I feel like this fails to consider my own valuing of my time.
Free Chocolate? Sure.
13¢ chocolate? I've gotta try to make change? An awkward amount no less. 3 pennies? They are getting hard to come by. I didn't even want a chocolate. I don't have any cash on me. Do you take card?
For instance, when I'm buying something off Facebook marketplace, if the items not a multiple of $20 bills and $50 bills, the denominations I can get from the ATM, I'm far less likely to buy it because I have to stop somewhere else on my way to the seller and try and make change. It's a pain in the butt.
I have literally overpaid for things from marketplace by a dollar or two to avoid making change.
But if my only options are 1¢ chocolate versus 13¢ chocolate, those are on way closer footing because either way I have to dig my wallet out.
I'd still take the Hershey kiss though because it tastes better.
Sorry I'm from Sweden and our banks have a service called Swish where we can send money on the phone. Paying in cash is extremely uncommon now a days. Every time I've bought or sold something on FB Marketplace the last decade I've used Swish. I thought you had something similar called Venmo in the US?
Before people read too much into this, just be aware that Ariely has been caught out a number of times falsifying data or even making up entire studies that he then writes about in publications. Not saying this was the case here, but you'd want to look for corroboration in sources not involving him. TFA mentions a few, but most of it is from Ariely.
I'll never understand the people who stand in line for an hour for "Free donut day" or something similar. You really value a $1.50 donut equal to an hour of your time?
I was hoping this would talk about the hordes of ungrateful users demanding more and more free labor from the unpaid volunteers of open source projects, but I guess we still don't know how to deal with that properly.
I thought the article would be about something like when you get $100 free chip, you are much more likely to gamble and lose it; or when someone win a lottery, they would quickly spent the money compared to if they had earned the money with hard work.
BTW, behavior economics people like DAN ARIELY in this article got bad reputation after being found fabricating data on the research about honesty
https://www.npr.org/transcripts/1190568472
Behavioral economics have repeatedly showed that humans are consistently irrational when it comes to buying and selling stuff.
Modeling humans as rational agents simplifies the economic reasoning and the equation a lot so it's not entirely worthless, but we must always keep in mind that this model is very far from the reality even if it's sometimes useful.
The Lindt packaging is ugly as sin, while the Kiss looks sleek and cool.
And unless you're constantly jingling around with a bunch of coins, using a common dollar means you're going to be jingling around with 87 cents of dirty coins, after pulling out your wallet.
The study they cite seems to be leaving out something: are the participantsforced to make a choice, or could they choose to not take either? If I were presented with the two choices they give, I'd probably take the free one in the first choice but not take either in the second because I just wouldn't care enough to buy the single small piece of chocolate for either price. If I were forced to make a choice, I might pick the Lindt, but I'd argue that then their experiment isn't actually testing the same thing. A forced choice been two things isn't the same as two options that can both be rejected.
For anyone who thinks there is flawed logic in this I encourage you to study JC Penney's pricing strategy failure. People, by and large, are not rational.
One interesting angle here is how “free” changes not just user behavior, but also how builders interpret demand.
In AI products especially, it's very easy to mistake “engagement” for “real demand” — because when things are free, people try everything. You get signals, but many of them are noisy or even misleading.
I’ve been thinking about this a lot in the context of marketing tools: instead of optimizing for more exposure or more content, maybe the harder problem is filtering out false positives — figuring out where genuine demand actually exists.
Otherwise, we might just be scaling irrational behavior on both sides: users consuming free stuff, and builders chasing the wrong signals.
60 comments
Free Chocolate? Sure.
13¢ chocolate? I've gotta try to make change? An awkward amount no less. 3 pennies? They are getting hard to come by. I didn't even want a chocolate. I don't have any cash on me. Do you take card?
For instance, when I'm buying something off Facebook marketplace, if the items not a multiple of $20 bills and $50 bills, the denominations I can get from the ATM, I'm far less likely to buy it because I have to stop somewhere else on my way to the seller and try and make change. It's a pain in the butt.
I have literally overpaid for things from marketplace by a dollar or two to avoid making change.
But if my only options are 1¢ chocolate versus 13¢ chocolate, those are on way closer footing because either way I have to dig my wallet out.
I'd still take the Hershey kiss though because it tastes better.
> the denominations I can get from the ATM
Is this for real? We can request 5s and 10s from the atm, along with 20s and 50s.
BTW, behavior economics people like DAN ARIELY in this article got bad reputation after being found fabricating data on the research about honesty https://www.npr.org/transcripts/1190568472
Modeling humans as rational agents simplifies the economic reasoning and the equation a lot so it's not entirely worthless, but we must always keep in mind that this model is very far from the reality even if it's sometimes useful.
The Lindt packaging is ugly as sin, while the Kiss looks sleek and cool.
And unless you're constantly jingling around with a bunch of coins, using a common dollar means you're going to be jingling around with 87 cents of dirty coins, after pulling out your wallet.
Fail to consider the transaction cost of paying the 13 cents for the Lindt, compared to the free Hersheys.
Plus Lindt sucks.
People give it me all the time as gifts. I give then give it away to random people like couriers.
Godiva on the other hand...
In AI products especially, it's very easy to mistake “engagement” for “real demand” — because when things are free, people try everything. You get signals, but many of them are noisy or even misleading.
I’ve been thinking about this a lot in the context of marketing tools: instead of optimizing for more exposure or more content, maybe the harder problem is filtering out false positives — figuring out where genuine demand actually exists.
Otherwise, we might just be scaling irrational behavior on both sides: users consuming free stuff, and builders chasing the wrong signals.