I suspect many of you don’t even know – or care all that much – where our data comes from. But I thought you’d enjoy a little discovery we made “under the hood” this past week.
For a quick overview: We ask people questions inside polls and quizzes they encounter while traversing the web. They answer for fun, to be heard, or to see an interesting result at the end. We don’t compensate respondents – we actually consider that a research sin. So, our entire lifeline depends on asking questions people want to answer. Studying the “engagement rate” of thousands of questions is fascinating meta-research, in and of itself.
What we found last week is that by optimizing the first question of a poll or quiz, using some algorithmic magic I would be fired for telling you about, we were able to increase the volume of our inbound data by over 20% overnight. You’d think the top-performing questions would all be things about Trump, immigration, or other hot-button issues. Certainly, some were. But others were head-scratchers. Here are some of the highest-performing questions in our entire system:
– Is fantasizing about someone cheating?
– Do you know someone who shouldn’t be allowed on the Internet?
– Do you trust Elon Musk?
– What do you call your in-laws? (Yes, this was one of your ideas. Thanks!).
Notice a pattern there? Me neither. People are weird.
Here are some more explicable things we’re seeing right now:
One thing people didn’t care about was the government shutdown. Our Economic Sentiment Index reached another record high in our most recent reading, which encapsulated the brief federal government hiatus that dominated headlines last week. While consumer confidence did fall slightly during the three-day shutdown drama, it rebounded instantly when the standoff ended. Much to do about nothing…
People are increasingly more likely to deal with stress by going to a spa – but eating & drinking are still our favorite mood-boosters. A cool little study we did this week looked at how Americans spend time and money to bolster their moods when feeling down. “Food and Drink” was the most popular choice, followed closely by entertainment like movies or concerts. Self-care options like exercise and spa treatments are less popular but growing, particularly among Millennials. One thing that surprised me: Going “Shopping” as a mood-booster is inversely correlated with income (ie. the wealthier you are, the less likely you are to do it).
Two things we learned about Amazon this week: 1) A tiny fraction of Alexa (or Google Home) owners actually use the devices to make purchases; And 2) Whole Foods’ brand favorability hasn’t budged since joining the Amazon family. Oh, and Amazon’s HQ2 is coming to Pittsburgh, although I have no data whatsoever to support that. It’s just positive thinking.
If like me, you’re not watching YouTube every day, you’re old. In 2017, a staggering 66% of U.S. teens watched YouTube daily and 80% watched at least several times a week. For contrast, a measly 26% of people over the ancient age of 35 watch it every day. Sometimes I feel like I live on a different planet.
A Random (But Not Trivial) Stat of the Week…
Look! People have been losing weight since Trump was elected! Well, not really. It’s much more likely that this chart reveals a shift in people’s perception of what it means to be overweight. Whether due to a societal movement to reject “body shaming” or an overall improvement in consumer confidence and mood, our attitudes about weight are changing. Hell, if the President is in “excellent health,” then so am I.
Pass the bacon.
Hoping you’re well.
JD
