, inventory, geographic coverage, contractual commitments and frequency limits. Growing investment in a region without delivery capacity can raise cost and degrade the experience. The media plan has to talk to the operational plan.
Compare scenarios, not just rankings
- Continuity scenario: what happens if investment stays close to the current level?
- Expansion scenario: where is there room to increase, and what evidence supports that hypothesis?
- Reduction scenario: which effects and commitments must be considered when pulling budget?
- Learning scenario: which test would reduce an important uncertainty ahead of the next decision?
Present assumptions and result ranges. Models can help estimate response curves, but extrapolating far beyond observed levels increases dependence on assumptions. Meridian is a public reference for modeling and optimization; using it requires assessing whether your data and specification are adequate.
Create a reallocation cadence
The review should follow the speed of the business and the time needed to observe effects. Operational indicators can change daily; strategic portfolio decisions do not need to follow every swing. Set variation limits and define the situations that call for an extraordinary review.
In a hypothetical scenario, a company might preserve brand investment and redirect part of its activation across regions with different commercial capacity. The justification is not simply to reward the channel with the highest ROAS: it is to improve the total result under the existing constraints.
What to bring to the committee
Show current investment, proposed change, evidence, uncertainty and review condition. State whether the comparison uses attributed results, incremental effect or modeled projection. These numbers are not interchangeable. Include the execution and production costs required for the reallocation to happen.
A well-documented decision lets you learn even when the result differs from what was expected. If the hypothesis was clear, the team can investigate whether execution failed, the context changed or the estimated relationship was wrong.
Is there an ideal split between brand and performance?
There is no ratio that works automatically for every business. Category, maturity, purchase cycle and objectives change the decision.
Should all budget go to the channel with the highest return?
Marginal return and constraints matter more than a historical ranking. The Media practice connects planning to incremental measurement.