[Blogg] Making sure your advertising really works

Taking a ruthless approach to learning as you go.


  • Lindsay Troy Service Line Leader, Creative Excellence
Get in touch

The authors of Freakonomics released a couple of interesting podcasts recently entitled Does Advertising Actually Work?

The argument that they put forward in the first of the two podcasts was that the majority of advertising doesn’t work. This based predominantly on the learnings from academic Anna Tuchman, an associate professor of marketing at Northwestern University’s Kellogg School of Management , who’s study on the top 500 fmcg advertisers in the US led her to conclude that “The vast majority of brands over-invest in advertising and could increase profits by reducing their advertising spending.”

It’s a pretty radical conclusion and if it’s true, as the hosts of the podcast would have us believe, then it has enormous implications for the industry.

Whilst the method the study used to derive the results seems clever, what is unclear is exactly how they are calculating the return on advertising spend. The indication is that their analysis is focused purely on sales uplifts.  If this is the case it throws doubt on their conclusion since it has long since been proven that some of the key benefits from advertising come not just from sales gains (in the short term) but also from the value it delivers in building or supporting the brand over the longer term, and crucially also, in justifying price premiums.

But it does raise some interesting questions about whether advertising doesn’t work as well on the whole as perhaps it’s assumed that it does. The hosts make a fair point that much of the available evidence for advertising’s effectiveness is based on individual case studies or meta analyses of the published cases available, which by their very nature tend to be cases where the advertising has worked. Those where it doesn’t work as well get put away in a draw never to see the light of day. And they also argue quite compellingly about what they call the Principal-Agent problem - that it is not always in people’s interests to talk about what happens when advertising doesn’t work - agencies, publishers or even to a certain extent buyers within the advertisers themselves have vested interests and their goals may not align directly with the profit maximising goals of the firm.

My view having worked with the evaluation and development of advertising for 20 years is that advertising very definitely can and does work.

Great business gains can be made but it’s not easy to develop a campaign that can create the kind of business gains that winners of the 100-Wattaren or IPA effectiveness awards are able to demonstrate. Ipsos’ experience tells us that for every big winner there are many more that don’t quite deliver the returns that the advertiser’s had hoped for – a critical reason, if ever there was one, for sticking with a winning formula for as long as possible if you find one.

And also a critical reason for finding a way to maximise your chances that you find a winning formula.

But why is it so hard?

Fundamentally because creating advertising that sells, builds brands and can create long lasting returns is not simple. It involves understanding the motivations of your target audience, identifying a powerful strategy, creative magic to convey the strategic idea in a compelling way, identifying the right media to reach your target and much more besides. There are unfortunately many pitfalls, many things that can go wrong – and frequently do – and while the rapidly changing media environment presents us with ever greater opportunities to reach people, it’s also getting demonstrably harder and harder to get our attention.

So, in the absence of published case studies to learn from explaining in detail what went wrong in the multimillion campaign that company X ran last year, what can you do?

The answer lies I believe in taking a ruthless approach to learning as you go. Stephen King, one of the founders of the Account Planning discipline first championed this approach back in the 1970s, explaining through what he called the The Advertising Planning Cycle how creating advertising should be a constant process of learning and development. At Ipsos we still champion this thinking today, helping our clients to learn and develop and optimise their advertising at each stage of advertising development journey. This could be helping to identify insights into a target audience that a campaign idea can be based upon, helping evaluate and optimise campaign concepts or ideas, or at the later stages in the process, evaluating finished advertising. Properly conducted research can help ensure that you learn at every stage of the journey, take the learnings with you and optimise your chances of developing powerful advertising that will drive large business gains.


We also know that the earlier in the cycle you start researching and learning the better. At Ipsos we see a 48% improvement in the performance of creative assets where we have conducted early stage creative development work vs those where we have not. Doing work early ensures your campaign has a solid foundation, increases the odds of success and allows you to save time and money by being more agile further along the cycle.

So, the answer to the question that the authors of Freakonomics posed: ‘Does Advertising Actually Work?’ is, from the perspective of people who spend their working lives evaluating advertising, very definitely ‘Yes, advertising can and does work’. But it is by no means a given, and advertisers would be unwise not to maximise their chances of success by working with a trusted research partner to learn at each step of the process and help maximise the return on advertising spend.



Does Advertising Actually Work? (Part 1: TV) (Ep. 440) - Freakonomics Freakonomics



  • Lindsay Troy Service Line Leader, Creative Excellence