Chances are you’re calculating ROI wrong, and you probably draw the wrong conclusions about the number derived (presuming you calculate it correctly). How ROI is evaluated needs to be based upon a serious calculation of ROI as well as then a critical understanding of the context of ROI in your business environment. I’ll explain both concepts, but let’s start first with calculating ROI. Tomorrow I’ll talk about context.
First, there are lots of ways to calculate ROI – 99% of them are absolute utter garbage. The most common that I’ve seen, and the one that most CMOs, brand leaders, and marketers, swear by, this tried and true type of formula: (revenue generated – ad spend)/(adspend). In other words, what they’re measuring is incremental revenue (some use incremental profit – but the same problems exist) as a function of advertising expenditures.
Why is this calculation bad? Well for one thing – what it the heck does it actually measure? What people THINK it measures is the amount of additional revenue a company receives as a result of its advertising activity. That’s absolutely an erroneous conclusion, in part because it’s based upon an erroneous assumption: without advertising, I don’t get the additional revenue. That assumption isn’t reasonable in most circumstances.
In most business cases, it’s folly to say that advertising is responsible 100% for revenues in a given period. Now, in my podcast with Scott Monty at Ford, he actually outlines how in the case of the Ford Festiva movement, it was easy to measure ROI because there was no other campaign activity going on – thus, all the benefits of the campaign were either in part or in whole as a result of the campaign’s activity. In the case of most businesses, it’s not true. Some revenue always comes from the advertising, some revenue comes from what was done in the past, and some revenue comes from just bumping into the marketplace of customers.
For example, I used to believe that McDonald’s could stop its advertising for three months and see no visible change in activity or sales. I mean, is there anybody on the planet that doesn’t know about McDonalds? Of course not. However, what I’ve been told by people who work that account, and who have worked that account, is that if McDonalds was indeed to stop – they would see an immediate plummet in sales. They know exactly why as well – if they stopped reminding people to buy McDonald’s, people would slow down their consumption. How does one measure ROI in that real world circumstance (the opportunity cost of not advertising)? Most brands haven’t a clue as to what pushes them forward, and what holds them back, so the ROI calculation is hindered as a result. The entire purpose of calculating ROI in 99% of all cases is to make two conclusions: 1) what accounts for most of our revenue (causal linkages), and 2) what is the efficiency of the activity (ostensibly to focus on optimal allocation).
Here’s another anecdotal problem – brand value. My brand value actually makes my advertising more effective in lots of ways, thus how do I understand the long term ROI on brand investment? What is the “output” variable to measure (which is in most ROI calculations, sales or profits). This is why most brands poo-poo the idea of “brand valuation” and see branding as “something we do, but it doesn’t really result in much.” However, for a brand like McDonald’s, continuing to reinforce the brand directly ties to floor traffic. Moreover, in the case of mega brands, like Coca-Cola, almost 100% of its market capitalization is brand value. Then of course, there is the situation where Brand is more valuable than the company – such a case being Pan Am or Pets.com. How is the ROI determined in figuring out the valuation and added-value of branding activities?
As a consequence of all these challenges, CMOs don’t have a good measure for understanding what is doing most of the heavy lifting in their marketing/advertising budgets. Hence John Wanamaker’s famous quip, “Half the money I spend on advertising is wasted; the trouble is I don’t know which half.” The other real reality that Wanamaker points to is – the way we calculate ROI obscures the reality of what actually produces revenue. Thus, we all know we get 80% of our results from 20% of our activities, but again, we have no idea what the 20% are, so we continue to do 100% of our activities, knowing that 80% of them do absolutely nothing (or nearly nothing).
Calculating ROI : Ok, how should I do it?
If you really want to calculate ROI this way – then there is only one way to do it properly (where you’ll really know the marginal revenue or profit derived from each dollar of advertising), devise a function that allocates marginal revenue and advertising activities in such a way that it holds all other activities constant (for those of you who are calculus inclined, what I’m talking about here is the first derivative of profit against whatever advertising variable one wants to use). If you in fact did that formula – you would really know what your ROI would be. You’d also be able to calculate the risk factors associated with the ROI and make assessments as to whether or not the activity was really worth the capital at risk.
Chances are, you wouldn’t really like the answer you received. In most cases, you’ll find your ROI is either break even or negative. You’ll find that most of your ROI’s are under 10%. That’s because advertising, on the margin, does little in the majority of cases to effect direct sales. If that doesn’t pass the giggle test for you, consider this fact. Advertising has gone up every year, world wide, for a decade. Has the industrial output (measured in GDP) of the world grown on pace with that of advertising? (Hint: It hasn’t. Here’s the general growth of the World’s GDP. Here’s the general growth of advertising spends.) If I just look at the amount of money being spent, if anything, it looks like the ROI overall is dropping on all marketing activities. At the macro level, at least, that would suggest that most advertising is an absolute waste. That doesn’t stop CMOs from spending, however, as they try to acquire new customers.
Rather than go the calculus intensive route, there is an easier route I think CMOs can use. It allows for both qualitative and quantitative measure, and the results also give insights as to what activities generate the most output. It also allows the rest of management (CEO, CIO, Sales, etc.) to understand the impact advertising and marketing efforts have on the overall revenue (and profit) generation of the company.
My basic formula for CMOs looks like this:
ROI of advertising/marketing is a function of:
- the costs of attracting more people (either virtual or in reality),
- efficiency in which it creates more leads (or opportunities for a sale)
- the cost of obtaining more sales relative to the activity (efficiency of conversion)
I’ll discuss tomorrow why those three areas may be the most insightful for CMOs in understanding “where am I really getting my results from” and how to apply it to advertising/marketing allocations.