.png)
Marketing spend decisions carry more scrutiny than they used to. Attribution has become less reliable under privacy changes, the number of channels keeps growing and finance increasingly wants evidence that budget choices actually drive revenue. The question underneath all of it is simple to ask and hard to answer well: how much should you spend and where? Getting it right is part of a credible marketing measurement approach, not a separate exercise.
Deciding how much to spend is really a question about marginal return: the point where the next euro in a channel returns about the same as the next euro anywhere else. Every channel has diminishing returns, so the goal is to spend up to the point of efficiency, not to the point of exhaustion. Default methods, last year plus a bit, the loudest voice in the room, the platform’s own recommendation, all ignore saturation and tend to leave channels over- or under-funded. Estimating true channel contribution with marketing mix modelling, then testing changes before committing, is what turns allocation from a guess into a decision you can defend.
Most budgets start with last year’s numbers plus a little, perhaps shifting some money toward whatever channel is in favour. It feels safe, but it quietly leaves returns on the table, because it assumes the market has not moved and ignores channel saturation. When you cannot see true channel contribution, you fall back on proxy metrics, which corrupt the allocation: last-click hands all the credit to search, view-through windows flatter display and brand looks expensive because it does not produce immediate clicks.
Most allocation leans on one of three flawed approaches. The historical-percentage method splits the budget the same way every year, which assumes conditions never change. The executive-opinion method hands budget to the loudest voice or the best slides, so politics replaces evidence. Platform recommendations can be useful for optimising within a platform, but they do not optimise across your total marketing portfolio. Meta can tell you how to allocate within Facebook or Instagram, for example, but not whether the next euro would work harder in YouTube, TV, Search or somewhere else entirely.
The most useful idea in setting spend levels is diminishing returns. Every channel saturates. As an illustration, the first €100,000 in paid search might return €500,000 in revenue, the next €100,000 around €300,000 and the one after that €150,000, until additional spend barely moves the result at all.
This is why the right level of spend is the one that equalises marginal return across channels. That is why marginal ROI matters more for allocation than average ROI: a channel can have delivered an excellent historical return and still be the wrong place for the next euro if it is already close to saturation. Finding it requires knowing actual channel contribution and the shape of each saturation curve, which is exactly what proxy metrics cannot give you.
Marketing mix modelling estimates each channel’s contribution to sales statistically, accounting for seasonality, promotions, competitor activity and the rest of the messy reality that makes measurement hard. Marketing mix modelling solutions do this without user-level tracking, so they keep working as privacy tightens and reach channels attribution cannot, such as TV, radio, out-of-home and sponsorship. Research on 212 large firms has linked greater use of marketing analytics to higher return on assets, which is the payoff of allocating on contribution rather than proxies.
Using modelling for budget decisions is well established in enterprise marketing because it gives a holistic view of performance that attribution cannot. Objective Platform combines Bayesian modelling with scenario planning, so you can test a change such as shifting €200,000 from search to video and see the projected impact before committing the money.
Budget has to work across awareness, consideration and conversion, yet most plans overspend at the bottom and starve the top, because conversion is easy to measure and awareness is not. Bottom-heavy budgets tend to push acquisition cost up as the easy wins run out, leave you invisible early in the journey and cede ground to competitors with a more balanced funnel. The right balance depends on category, market maturity and growth stage, but very few brands do well on performance alone.
There is no universal brand-to-performance ratio. The right mix varies by industry, competitive intensity and deal-cycle length, which is why testing your own mix matters more than any rule of thumb. Balancing brand and performance depends on measurement that captures both the immediate response and the slower-building brand effect.
A practical way to test the current mix is an incrementality check: pull spend back in one channel and measure the total business impact. A geo experiment or user-level holdout can show whether attributed conversions materially exceed the conversions genuinely caused by the activity. Experiments in marketing measurement set out how to design these tests.
Short-term return and long-term brand contribution both matter. Paid Search might show a 800% platform-reported return while an incrementality test finds that a meaningful share of those conversions would have happened anyway. TV might show a lower immediate return while also contributing to brand effects that influence future demand. A sound spend level has to weigh both time horizons.
A spend plan cannot be static: conditions shift, competitors launch, headwinds appear. Build in quarterly reviews and reallocation triggers so that a channel underperforming for two months can be adjusted without waiting for the next planning cycle. Reserving part of the budget for rapid reallocation gives you room to act on an opportunity or defend against a competitor. The annual planning challenges most leaders face, proving ROI to a sceptical CFO and forecasting in a volatile market, get easier when you can model scenarios rather than argue from instinct.
When budgets are cut, the common response is an even reduction across every channel, which feels fair but erodes the efficiency of the mix. Smarter cuts rank channels by marginal efficiency and take from the bottom, sometimes removing a weak channel entirely rather than weakening all of them. Maintaining share of voice through a downturn tends to pay off later, so it is usually worth protecting brand investment rather than slashing it, a point reinforced by building a budget plan that holds up all year.
Allocation is not a one-off decision but a continuous process, closer to managing a portfolio than setting an annual plan. Regular scenario planning, modelling different allocations and comparing projected outcomes, builds institutional knowledge about how the mix behaves under different strategies.
Predictive media planning turns allocation from an annual ritual into an ongoing loop: predict performance, adjust the mix, measure results and feed those results back into better predictions.
AI and reinforcement learning for budget decisions
AI is increasingly part of how allocation gets done. Reinforcement learning and similar methods can learn from historical data and weigh far more variables than a spreadsheet, surfacing allocations a manual process would miss, and AI assistants are starting to let marketers explore those scenarios in plain language rather than wrestling with dashboards. What makes this genuinely useful is pairing it with transparency: when you can see why a model recommends a shift, you can act on it with confidence and explain it to the board. The strongest setups combine that automated optimisation with human judgement on the things a model cannot see, competitive positioning, brand strategy and upcoming launches.
Allocation decisions only hold if you can show they work. Finance wants business outcomes rather than marketing metrics. That means translating channel performance into revenue: showing, for instance, that a €500,000 investment in brand building is expected to generate around €2 million in incremental revenue over two years, with a stated confidence range. Gaining full control of your marketing budget rests on this shift from marketing metrics to business outcomes.
With budgets under pressure (Gartner found 59% of CMOs say theirs is too small to deliver their strategy), a business case built on investment, expected contribution, time horizon and a sensitivity range changes the conversation. You stop defending marketing and start discussing investment returns. A simple monthly view, spend against budget, actual against projected return and saturation by channel, keeps that visible and builds the credibility you draw on at the next planning round.
The measurement backdrop keeps shifting. Third-party cookies have not disappeared altogether, Google has kept them in Chrome, but relying on user-level tracking as the backbone of measurement has become less attractive, so the methods that set spend levels need to work without it. Preparing for measurement in a privacy-first world means leaning on approaches that do not depend on following individuals: marketing mix modelling, incrementality testing and brand studies all keep working as tracking fades. Investing in those now, rather than when attribution finally breaks, is what keeps spend decisions grounded.
Deciding how much to spend comes down to measuring true contribution, understanding where each channel saturates and testing changes before committing to them. Objective Platform brings marketing mix modelling and scenario planning together, so you can estimate channel contribution, test budget scenarios and set spend levels on evidence rather than on last year’s number.
How much should you spend on marketing?
There is no fixed percentage that fits every business. The useful way to frame it is marginal return: keep investing in a channel only until the next euro returns about the same as the next euro elsewhere. That requires estimating each channel’s true contribution and the shape of its saturation curve, rather than copying an industry benchmark.
How do you decide marketing budget allocation?
Allocate toward marginal return. Estimate each channel’s incremental contribution with marketing mix modelling, compare saturation curves and move budget from channels that have saturated toward those with higher marginal return. Test the shifts with experiments before committing, then revisit the allocation regularly rather than once a year.
What are diminishing returns in marketing spend?
Diminishing returns describe how each additional unit of spend in a channel tends to add less than the one before it. Early spend captures the most responsive demand; beyond a point the channel saturates and extra spend barely moves outcomes. Knowing where that point sits for each channel is what tells you when to stop adding budget.
How do you prove marketing ROI to finance?
Translate spend into expected revenue contribution with a confidence range and a time horizon, using measurement finance trusts, typically marketing mix modelling validated with experiments. Framing spend as an investment with projected returns and a sensitivity range turns a budget defence into an investment discussion.
How much of the budget should go to brand versus performance?
There is no universal ratio. The right balance depends on category, competitive intensity and deal-cycle length; it also shifts over time. Rather than adopting a rule of thumb, test your own mix: estimate long-term brand effects with modelling and run incrementality checks on performance channels to see what each is genuinely adding.
