Crowning the Ultimate Halloween Candy with MaxDiff
It’s that time of year when some of us decide to partake in the (over) consumption of candy and chocolate - all in the spirit of Halloween of course!
And many have taken a stab at crowning the ultimate Halloween candy. A recent poll from Monmouth University named Reese’s Peanut Butter Cups as the Halloween favorite. And in 2017, @fivethirtyeight posted an Ultimate Halloween Power Rankings, agreeing that Peanut Butter Cups reign over 85 other chocolates and candies.
But there is a better way to rank these candies than just asking about “favorites”, like in the Monmouth poll, or comparing two candies at a time, like in the @fivethirtyeight experiment.
MaxDiff is an approach for measuring consumer preference for a list of items. Items could include messages, benefits, images, product names, claims, brands, features, packaging options, and more! In this case – we have 80+ different options a trick-or-treater could receive in their Halloween bag.
In a MaxDiff exercise, respondents are typically shown 2-6 items at a time, and asked to indicate which is best and which is worst, or most appealing/least appealing, most likely to purchase/least likely to purchase, etc.
This task is repeated many times, showing a different set of items in each task.
What is so great about MaxDiff is that, from just one task, we can conclude from this respondent’s answers that...
Reese’s Peanut Butter Cup > Tootsie Pop
Reese’s Peanut Butter Cup > Jawbusters
Reese’s Peanut Butter Cup > Almond Joy
Reese’s Peanut Butter Cup > Milky Way Midnight
Jawbusters > Tootsie Pop
Almond Joy > Tootsie Pop
Milky Way Midnight > Tootsie Pop
We just don’t know…
Jawbusters ??? Almond Joy
Jawbusters ??? Milky Way Midnight
Almond Joy ??? Milky Way Midnight
Tip #1 - Ask respondents about 3 to 5 items at a time in a Best-Worst experiment to capture more information quicker than asking about just 2 items.
Tip #2 - Try and show each respondent each candy at least once. If your list of items is small, try and show each candy 3 times to each respondent. Balancing the length of the exercise with as much information to inform your model as you can get!
Next, we can use the power of math, hierarchical Bayesian regression to be exact, to build a model that allows us to predict how each individual would choose in every candy match up, regardless of if they actually saw that match up in their experiment or not!
The resulting model offers a score for every item for every individual. Better yet, the results are ratio-scaled so that a score of a 10 means that item was twice as preferred as a score of a 5!
And, since we have individual-level data, we can look and see if there are specific clusters of people interested in one candy versus another (i.e. chocolate vs. sour). We can even run a TURF (Total Unduplicated Reach and Frequency) Analysis on the MaxDiff scores, and find out that we should actually buy Reese’s Peanut Butter Cups, Twix, and Sour Patch Kids to appeal to the majority of Trick-or-Treaters!
By using MaxDiff, your house will be remembered as the one that made everyone happy on Halloween Night!
(Unless of course one of your neighbors is offering King Size candy bars...)
Want to take our survey and help us crown 2019’s Ultimate Halloween Candy using MaxDiff? Click here and feel free to share it with anyone you know.
Want a sneak peak into last year’s results? Check out my article on @SawtoothSoftware here.