Elyse Foley.
Elyse Foley, PHD National Head of Planning and Brendan Hewitt, PHD Sydney Head of Strategy
Birth, death, taxes…and the reflexive urge of media professionals to stress-test established doctrines. Some things in life are certain.
PHD’s National Head of Planning Elyse Foley and Sydney Head of Strategy Brendan Hewitt have decided to debate the enduring relevance of a couple of marketing and media fundamentals. As markets, technology, and consumers shift rapidly, do the old rules still hold? Let’s start with the fundamental that brand growth is broadly predictable based on investment.
To keep things honest: Elyse is arguing in favour of their continued relevance, while Brendan attempts to challenge them—despite both being, in principle, believers.
Elyse takes the affirmative: Brand growth is broadly predictable based on investment
Determining the optimal investment to achieve business objectives has always been marketing’s holy grail. Historically, we’ve approached this in a few different ways:
- Task-led budgeting, where reach and frequency targets are used to set media weight and define budgets. The issue? Most brands have “sufficiency” guidelines, but there’s no magic number for reach that guarantees brand or business success.
- Advertising-to-sales ratio, which divides total ad budget by net sales. This often serves as a category benchmark. The problem is these ratios don’t account for effectiveness—they rely on historical averages, not impact.
- SOV/SOM, a more robust but still flawed approach. It compares your share of voice (SOV) to your share of market (SOM), aiming for excess SOV to drive growth. Yet again, correlation is not causation.
These tools are useful and familiar. But there’s a better way.
Econometric modelling (or MMM) provides the predictive power to estimate returns on advertising and, crucially, the marginal return on incremental spend—helping marketers judge whether that extra 25% investment is worth it.
It’s fair to say advertising-driven revenue shouldn’t be viewed in isolation. But today’s modelling capabilities are exceptionally strong. At PHD, we factor in priors and forecasted variables like category growth, brand growth, media inflation, current SOV and SOM. These shape scenarios that guide smarter decisions. We use techniques like adstock and diminishing returns curves to reflect real-world media impact over time—and incorporate Bayesian modelling to integrate uncertainty and update models dynamically as new data comes in.
Even without a bespoke model, investment planning tools can play a positive role here. With thousands of MMMs analysed across categories, we can forecast likely outcomes using just a few inputs every marketer has. With the right tools and support, brand growth becomes not only predictable—but accessible.
Brendan on the counterpoint: Marketing isn’t linear, and humans aren’t predictable
Overall, the issue is not the principle itself, but how it’s often interpreted. The type that views marketing like a vending machine: insert coins, receive growth.
But marketing isn’t linear or deterministic. It’s a complex adaptive system, where outcomes emerge from the dynamic interaction of many moving parts. Growth occurs when these align and not simply when money is spent.
Linear systems assume predictability. They are context-independent and static. That’s not how media, or the world, works.
We all understand that audience attention is fragmented across platforms and formats, with shared tastes and cultural consensus increasingly elusive. Its algorithmic reality, and not solely the schedule, that determines impact.
The conditions under which advertising is received, shared, and acted upon are increasingly volatile and context-dependent. In complex systems, small inputs can have outsized effects, and big budgets can fall flat. More conditions, less control.
Elyse’s point on Econometric modelling is well taken. These are powerful tools and vital for many categories. However, it’s worth noting a key limitation - the impact of creativity. That might sound a bit abstract, but it is grounded in evidence. We routinely see campaigns with similar spend and media mix deliver wildly different ROI, driven largely by creative choices and the context in which advertising appears. These factors shape responsiveness and lead to significant swings in effectiveness.
Spending isn’t the same as effectiveness. Creativity is one lever, and so are cultural timing, distinctiveness, relevance, and community influence.
Yes, investment matters. It can create the conditions for growth—particularly for established brands. But we should be careful not to mistake the map for the (increasingly complex) terrain.
Conclusion
Econometric models do offer invaluable, data-driven insights that help us optimise investment and predict outcomes, but as marketers we should always be ready to embrace the often chaotic nature of audience behaviours and staying in sync with cultural trends. By balancing both sophisticated analytical tools and a deep understanding of human nature, brands can unlock that transformative and enduring growth we’re all after.
\
Brendan Hewitt