Why you should never ask customers whether they think they will buy your new product

Written by Dr Amantha Imber

All around the world today, thousands of companies are running focus groups to ask customers whether they will buy the new product that they have just designed. And these focus groups all look very similar.

There will be a moderator, employed by the company or their market research agency, and they will be sitting on a table with seven or eight potential customers. They hold up a mocked up picture of the new product, perhaps a new chocolate bar. The moderator describes the new bar in generous (and delicious) terms.

The moderator then asks what they think is their killer question: "Imagine you saw this chocolate bar on the shelf of your local supermarket. How likely do you think you would be to purchase it?"

Each person thinks to themselves, "I like chocolate. And this bar sounded tasty." They say out loud, "Yes, I would be very likely to buy it".

Executives in the company rejoice after hearing the results of the research: Eight out of 10 people said they would definitely buy this new bar.

Six months later, when the bar finally makes its way to supermarket shelves, sales are dismal. Executives ask what has happened. "Didn't 80% of customers say they would buy this bar?"

The marketing team shrug their shoulders and say yes, that is what customers told us.

Nine months later, the bar is pulled from shelves due to lack lustre sales and the product is labeled as a failure.

The reason focus groups and other research methodologies that involve asking customers what they think are so poor at predicting the future is because people are bad at predicting how they will behave in the future. Research into the relationship between our intentions versus actual behaviour shows that there is only a weak correlation between what people say (i.e. their intentions), and how they will actually behave in the future.

In order to more accurately predict the success of new products and services, organisations need to do the following.

Put a ban on focus group testing for new product development

While focus groups can be useful for understanding how people say they feel about something, they are notoriously poor at predicting how people will behave in the future. Save focus groups for research projects that are aimed at understanding the present, such as exploring people's brand perceptions, rather than asking them to be a clairvoyant about their own behaviour.

Stop asking questions beginning with "How likely would you be€¦"

Outside of focus group methodology, researchers need to stop asking questions that begin with the words "How likely would you be€¦". This question is only a measure of how someone intends to behave in the future, as opposed to whether they actually will behave this way in the future.

The only thing this question generates is false hope. People often have the best intentions, but this often doesn't play out in real life.

Run experiments to test actual behaviour

When a good idea is born that needs to be tested, run an experiment to test actual behaviour. Experimentation involves setting hypotheses as to why an idea will add value to the customer and creating a minimum viable product (MVP), which is the most basic version of the idea that will still allow for learnings. Experiments can then be set up to test hypotheses using the MVP. Because experiments are designed to measure cause and effect, good experiments measure behaviour, not intentions. And then based on the results, organisations can iterate or change course accordingly.

Because experimentation measures actual behaviour, it is a very effective way to de-risk innovation. Companies from Google, through to Lendlease, through to Commonwealth Bank all embrace experimentation as a key stage in their innovation process.

Dr Amantha Imber is the Founder of Inventium (www.inventium.com.au), the innovation consultancy that assesses and compiles the AFR Most Innovative Companies list. Enter this year's list at www.mostinnovative.com.au