when good ideas fail… and how to mitigate that risk!
I have a question for you: if most everyone tests their innovation through concept testing prior to a launch, and if we assume that every new innovation being launched has excellent concept test scores, then why do so many new product launches in CPG fail? The answer is simple - concept test results do not guarantee in-market success. But why? Mission Field has been diving deep into our history of in-market transactional tests and now we have data that reaffirms our intuitive and experiential knowledge.
Over the past 10 years, Mission Field has conducted hundreds of tests of new innovations in live retail settings. The vast majority of them were great ideas that just needed a little tweaking. Some performed a little worse than expected - meaning they still sold an average amount within their categories,just not as good as hoped. But then there was a third group that clearly stood out because it was a pile of failures… and what was shocking was that each idea had previously scored incredibly well in their concept tests.
To dive deeper beyond our personal intuition we analyzed our deep history of hundreds of in-market tests against their competitive categories. This included sales of the test products against the competitive set in the time they were tested as well as how they impacted their categories vs the control store experiences. Our analysis aggregated millions of data points and required a machine-learning expert from Columbia University to help us decipher it all. And what we learned shocked us!
The results of how all the transactional tests performed against their categories did not play out into a standard bell curve- rather they produced a bi-model curve in shape - meaning that there was a large bell curve of ideas with above-average success, and a smaller bell curve of ideas that struggled to gain traction. And that second, smaller, bell curve of failure was far below the average results of the category… these ideas cratered in their performance against the category norms be it unit sales, $ sales, price-value, time to growth, and more. Digging deeper into the data, we were able to break down all of our testing cases and describe 6 components that helped us describe where ideas worked well and where they went wrong… and recall that every one of the ideas that failed went into their testing experinece having had excellent concept test results.
While we can’t get into the details of each case here - the key takeaway is that we now have new data-driven insights around how amazing ideas can fall short in execution, competitive context and generating consumer interest. We also know a lot more about what can be done to correct those challenges before a launch. Overall our transactional testing model is now showing a 90.4% predictability in terms of identifying a transactional test’s exact level of success and failure, which is something we are now allowing our clients to leverage in their tests of innovation & renovation. If you are planning to launch a new idea that has excellent concept scores, pause and consider the possible impact of a launch that fails… and then call us and we’ll show you how we can help you reduce that risk of failure!