One note before the story. This article involves compulsive spending, a drained retirement account, and a marriage that ended over it. Compulsive spending and gambling are close enough that some students will have watched a family member go through something like this. I don’t think that is a reason to skip the case — it is a reason to introduce it as a design and ethics problem rather than as a story about one man’s foolishness, and to let students take it seriously without anyone having to say why.

Sean Harding is an accountant. Over four months he spent $1,362,687.31 on a live-shopping app, working through his savings, then personal loans, then money borrowed from a friend, and finally his employer’s corporate credit card. He resigned. His wife filed for divorce. He sold the house, the car, and the cards. Then, this year, he spent another $37,000. Harding’s story opens a recent Wall Street Journal investigation of Whatnot (“Whatnot, the Live Shopping App Where Some People Bid Until They’re Broke,” August 4, 2026), and the line worth bringing to class is his own explanation of how it happened: it was easy to swipe, and you don’t remember how many times you have done it.

Whatnot auctions almost anything live inside the app — trading cards, coins, sneakers, designer fashion, houseplants. The company says it moved roughly $8 billion in goods last year and holds close to 60 percent of live commerce in North America and Europe. Auctions run on a 45-second countdown; rather than typing a bid, buyers swipe a bar that automatically offers a few dollars over the current high. One swipe, no confirmation, no checkout screen. During one Pokémon auction, a card climbed to $170 in the closing seconds while the host insisted he had paid more himself. That card in that condition regularly sells for under $80. Mark Leiser, a deceptive-design researcher who taught digital law at Leiden University, calls the cumulative effect a “degradation of consent” — not one deceptive act, but the steady removal of the moment when a person decides. CEO Grant LaFontaine rejects the framing, says the company is building a good user experience, and points to a 4.7 App Store rating and internal data where overspending has not registered as a significant problem. Both accounts appear to be true, and that is the tension worth handing to students.

Claude was used to generate a first draft of this blog post.

Relevant Chapters in Essentials of Marketing

Chapter 5 notes that the excitement of an on-site auction may stimulate impulse buying more than checking out an auction online, which might lead to more consideration. Whatnot’s entire design is an argument against the second half of that sentence — the countdown clock, the host, the chat, and the one-swipe bid import the arousal of a live room into the channel the chapter treats as the more deliberate one. Chapter 5’s problem-solving continuum sharpens it further: a $170 purchase of an infrequently bought, hard-to-value collectible sits toward extensive problem solving, where consumers search for information and weigh alternatives, yet the app compresses it into something closer to routinized response behavior. Leiser’s “degradation of consent” is a name for that gap. And Chapter 5’s own #AI4M box asks the question this case answers — it notes that anchoring, scarcity cues, and social proof are now optimized by algorithms, and asks at what point a gentle push becomes a mighty shove.

Chapter 12 (retailers, wholesalers, and their strategy planning) is the second anchor, and the useful thing is that Whatnot does not fit its categories cleanly. Chapter 12 defines social commerce as an entire shopping experience occurring within a social media platform, and Exhibit 12–7 groups social commerce and influencer storefronts under new retail models and marketplaces. Whatnot inverts the order: it is not a social platform that added shopping, it is a shopping platform that added the social mechanics. Students can argue whether it belongs in that category, in platform-based e-commerce, or in a row the exhibit does not have yet. The chapter’s comparison of online and in-store advantages is also worth revisiting, since Whatnot deliberately imports several in-store advantages — the personal help, the social experience, the instant gratification — into an online format that keeps the low costs and wide assortment.

Chapter 1 (marketing’s value to consumers, firms, and society) carries the ethical weight, and it is the chapter that keeps this from becoming a simple villain story. The marketing concept means that an organization aims all of its efforts at satisfying its customers — at a profit, and LaFontaine’s defense is essentially a marketing-concept defense: the ratings are high, the complaints are reviewed, the problems get fixed. Chapter 1 also frames macro-marketing as a social process that directs an economy’s flow of goods and services in a way that matches supply and demand and accomplishes the objectives of society, which is a different and harder test than customer satisfaction scores. The chapter’s treatment of marketing ethics is explicit that it will not moralize, and the AMA code in Exhibit 1–7 gives students something more useful than instinct to argue from.

Class Discussion Ideas

The trap with this case is that students arrive already agreeing — the app is predatory, the man should have stopped. That conversation is over in four minutes. The better use of the class period is to make them locate exactly where the harm enters, because every individual feature here is ordinary. Countdown auctions are old. Enthusiastic sellers are old. One-click purchasing is standard and generally considered good design. The activities below are built to keep students from resolving the question too early.

In-Class Activities

Find the Missing Step. In small groups, have students lay out the steps a careful buyer would take before spending $170 on a collectible — establish a budget, check recent sold prices, assess condition, compare sellers — and then mark which steps a 45-second swipe auction removes and which it merely compresses. The payoff is that most of the process survives in theory and none of it survives in practice. (Chapter 5)

Price the Friction. Teams design three specific interventions that would slow a purchase — a confirmation screen, a daily cap, a cooling-off period on high bids — and then estimate what each would cost Whatnot in revenue and rank them by that cost. Require a recommendation at the end. This forces students to see why a firm acting in good faith still finds these decisions hard. (Chapter 5)

Fix Exhibit 12–7. Live auction commerce does not appear by name on the chapter’s trends exhibit. Have students decide where it belongs — social commerce, platform-based e-commerce, or a new row they write themselves — and defend the placement using the definitions in the chapter. Good short activity, and students enjoy being asked to amend the textbook. (Chapter 12)

Argue the CEO’s Case. Assign students to build the strongest version of LaFontaine’s defense using the marketing concept and the AMA code of ethics in Exhibit 1–7, then have a second group respond using the same two sources. Requiring both sides to argue from the same standard produces a much better discussion than asking whether the app is ethical. (Chapter 1)

Discussion Questions

1. Chapter 5 suggests that an on-site auction may stimulate impulse buying more than an online auction, which might allow more consideration. Does Whatnot fit that contrast? (Chapter 5)
   • Answer: No, and that is what makes it worth discussing. The chapter’s distinction assumes the online channel is quieter — a screen, a page, time to think. Whatnot rebuilds the arousal of the physical auction room inside the phone: a live host, a visible crowd in the chat, a running countdown, and other bidders acting in real time. Students should identify which specific features do that work. The broader lesson is that the channel itself was never what created deliberation; the pace and the social pressure were, and those can be moved.
2. Using the problem-solving continuum, where should a $170 collectible purchase fall, and where does it actually fall on this app? (Chapter 5)
   • Answer: Chapter 5 places infrequent, expensive, higher-risk purchases toward extensive problem solving, where buyers search for information, consider more alternatives, and apply more evaluation criteria. A collectible whose value depends on condition and grading is exactly that kind of purchase. In practice the app produces something closer to routinized response behavior — a repeated, low-effort swipe. Strong answers will notice that the mismatch is the whole problem: the buyer is using a decision process suited to buying milk for a decision that resembles buying a used car.
3. Chapter 5 asks at what point a gentle nudge becomes a mighty shove. Where would you draw that line here? (Chapter 5)
   • Answer: Open-ended by design, and the most valuable question in the set. Push students to rule on individual features rather than the app as a whole, because the features are ordinary in isolation: a countdown clock, an enthusiastic host, a visible chat, one-click purchasing, an algorithmic feed. Most students will locate the line at either the removal of the confirmation step or the combination of a randomized outcome with a countdown. Ask what principle they are using — asymmetry of information, absence of a decision point, exploitation of a known vulnerability — since the principle is what transfers to the next case.
4. Chapter 12 defines social commerce as an entire shopping experience occurring within a social media platform. Is Whatnot social commerce? (Chapter 12)
   • Answer: Only awkwardly, which is the useful part. Whatnot is not a social platform that added a purchase flow, like Instagram or TikTok; it is a commerce platform that adopted social mechanics — hosts with followings, live chat, an algorithmic feed. By the letter of the definition it may not qualify; by function it plainly belongs alongside the others. Students who argue for a new category on Exhibit 12–7 should be asked what defines it, and “live” is a weaker answer than “the seller is a performer and the audience is part of the product.”
5. Whatnot takes about 6 percent of each sale and does not own the inventory. In Chapter 12’s terms, what kind of business is it, and who is responsible when goods are stolen or misrepresented? (Chapter 12)
   • Answer: It behaves like a marketplace or platform rather than a retailer that buys and resells, which is precisely why responsibility is contested. The article describes coins that had been cleaned or altered and sold at prices they could not support, and police charging sellers with moving stolen merchandise. Whatnot says it prohibits misrepresentation, refunds buyers, bans sellers, and has tightened vetting. Students should weigh what a platform can reasonably police at 550,000 hours of live video a week against the fact that it collects a percentage of every one of those sales. There is no settled answer, and the honest ones acknowledge that scale is both the defense and the problem.
6. LaFontaine points to a 4.7 App Store rating and says overspending has not appeared as a significant issue in the company’s data. Evaluate that defense against the marketing concept. (Chapter 1)
   • Answer: The marketing concept means an organization aims all of its efforts at satisfying its customers, at a profit, so by its plain terms his defense is responsive rather than evasive. The weakness is measurement. An average rating describes the middle of a distribution and is close to silent about its tail, and the customers in this article are the tail — one of whom is unlikely to be filing an App Store review while hiding loans from his wife. Students should be pushed to name what Whatnot would need to measure to detect the harm it says it does not see. Spending velocity, session length after midnight, and share of revenue from the heaviest one percent of accounts are all reasonable answers, and the article notes the company is beginning to look at the first of these.
7. Is this a marketing problem or a personal responsibility problem? (Chapter 1)
   • Answer: The right closing question, and students will likely split hard. Chapter 1 gives them a standard better than instinct: macro-marketing is a social process that directs an economy’s flow of goods and services in a way that matches supply and demand and accomplishes the objectives of society. That test is about the system, not the individual, and it can return a failing verdict even where every participant acted within the rules and no one was defrauded. Strong answers will hold both ideas at once — that Harding made every one of those swipes himself, and that a system reliably producing that outcome is not excused by the fact that its users pressed the button.