Marketing teams spend a lot of time optimizing performance. What is the right CPA? What is the right ROAS goal? How much should we spend? These questions seem practical. They are measurable. They give teams something concrete to anchor on. But they miss the real decision.
Budgets often start with a fixed amount of expenses to optimize. This is normal in less mature programs. But as more data becomes available, this approach is likely to evolve.
Marketing expenses are one of the primary ways a company deploys its capital. When it’s not treated as a capital allocation decision, teams make choices that don’t align with what that capital is supposed to yield. As spending increases, the trend is fairly consistent:
- Incomes are increasing.
- Efficiency decreases.
- The profit from contributions increases, then eventually decreases.
- Each additional dollar becomes less productive than the last.
None of this is controversial. Most teams have noticed this in their own data, even if they have not formalized it. The disruption occurs in the way these dynamics translate into decisions.
Your customers are searching everywhere. Make sure your brand introduces himself.
The SEO toolkit you know, plus the AI visibility data you need.
Start free trial
Start with
The same data supports multiple “correct” answers
Consider a data example: As expenses increase, performance follows a typical diminishing returns curve. The first few dollars are very effective. Subsequent dollars continue to drive growth, but at a lower rate. This is where I see many teams stopping. But they miss two crucial steps:
- Translated into marginal ROAS.
- Transition to contribution benefit.
Without going beyond expenses, revenue, and ROAS, it’s difficult to convince yourself to take meaningful action from data.
Even if you are tracking ROAS/iROAS, you may still be looking at the data too passively when it should be sparking an active discussion: where should we spend?
The answer is everyone’s favorite: it depends. It depends on what you’re optimizing for. You need to map the impact on contribution profits and what is happening outside of your spending.
If your goal is to maximize your income
You would continue to spend. In theory, you would spend forever.
In practice, no company does this. But from a directional perspective, if the goal is only revenue growth, the answer is always more (in the first chart, the revenue line never goes down).
If your goal is to maximize profit
The answer seems very different. Let’s draw this table in more detail:
In this example, the benefit of contributions peaks at expenses of approximately $6,000 per day. So far, every dollar represents meaningful work for the company:
- Generate additional income.
- Coverage of variable costs.
- Contribute to profits.
Beyond that, income continues to grow, but less efficiently. Eventually, you no longer add to your results. You trade profits for additional income.
If your goal is to maximize revenue without becoming a P&L drag
There is a third option that often falls in between. You could choose to spend up to around $18,000 per day. At this point:
- Income continues to rise
- The profit from the contributions was entirely reinvested
You make a conscious decision to prioritize growth, while avoiding losing money.
Same curve, very different decisions
These are very different answers:
- ~$6,000/day → Maximize profits
- ~$18,000/day → Maximize revenue without becoming a drag on P&L.
- “Spend forever” → Maximize income at all costs (not a real goal, but directional what that implies).
I like to plot yield curves with these extra columns because it forces you to look closer than the typical “earnings flatten” point from a standard curve.
Where Most Teams Go Wrong
Most teams never explicitly make this choice because they never bring this data to the table with financial stakeholders. Instead, they use proxy metrics by default:
- A ROAS objective that seems fair.
- A CPA threshold that has been used historically.
- Performance reported by the platform as a replacement for business impact.
These seem to be the right optimization decisions. In reality, these are implicit choices about where to be on the curve without actually thinking about or measuring the curves.
In some cases, teams underinvest. A high ROAS goal (“to ensure we generate plenty of revenue”) could limit spending to between $3,000 and $4,000 per day, although the company could profitably spend $6,000 or intentionally push to $18,000 depending on its goals.
No one says, “We choose to leave growth on the table” when setting a high ROAS goal, but that’s often the result.
In other cases, teams overinvest without realizing it. They rely on average performance instead of thinking about what’s happening at the margins. So they might say, “We need to maintain ROAS of 2.0 (because we need every dollar to deliver). Keep spending until blended performance reaches that number.”
But at this point, the marginal dollar will likely have a very different effect. If your average ROAS is 2.0, your marginal ROAS could be closer to 1.4. Early dollars are often very profitable, especially in successful digital channels. Later dollars do a lot less work. You may already be past the point where the profit from contributions is maximized.
If the goal is to maximize profits, the right question is not where the average ROAS is, but rather where the marginal return meets your break-even point.
This is a capital allocation decision
This is where marketing and finance need to be aligned. It should be considered like any other capital expenditure, in terms of the trade-offs: Is there a better way to deploy this capital to maximize return?
For many businesses, especially consumer brands, marketing is one of the primary uses of capital. A significant portion of revenue is reinvested in paid media and other growth initiatives.
This makes marketing one of the company’s most important capital allocation levers. Yet it is often managed through isolated platform metrics and optimization decisions, disconnected from how finance thinks about returns, risks and trade-offs.
It is in this gap that the problems appear. Too much focus on efficiency leads to underinvestment, while too much focus on growth quietly erodes profitability. Both are symptoms of the same problem: the decision was never made explicitly.
More often than not, marketing hasn’t gotten a seat at the table because it hasn’t brought the right data to the table. What happens then?
Marketing is given a budget and must spend it efficiently, drive growth, and do it profitably. It becomes a one-way street between finance and marketing, rather than a shared decision based on a clear understanding of what each additional dollar actually brings to the business.
Marketers have the opportunity to change that. But it starts with making the right data available and using it to make better decisions about how that capital will be deployed.
