An unconstrained optimum is rarely an executable plan
A model can recommend moving budget between channels without knowing contractual commitments, audience availability, creative capacity or the time needed to ramp delivery. Those conditions belong in the decision before optimisation, not in a list of excuses after the model returns an answer.
- Define the objective
- Record commitments and bounds
- Optimise feasible scenarios
- Approve a staged move
Use the allocation guide for the mechanics. This checklist turns the business conditions into explicit constraints that can be reviewed and tested.
A budget recommendation needs credible response estimates, consistent value assumptions and feasible constraints. The marketing budget decisions learning path connects this idea to the assumptions, practical checks and next steps.
To work through the connected decisions, read MMM for profit: choose contribution over revenue alone and MMM saturation curves: know when you are extrapolating.
Write the budget identity first
Decide whether the total budget is fixed, capped or allowed to vary against a return target. Specify the planning horizon and currency. Make sure channel lower bounds do not already exceed the total and upper bounds leave enough capacity to allocate it.
For a fictional £100,000 plan, suppose television has a £40,000 committed minimum, search a £35,000 maximum and social a £30,000 maximum. The remaining £60,000 must fit within the available search and social capacity. If another requirement sets social's minimum to £35,000, the configuration is inconsistent and should fail before optimisation.
Distinguish hard constraints from preferences
A signed contract may be a hard minimum. A preference to avoid moving more than 10% may be a policy or a risk-control choice. Record the type, owner and rationale for each bound. Otherwise an allocation can appear model-driven while mostly reflecting undocumented preferences.
Consider ramp rates, minimum viable campaign sizes, inventory and sales capacity. A weekly recommendation and a quarterly commitment operate on different calendars. Explain how the plan will be implemented through time rather than assuming the full change happens immediately.
Compare robust scenarios
Run the same constraints across plausible response curves and cost assumptions. Use marginal return for the next unit of spend, while keeping historical average ROI as a separate descriptive measure. If a channel sits at a bound, explain whether that reflects a commercial restriction or the supported response range.
Report the current plan, proposed plan, expected incremental outcome, uncertainty and binding constraints. A constrained optimum is conditional on all of those inputs. It is not a guarantee of the maximum achievable business result.
Approve a staged move with a review date and rollback condition. The scenario-planning guide helps make the decision legible to finance and channel owners, including the trade-offs introduced by constraints.
Sources and further practice
This article’s worked example is illustrative, not a measured client result. Use the linked methodology to check the assumptions before applying it.