Les Binet and Sarah Carter get a little bit angry about some of the nonsense they hear around them… like the idea that we learn best from success.

Recently, we found ourselves at the IPA 'Eff Fest', a celebration of all things related to advertising effectiveness. Professor Paddy Barwise from London Business School in one session showed a collection of social media case studies. The cases were interesting he said, but would be even more interesting if they were about failures. Because we learn so much more from failure than from success.

Much as we value effectiveness case studies, we agree. And in that spirit, let us offer a case study of our own.

Once upon a time, a group of clever agency people had a great new idea for promoting an FMCG brand. It essentially was a price promotion, but was delivered in a cool, new digital way. The client loved the idea too, and pretty soon, software developers were hacking through the night to get the thing ready for the brand's peak season.

But just before launch, a kill-joy consultant heard about the project. Don't do it, they said. It's going to lose money. But no-one believed them and it was too late to stop anyway. The project was indeed a failure. Sales volumes were tiny and were so heavily discounted that they were deeply unprofitable. The client vowed never to do anything like it again.

So what can we learn from this story?

Firstly, beware of fads and fashions. Clever people chose a flawed technological solution, less because it was right, more because it was 'cool'. Secondly, beware 'groupthink'. Everyone involved in our example loved the new promotional idea. Nobody sought an outside view. The consultant who correctly predicted failure only heard about it by accident.

Remember thirdly, it's not your money. The team thought about their problem in a very 'Silicon Valley' way: create some cool techy stuff, get it out there quickly, and let the money take care of itself. That may be fine if you're a Google or a start-up with patient investors. But it rightly doesn't play well with established FMCG company boards. Fourthly, beware hasty decisions. Agencies always have short deadlines, but by the time it was clear the project would fail, it was too late to stop it.

But the biggest lesson of all is to learn from our mistakes. Let's hope internally that has happened in this case. But the project is still trumpeted by many as a great example of creativity.

The tendency to sweep failure under the carpet is pretty universal in our industry. Occasionally, we do get hints of failure. Why did Heinz go back to TV advertising after very publicly putting all its money into DM back in the 1990s? Why did Pepsi row back from its decision to make social media the core of its marketing? But public admissions are rare, and case studies non-existent.

Some industries are better at this than we are. Military failures tend to be very obvious and very public, as do failures of government. As a result, these are two areas where failure has been studied in depth. A recent book, The Blunders of Our Governments (King and Crewe) categorised factors causing governments to stumble, including groupthink (a concept arising from the study of the US failure in Vietnam), intellectual fads, hasty decisions and a failure to learn from past mistakes. Our little case study exhibits all of these.

But there are some examples in our world of enlightened companies brave enough to embrace failure. A G Lafley at P&G, for example, introduced a President's 'Fail Forward Award' for teams or individuals that enable the organisation to 'significantly learn from a failure'.

So maybe it's time for agencies too to man up and start admitting to some mistakes. When the IPA holds its inaugural IPA 'Ineff Fest', we'll be in the front row.