What is budget optimization?

Once a model has a response curve for every channel, the budget becomes an optimization problem: spread a given total across channels and weeks so that the marginal returns are as level as the constraints allow, or find the total at which the last unit still clears the return the business requires. The first is allocation; the second is the budget level. Both run on the same curves.

Allocation mode takes a fixed total and distributes it across channels and time periods, inside the constraints you set: a TV contract already signed, a floor on brand, a channel held steady for a test, a channel excluded this quarter. Each constraint gets a price. If locking TV at its contracted level lowers the forecast return, the scenario shows by how much, so the trade-off is visible rather than hidden.

Level mode takes a return target instead. Set a minimum marginal ROAS or a maximum marginal CAC, and the optimizer finds the total where every additional unit still clears it. That turns "marketing wants more" into "at what return do we stop?", a question a CFO can answer and defend.

Reallocation alone, before any increase, typically gives 5 to 15% more effect from the budget you already have, by moving money from channels that have flattened to channels with headroom. Because digital and offline sit in one model, TV, radio and out-of-home are optimized on the same terms as search and social.

The optimizer treats channels as independent saturable resources; it does not model interactions between them. A reallocation is also a forecast with a range, verified against actual sales the following months, which is why big moves are best made in steps or as deliberate tests. Scenarios can be run any time in the Odins platform or from your own AI assistant; the monthly recommendations are the reviewed version.

• Scenario planning: Forecasting what a budget would return before committing it, and comparing several futures side by side.
• Marginal ROAS: What the next unit of spend in a channel earns at today's level. The number a budget change actually asks for.
• Marginal CAC: The cost of the next customer in a channel at today's spend. The CAC version of marginal ROAS, for businesses that count customers.
• Saturation curve: The response curve per channel: how sales respond to spend, and where the next unit stops paying.
• Diminishing returns: Each extra unit of spend in a channel earns a little less than the one before. Why the average is the wrong number to plan on.

Go one level deeper: Scenario planning and budget optimization. Allocate budget across channels. Or go back to all terms in the glossary.