Crowdsourced studies
/Getting to the bottom of something is really easy if you do the right amount of tests and studies, but these are costly and rarely are completed for all the things you wish they would be for. In comes crowdsourced based studies. Think kickstarter meets double blind placebo to beat the marketing hype and get down to the real data. What price would be needed to run a study, then divide that by what percentage of the product people would be willing to invest in to get the info, maybe 5%, and that gives you the number of the people you need to contribute.
Foods / Supplements
Exercise equipment
Consumer products
A/B tests on tasks with a battery of products
There is a big marketplace of people that do this type of research self funded who go and sell the info in the form of ads or affiliate codes on the backend. While this can work, there is likely a huge amount of sunk money in the research or in the actual products allowing for a wide variance in the profitability or sustainability of the enterprise. Standardize the studies and price properly.