The marketing team at Brand A delivers a campaign with an ROI of 1.6x. The team at Brand B return 2.0x. Which company has the happiest CFO?
This scenario I teach at adchool, highlights the fundamental problem with ROI.
The answer is, “we need more information”. Because these ratios, by themselves, are largely useless.
In the tutorial, I then reveal Brand A invested $5 million, delivered $20 million in incremental revenue on a margin of 40%. Their profit is $3 million. Whereas Brand B invested $2 million, delivered $10 million in incremental revenue, with the same margin. They have a higher ROI, but their profit is only $2 million.
The happier CFO is the one with the higher profit. Because a higher ROI does not necessarily mean more effective.
A measure of efficiency, not effectiveness
Return on investment is a measure of efficiency. That is, an indicator of how well resources were allocated. It is not a measure of the net result. Which is why ROI is broken.
You may choose to pursue a lower ROI because it’s still delivering incremental profit. A campaign may hit diminishing returns, and yet it’s still generating positive returns. Efficiency is sacrificed, but the net effect continues to grow.
A team, or a department, or a company, optimised to improve ROI, will seek to maximise efficiency, rather than effectiveness. And the fastest way to improve your ROI is often by reducing the ‘I’.
In his 2004 paper, 'ROI is dead, now bury it',Tim Ambler shows how a cut in marketing expenditure can mostly maintain sales, thus improving ROI but actually creates a significant decline in net profit. Not good. And that doesn’t factor in the long-term benefit of building future demand, not captured in this year’s financials.
He also playfully suggests if you cut all marketing investment and still achieve some sales you can achieve infinite ROI. Les Binet says “The most efficient way to run a business is not to have one. Zero revenues, zero employees, infinite efficiency.”
Waste isn’t a bad thing
I’ve long been a fan of Ambler, who I first came across in a different 2004 paper 'The waste in advertising is the part that works.' Here he introduced (to me at least) the concept of signalling theory.
Borrowing from biology, Ambler shows how animals use wasteful characteristics to signal their superiority when attracting a mate, or fending off a predator. And brands work the same way - extravagant displays of expense signal fitness and credibility, strengthening effectiveness.
The inefficiency of a Superbowl ad, is where its value lies. Which would be penalised if you were only optimising to ROI.
Not just capturing existing category demand
You advertise for many reasons. Yes, sometimes it’s to drive an immediate short-term response, capturing people who are ‘in market’. But we know most people are not ‘in market’ and brands can deliver more effectiveness when they’re thinking longer term, building future demand. ROI tends to miss this.
ROI also overlooks the role of advertising in sustaining sales rather than driving growth. Brands who do not maintain mental availability will leak sales to a competitor who does. Ambler says “We do not paint the warehouse because it has a positive ROI, but because not painting it will cost us more in due course.”
And ROI overlooks advertising’s ability to drive pricing effects. On this Ambler says “Advertising typically influences price more than volume”. He also points out an increase in sales leads to an increases variable costs, but an increase in price lands straight on the bottom line.
If not ROI, then what?
ROI is hard to get away from. Media platforms report on return on advertising spend. Effie entries conclude with that big final number. And it’s often touted in public case studies and keynotes because it’s largely meaningless, unlike a commercially sensitive profit.
Maybe there’s a better way.
The most obvious answer is net profit. The metric our non-marketing friends get excited about. But this continues to short-sighted.
Instead, some have proposed a metric called Discounted Cash Flow. A valuation method that estimates the value of an investment, based on its expected future cash flows. And it’s ‘discounted’ because future cash isn’t quite as good as cash right now. Also predicted cash isn’t guaranteed cash. So you take a little off.
This achieves a few things. It treats advertising as an investment, not an expense. It looks long-term, appreciating the future value of the investment. And it factors in the changing value of money ($1 today is not worth $1 next year).
It’s more complex, sure. And perhaps a bit scary for the non-accountants among us. But maybe borrowing some finance fundamentals gives marketers some much-needed credibility.
One last Ambler quote: “If advertisers seek respect for their proposals, then they should use respectable metrics.”
Apparently, we couldn’t bury ROI in 2004. Maybe it’s time to redig the grave in 2026.
Zac Martin is a planning director at TBWA\Melbourne
