Impact report Adding a game element encourages participants to answer in more detail generating richer data embarrassed by their own behaviour. This means being more empathetic when designing surveys and improving communication with those taking part. He cites the example of a project his team did with Mintel to evaluate consumer attitudes and brand awareness of new products. Mintel identifies and catalogues around 18,000 new food and drink products in the United States each year for its global new products database. The team decided to add consumer evaluation research, and wanted to measure reactions to each of these new products as they were released, and give their global client base live access to the insight. Mintel initially ran the research in the US before expanding it into Australia. The company is currently using it to evaluate more than 2,000 products a month across the two markets. Consumer testing 18,000 products a year was an ambitious idea; it would mean fielding 400 surveys a week. If 100 people were to evaluate each product, it would require two million completes a year an impossibly large number. The Kantar QuestionArts team had to find a way to conduct the evaluations more efficiently, and design a survey that participants, once they had completed it, would want to do again. Respondents were struggling to concentrate because the work involved evaluating thousands of products, taking photographs, and completing a lengthy survey. The first task was to slash the time taken to complete a typical new product evaluation survey from more than 10 minutes to just two. All the remaining detailed questions normally found in a new product development survey were replaced with one simple, open-ended question, asking people to explain what they like or dislike about each product. A text analytics protocol was developed to convert this feedback into closed metrics, by analysing more than 150,000 of the comments, isolating the recurring words and phrases, and classifying them into categories. Kantar also came up with a gamification element, with participants winning points by guessing if they felt a product would be successful or not, and if their view matched that of the market. The survey was also tightened up and made less repetitive, with the number of questions relating to how someone evaluated a brand cut significantly to reduce boredom levels. In head-to-head tests, Kantar discovered that people completing the redesigned, gamified version of the survey evaluated 25% more products voluntarily. The rating they gave to the survey increased from 7.8 to 8.9 (out of 10), putting it in the 22 Impact ISSUE 36 2022_pp18-27_Report.indd 22 08/12/2021 09:59