Okay. So before we get into what Daniel actually sent us, I want to close a loop from earlier, because he opened this whole thing by referencing our conversation about grazing at public markets. The dates, the olives, the question of whether it's actually unhygienic or just feels that way.
Which it mostly is. Feels that way.
Mostly. And Daniel's point, and I'm paraphrasing here, is that market grazing is pretty entry-level in the world of ridiculous things consumers try to pull off.
Entry-level. That's generous.
He wrote in to say that Amazon has spent years tightening its return policy after being exploited by people with, and I'm quoting him directly here, "a flexible moral compass around what's a reasonable reason to return a product." And he's clear it isn't just Amazon. It's a global reaction to wild overreaches. People ordering clothing just to try it out. Folks who treat the return window as a legitimate way to furnish a rental temporarily.
That's a real thing. That's a documented thing.
He says it plainly, he doesn't advocate for the generosity of retailers, but he wants to collect the most ridiculous documented examples of this behavior from around the world. Two questions, really. What's the funniest boundary-pushing on the sampling side at markets, and what are the wildest documented cases of return policy abuse?
So a spectrum.
A spectrum. Petty at one end, profound at the other. And I think we start at the shallow end of the pool before we dive into the deep, dark waters of e-commerce fraud.
The shallow end is the market. And the interesting thing about the market is that nobody wrote the rules down. There's no posted policy at a date stall in the Shuk. There's a social contract, and it's entirely enforced by eye contact.
Which is why it works. You take one grape. You don't take the bunch.
You take one grape, you make eye contact with the guy who owns the grapes, and he nods. That nod is the entire terms of service. It's the same assumption of good faith that sits underneath every return policy Amazon has ever written, except Amazon wrote theirs down and then discovered that a small number of people will read a written policy as a challenge rather than an agreement.
There's a line I keep thinking about, from a discussion thread about exactly this. Somebody asked, if someone sets out a big container of candy on Halloween, do you empty the whole container? Do you also take the money out of the tip jar?
The tip jar is the tell. Because the candy is offered. The tip jar is just... sitting there. And the person who takes from the tip jar has decided that the absence of a lock is the same as permission.
And that's the whole episode, isn't it. The difference between a thing being offered and a thing being available.
That's the line. And it's hard to draw in a written policy, because a written policy has to cover every case, so it defaults to being generous. It says yes to everything, because saying no to a legitimate customer costs you more than saying yes to a fraudster. That's the math. Until it isn't.
Until the fraudsters scale.
And then you get a company that built its entire brand on being the most customer-centric company in the world discovering that customer-centricity is a vulnerability you can drive a truck through.
So let's do it. Shallow end first. What's the actual market-grazing behavior, and where's the line?
The honest answer is that most market grazing is completely fine and culturally normal. Free olives at a tapas bar in Spain. A date from a tray in a Middle Eastern market. The Costco sample circuit in the States, which is its own subculture. That's all sanctioned. The vendor is buying your attention with a bite of food, and it's a rational trade.
And the abuse is when the sampling stops being a transaction and becomes a meal.
Right. There's a documented pattern, and it's mostly anecdotal because nobody's running a study on it, of people who will make a full circuit of a market and eat enough free samples to constitute lunch. Which is, honestly, funny. It's petty. It's low-stakes. It's the kind of thing you'd tell a story about at a dinner party.
And it's entry-level precisely because the stakes are so low. The worst outcome is a vendor telling you to move along.
The worst outcome is social embarrassment. Which, for most people, is enough. That's the thing. The market works because the cost of being caught is shame, and shame is a real cost for most people.
For most people.
And the fascinating thing is that when you move from a market stall to an e-commerce platform, you strip out the eye contact, you strip out the shame, and you replace it with a form that has a dropdown menu for "reason for return." And the dropdown menu is where the trouble starts.
Because the dropdown menu has an option that says "no longer needed," and "no longer needed" is a lie that costs nothing to tell.
It costs nothing to tell. And that's the gateway. That's the entry-level drug of return abuse. You buy a shirt, you wear it, you decide you don't want it, and you select "no longer needed" instead of "I wore this to a wedding." And once you've done that once, and nothing happened, the moral floor drops a little.
So walk me through the escalation. Because I want the actual documented cases.
The best one, and I mean this sincerely, the single most elegant example of return abuse I've ever come across, involves Kindle books.
E-books. Okay.
A woman, documented in a first-person account from someone who knew her, spent over two years buying Kindle books and returning every single one of them. Well over a hundred books.
Every one?
Every one. And the mechanism is the beautiful part. She'd buy the book. She'd immediately request a return. And then, and this is the trick, she'd turn off the Kindle's internet connection so the book wasn't remotely deleted while she finished reading it.
Hold on.
Yeah.
Say that back to me, because I want to make sure I understood it.
She returned the book before she read it, and then disabled the connection that would have removed it, so she could read the book she had already returned.
That's not shoplifting. That's time travel.
It's a refund on a thing you haven't consumed yet, followed by consuming it. The return window and the reading window were the same window, and she just... separated them.
How long did this go on?
Over two years. Over a hundred books. And eventually Amazon banned her, but here's the detail that I think is the real story. They banned her from Kindle returns specifically. They did not ban her from the retail side of the account.
They segmented the ban.
So she kept buying physical goods and kept returning somewhere between twenty and fifty percent of them, and according to the account, she even received courtesy credits during that period.
Twenty to fifty percent.
Twenty to fifty percent of everything she bought.
And they gave her store credit.
For the inconvenience.
That is a company so committed to its own brand promise that it will hand a gift card to the person actively robbing it, because the alternative is admitting the promise has limits.
Which is exactly the tension. The brand is "we will make it right." And "we will make it right" doesn't have a clause that says "unless you're the reason it's wrong."
So that's the Kindle case. What else have you got?
Wardrobing. Buy, wear, return. There's a first-person account of a colleague who would buy designer shirts, wear them to company social events, and return them within a week or two marked as unused. Repeatedly. Sometimes to the same salesperson who sold them to him.
The same salesperson.
And when confronted, his defense was, and I quote, "a lot of people do it."
Which is the universal moral defense of the person who has already decided.
It's the appeal to the norm. And the norm might even be partially true, which is what makes it insidious. If ten percent of people do a thing, it feels like a rule rather than an exception.
What about the scale of it? Is wardrobing small?
It's estimated in the billions annually, industry-wide. It's not small. It's one of the reasons clothing retailers started putting those big black plastic tags on formalwear, the ones that are visible if you wear the garment out.
The tag that says "you cannot wear this and return it."
The tag that makes the lie visible. It's a physical enforcement mechanism for a social contract that broke.
Okay, keep going. What's the one that made the news?
There's a documented case from a bookstore owner, reported in the press, of a customer who returned eight hundred dollars worth of books after using them to stage a home for the holidays.
Stage a home.
Staging a home. You know, when you're selling a house and you bring in furniture and props to make it look lived-in. She used eight hundred dollars of books as set dressing, then returned them.
So the books were props.
The books were props. And the store took them back, presumably, because the policy said they could.
Did they look used?
That's the thing. Books used as props on a shelf for a few weeks look exactly like books. There's no wear. There's no tell. It's the perfect crime, except that it's not a crime, it's a policy.
It's a policy. Which is worse, in a way, because you can't arrest someone for it.
You can only ban them. And that's where this whole thing is heading. But before we get to the bans, there's the general pattern, which is the "free rental."
The free rental.
Ordering several similar items with the intention of keeping one and returning the rest. Which is, on its face, reasonable. You want to try three pairs of shoes, you order three pairs of shoes, you keep the one that fits. That's what the policy is for.
But it slides.
It slides when it becomes "I need a drill for a weekend, so I'll order a drill, use it, and return it." Or "I'm moving apartments and I need furniture for two weeks." The return window becomes a rental period with a zero percent rental fee.
And at that point you're not sampling. You're borrowing.
You're borrowing with extra steps. And the retailer is paying the shipping both ways.
So where does it tip from policy abuse into actual fraud? Because there's a line there.
There's a clear line, and it's when the item you return isn't the item you bought. That's the cottage industry. There are people running operations where they'll compromise an account, request a gift-card refund on a high-value item like a laptop, and then buy a replacement shipped to a freight forwarder.
So the refund goes to a gift card, the gift card buys a new laptop, the new laptop leaves the country.
And the original account holder, who may have nothing to do with any of it, is left holding the bag. That's not a person with a flexible moral compass. That's a criminal enterprise. And it's the reason the whole system is tightening, because the platforms can't easily distinguish between the woman returning a hundred Kindle books and the crew laundering laptops through gift cards.
They can't distinguish because both look like a pattern of returns.
Both look like a pattern. And that's the tragedy of the commons here. The egregious fraudster and the enthusiastic wardrober get swept up in the same net, because the net is a machine learning model, and the model doesn't care about intent. It cares about the shape of the data.
These stories are funny. The Kindle thing is funny. But they have a cost, and that cost is now being calculated by retailers, by the environment, and by the rest of us.
It's being calculated, and it's being calculated in a way that's going to affect everyone who never abused anything.
So let's talk about the reckoning. What does the crackdown actually look like?
It started with the bans. Amazon has banned customers who return too many items. This was reported in the Wall Street Journal back in 2018, and it's been happening quietly ever since. The bans can be account-wide. They can be effectively permanent. And there's no clear human appeal.
No human appeal.
You get an email. You're told your account has been closed due to unusual return activity. You can email back. A person may or may not read it. And that's it.
That's a strange thing for a company whose entire brand is customer service.
It's a very strange thing. And the algorithm, according to people who've looked at it, is mostly targeted toward people who are egregious and, more importantly, lose Amazon money. Returning a lot of high-end items, DSLRs, Apple laptops, triggers it fast. Returning a lot of cheap items might not.
So it's not about the number of returns. It's about the dollar value.
It's about the dollar value, and about the ratio. And that's a distinction that matters, because it means the enforcement is economic, not moral. It's not punishing bad behavior. It's punishing unprofitable behavior.
Which is a different thing wearing the same costume.
It's a different thing wearing the same costume. And it gets stranger, because Amazon isn't the only player. There's a whole third-party industry built around this.
Name it.
The Retail Equation. Best Buy and a number of other chains pay this service to score customers' shopping behavior and impose return limits. And the key detail is that these limits override the store's own posted policy.
So the sign on the wall says one thing, and the system behind the register says another.
The sign on the wall says "thirty days, no questions asked." The system behind the register says "this customer has been flagged." And the cashier can't override it. The manager can't override it. It's a secret score, and you don't get to see it.
There's a case I want you to tell, because it's the one that sticks.
A real-estate agent returned three phone cases. Three. And she was banned from returns and exchanges for a year.
Three phone cases.
And she couldn't get a straight answer about why, because the scoring system is proprietary. She was told she'd exceeded a limit. She hadn't been told there was a limit. And the limit was, from any reasonable perspective, absurd.
Three phone cases is not a pattern of abuse. Three phone cases is a Tuesday.
Three phone cases is a person who bought the wrong size and then tried a different one. And the system flagged her, and the store enforced it, and she had no recourse. That's the failure mode of automated enforcement. It's mostly right, and when it's wrong, it's wrong in a way nobody can fix.
So the pendulum is swinging. The generosity that built these companies is being clawed back. And I want to talk about the environmental side, because that's the part that's least funny.
It's the least funny and it's the most concrete. There was a CBC investigation a few years back that found returned online purchases are often sent to landfill rather than resold.
How often?
That's the honest problem with the reporting. The investigation established that it happens, and that it happens at scale, but it didn't put a clean percentage on it. And the reason is that the answer depends entirely on the item. For low-value items, it costs more to inspect and restock than the item is worth.
So they destroy it.
They destroy it. A returned ten-dollar item that costs twelve dollars to process gets thrown away, because the rational move is to eat the loss. And that's not malice. That's arithmetic.
It's arithmetic that produces a landfill.
And it means the environmental cost of a return is often total. The item was made, shipped, returned, and destroyed. Four steps, one outcome.
And who pays for that?
Ultimately, the customer who doesn't abuse the policy. Because the cost of returns gets priced into everything. The reason a shirt costs what it costs includes the cost of the shirts that came back and got thrown away.
So the abuser and the honest shopper are paying the same tax, except one of them earned it.
One of them earned it.
I want to push on something, though, because I don't think the consumer is the only villain here. There's a counterpoint that I think is worth taking seriously.
Go ahead.
Amazon itself resells returned goods as new. There's a well-documented pattern of items arriving that have clearly been opened, used, and resealed. And the company's customer-centric model isn't just generosity. It's a business strategy. It's a moat. It's a way of making competitors look stingy.
That's fair. The generosity is a weapon. It's a way of saying "shop with us, because if you don't like it, we'll take it back, and our competitors won't." And that weapon has a cost, and the cost is the abuse. So the company that built the trap is now complaining about the mice.
That's a little harsh, but I take the point.
It's a little harsh. But the truth is that the retailer is not a victim in the way a small bookstore is a victim. Amazon can absorb the abuse. The bookstore owner who took back eight hundred dollars of staging books feels it. Amazon doesn't feel a hundred Kindle returns. So the moral picture is more complicated than "bad customers, good company."
And there's a knock-on effect here that I think is the real story. The tightening isn't just about money. It's about trust. The whole system ran on an assumption of good faith, and the assumption is being withdrawn.
It's being withdrawn, and it's being replaced with scoring. And scoring is opaque. And opaque scoring is a different kind of relationship. You're no longer a customer. You're a risk profile.
Which brings us to the question of where this all goes. But before we get there, I think somebody here has actually been on the front lines of this.
Hilbert: A Dyson V8, three hundred and forty nine dollars.
Sorry, what?
Hilbert: That was the one that broke me. Not the drone. Everyone remembers the drone. The drone was a Tuesday.
You worked returns.
Hilbert: One holiday season. Big-box electronics. I don't name the store, they don't need the publicity. I was on the returns desk from the week before Thanksgiving through the first week of January. And the drone guy, sure, the drone guy was memorable. He brought back a quadcopter covered in mud with a bent propeller and told me it didn't fly right.
Covered in mud.
Hilbert: And when I pointed at the mud, he said the damage was pre-existing and he was just testing it out. I had to get a manager. The manager approved the return to avoid a scene. That's the phrase. To avoid a scene.
So the policy wasn't enforced.
Hilbert: The policy was never enforced. The policy was a suggestion that got overruled by whoever was loudest. That's how it actually works. The sign says thirty days. The reality says whoever makes the most noise gets the refund.
So the drone guy wasn't the worst.
Hilbert: The drone guy wasn't the worst. The worst were the swappers. The people who buy a new item, take it home, put their broken one in the box, and bring the box back. That's a different category. That's not a person with a flexible moral compass. That's a person running an operation.
And you caught them?
Hilbert: Sometimes. You learn to check the serial number. You learn to feel the weight of the box before you open it. And one time I opened a laptop box and found a brick inside.
A brick.
Hilbert: A brick. Same weight as the laptop, roughly. Wrapped in a towel so it wouldn't rattle. Somebody had bought a laptop, taken the laptop out, put a brick in, and returned it. And the person at the counter before me had taken it back without opening the box, because it was sealed and the seal looked fine.
How long had it been going on?
Hilbert: Long enough that I still check the seals on boxes. To this day. Any box. If I'm buying something, I check the seal. My wife thinks it's a tic. It's not a tic. It's a habit I earned.
What happened to the brick?
Hilbert: We kept it under the counter. Used it as a paperweight. Best thing that came out of that job.
So the lesson is that the policy was never the problem. The policy was fine. The problem was that nobody was willing to enforce it, because enforcing it meant a scene, and a scene meant a complaint, and a complaint meant a bad survey score.
Hilbert: That's the whole thing. The survey score ran the store. Not the manager. The survey. And a customer who's told no gives you a one. So nobody told anyone no. And the swappers figured that out faster than the honest customers did.
Which is exactly the incentive structure we've been describing. The company optimizes for the metric, and the metric rewards never saying no.
Hilbert: And the brick is what you get. Anyway. I have an appointment.
Now?
Hilbert: I'm already late for it.
Okay.
Hilbert: Check your seals.
The future of the good faith economy. That's where I want to land, because I think we're watching the end of something.
We're watching the end of the assumption. The next phase is tiered trust. The customers with clean histories get the generous policy. The customers with flagged histories get the strict, no-exception, restocking-fee, store-credit-only policy. And the tier you're in is determined by a model you can't see.
The model is trained on data that includes people who did nothing wrong.
It includes people who did nothing wrong. The real-estate agent with three phone cases is in the training data. And she doesn't know why she's in the strict tier, and she can't appeal, and the company can't explain it, because the model can't explain itself.
Which is a strange place to end up. We started this episode at a market stall, where the entire enforcement mechanism was a nod and a bit of eye contact. And we've ended up at a secret score that decides whether you're allowed to return a shirt.
The nod was cheaper. It was cheaper and it was kinder and it worked, mostly, because most people are decent. And the whole system we've been describing is what happens when a small number of people aren't, and the rest of us pay for it.
There's a line from one of the threads on this that I keep coming back to. Somebody wrote that there are a lot of nice things in life that work because most people aren't terrible, and they don't want to live in a world dominated by policies designed to protect against the ones who are.
That's the tension. The convenience we demand and the trust we're willing to extend are pulling in opposite directions. And the environment is quietly paying the bill, because a returned item that gets destroyed is a total loss, and that loss is priced into everything.
The regulatory angle is coming. If the waste keeps showing up in landfills, you're going to see return taxes, mandatory restocking fees, something that puts a price on the transaction.
Then the free rental stops being free. Which, honestly, might be the cleanest fix. A price.
A price is honest. A secret score is not.
A price is honest.
Alright. That's the episode. Thanks to Hilbert Flumingtop, our producer, who has an appointment he is now very late for.
He's going to make it. He always makes it.
This has been My Weird Prompts. If you enjoyed this one, a review helps other people find the show. You can find everything at my weird prompts dot com. We'll be back soon.
See you then.
