Five marketing lessons from the most punchable man on Madison Avenue – and one very expensive lesson about all-nighters.

Nobody wants to be Pete Campbell. That is more or less the point of him. Matthew Weiner built the character to be irritating on contact: the voice, the hairline that recedes like a punishment for something he did in a previous life, the rifle he acquired by returning a wedding gift. Don Draper gets the monologues. Pete gets the wince.

But run the seven seasons of Mad Men as a business audit instead of a drama and something inconvenient turns up. Don made the ads. Pete made the money. He is the only character who consistently does the boring, unglamorous, revenue-generating work of finding clients, keeping clients, and noticing things about clients that nobody asked him to notice. He is also, not coincidentally, the character who most clearly shows you how a marketing organization gets itself killed.

1. THE ADMIRAL TELEVISION MOMENT: THE INSIGHT WAS ALREADY IN THE ROOM

Season 3, episode 5, “The Fog.” Admiral Television’s national sales have flatlined. Pete goes through the numbers and spots something the client apparently never bothered to look at: Admiral sets are selling disproportionately well to Black consumers in specific markets. His proposal is not complicated. Buy the cheaper ad space in Black publications and speak to the people who are already buying the product.

The Admiral executives are appalled. Roger, Bert and Lane haul Pete in and dress him down for embarrassing the firm, then quietly concede that he was right about the idea and wrong about the customer he tried it on. Nineteen sixty-three, and the account man everybody hates has just invented multicultural marketing, badly.

Sixty-odd years later the arithmetic is not remotely in dispute. The Selig Center for Economic Growth, cited in Nielsen’s 2026 reporting, projects Black buying power in the United States to top $2.1 trillion in 2026, up roughly 2.4 times since 2000. Nielsen’s January 2026 study found that 67% of Black audiences pay more attention to advertising in media that reflects their culture, against 46% of the general population. Among Black consumers aged 18 to 34 – the demo every brand claims to want – 74% say they wish they saw more representation of their identity group in the content they consume. And 70% say they will stop buying from a brand they believe devalues their community, up from 66% in 2023.

The takeaway is two-sided, and the second half is the one people skip. Pete did not commission research. He read the client’s own sales report, which anyone could have read. Most of the best audience insight in your business is already sitting in a spreadsheet somebody exports every month and nobody interrogates. But Pete also detonated the idea by walking it cold into a room that was not ready to hear it. He had the data and no air cover. Sequence is strategy: socialize the uncomfortable insight internally, arm a champion, then present. The idea that dies in the room is rarely a bad idea. It is usually a badly staged one.

2. THE 71% PROBLEM

In the season 4 premiere, Lane Pryce says the quiet part with an accountant’s calm: Lucky Strike now accounts for 71% of the agency’s billings. Eleven episodes later, Lee Garner Jr. takes the account elsewhere and the firm nearly dies on the spot. Potential clients tell Don some version of the same thing – call us in six months, if you still exist.

Everybody in marketing can quote the retention numbers. Frederick Reichheld’s research at Bain, popularized through Harvard Business Review, holds that a 5% lift in customer retention raises profits somewhere between 25% and 95%, and that acquiring a new customer runs 5 to 25 times the cost of keeping one you already have. Fine. True. Printed on a thousand agency slides.

Almost nobody puts the companion number on a slide: what percentage of revenue walks out the door on one phone call from one person who is currently annoyed with you? Concentration is the risk that looks like success right up until the morning it doesn’t.

Pete is the character who instinctively hedges it. He is forever dragging in the accounts everybody else finds beneath them – Mohawk Airlines, Clearasil, Vicks Chemical, Sugarberry Ham. Roger sneers at every one of them. Those unglamorous logos are a meaningful part of why there is still an agency after Lucky Strike leaves. And when Pond’s demands that the firm drop Clearasil over a conflict in season 4, Pete does not simply eat the loss; he trades the small account away and comes back with the entire Vicks Chemical relationship. That is the account management move: when you are forced to give something up, make sure you are negotiating, not apologizing.

3. THE CHEVY ALL-NIGHTERS: CRUNCH AS A COSTUME

This is the part of the playbook that reads like a startup post-mortem, because it is one.

May 1968. Don and Ted, pitching against each other for a top-secret Chevrolet project code-named XP-887, work out over drinks that General Motors was always going to hand the business to McCann-Erickson and is only running the process to harvest ideas from smaller shops. So they merge two entire agencies overnight to win a car they have never seen. It works. Everyone is thrilled. Peggy writes the press release.

Then the account starts. Chevy rejects everything. Don orders the whole staff in for the weekend. The partners bring a doctor into the office to inject the creative department with an “energy serum” promising “24 to 72 hours of uninterrupted creative focus.” What that weekend actually produces: Ken Cosgrove tap-dancing in Don’s office, an X-Acto knife going into Stan Rizzo’s arm, and a pitch Don writes that turns out to be a love letter to his neighbor’s wife with a soup ad stapled to it. Total usable output for the weekend: approximately zero.

Then Ken comes back from Detroit with the calendar. Three years of monthly deadlines, every idea climbing an approval ladder through all of General Motors until it reaches God, with a final stop at God’s wife. Ken later gets shot in the face on a client hunting trip and steps off the account. Pete volunteers to replace him, because of course he does.

Here is how the heroic sacrifice pays off. In season 7, GM pulls the account and gives it to Campbell-Ewald. And the car? XP-887 reached showrooms in 1970 as the Chevrolet Vega. Motor Trend named it Car of the Year for 1971. In April 1972 Chevrolet recalled about 130,000 of them over a carburetor fault that could cause backfires and engine fires; that July it recalled roughly 500,000 more because the rear axle shafts could separate from the housing and take a wheel with them. The car rusted famously. Production ended in 1977 having damaged GM’s reputation for building a reliable small car.

Two years of all-nighters, one gunshot wound, one lost account, one punchline of a product. If there is a cleaner parable for hustle culture, I have not found it.

The research on this is not ambiguous, and it has not been ambiguous for a while:

• The hours are physically dangerous. The World Health Organization and the International Labour Organization estimated in 2021 that long working hours caused 745,000 deaths from stroke and heart disease in 2016 alone, a 29% increase since 2000. Working 55 or more hours a week carries an estimated 35% higher risk of stroke and 17% higher risk of dying from ischemic heart disease compared with a 35-to-40-hour week. Roughly 488 million people, about 8.9% of the global population, work those hours. Long hours are now the single largest occupational disease burden there is – about a third of the total.

• The extra hours are not producing work. Stanford economist John Pencavel, publishing in The Economic Journal in 2015, found output per hour falls sharply past about 50 hours a week and falls off a cliff past 55. A 70-hour week produces roughly what a 55-hour week produces. Fifteen hours of someone’s life, converted directly into nothing.

• The people are leaving. Gallup finds 76% of employees experience burnout on the job at least sometimes, and 28% report feeling burned out very often or always. Burned-out employees are 63% more likely to take a sick day, 23% more likely to visit an emergency room, and 2.6 times more likely to be actively hunting for another job. The resignation is the lagging indicator; the all-nighter is the leading one.

• And the industries that romanticize it know exactly what they are doing. The International Game Developers Association’s 2015 satisfaction survey found 62% of developers said their job involved crunch. Of those, nearly half worked more than 60 hours a week and 17% topped 70. Thirty-seven percent received nothing extra for it. Not overtime – nothing. In a later industry survey, 89% of respondents in crunch got no paid overtime, only perks like catered dinners.

The Chevy arc is a complete crunch cycle compressed into one client: prestige logo, heroic mobilization, chemical assistance, injury, three years of misery, account lost anyway, product remembered as a joke. The question to ask when your team is pulling all-nighters is not whether they are committed. Commitment is not scarce. The question is whether anyone has audited whether the account is worth it and whether the hours are generating anything at all. In Mad Men, nobody asks. In most companies, nobody asks either.

4. PETE’S ACTUAL SUPERPOWER: HE CALLED BACK

Pete cannot write. He has no taste. His pitches are competent at best. What he has is a total inability to stop asking. He works his college network, his in-laws, his wife’s dinner parties, his father-in-law’s business, and every stray connection in a Rolodex he tends more carefully than his marriage.

The widely circulated sales figures – originally compiled by groups including the Brevet Group and the National Sales Executive Association, and worth treating as directional rather than gospel – say roughly 80% of non-routine sales require five or more follow-ups, while 44% of salespeople give up after one, another 22% after two, and about 92% have quit by the fourth attempt. Only about 8% keep going past five, and that 8% closes the large majority of the business. It takes an average of eight attempts just to reach a cold prospect on the phone.

Set the exact decimals aside; the shape has been consistent for thirty years. Persistence is not a personality trait in sales, it is an arbitrage. Pete Campbell is unpleasant, mediocre at the craft, and structurally underestimated by everyone around him, and he still out-earns better men because he makes the eighth call. That is the whole trick.

5. THE REPUTATION LEDGER, WHICH PETE KEEPS BADLY

Every one of Pete’s self-inflicted disasters comes from the same error: he treats leverage as a substitute for standing. In season 1 he brings Bert Cooper proof that Don Draper is not Don Draper, expecting to be rewarded, and gets the most deflating response in the series – Cooper simply does not care, and asks who exactly is going to want Pete after that. Information is not power when you have no credibility to spend it from.

The same logic runs the other way, too. In season 6, Pete’s personal conduct at a brothel – where he runs into his own father-in-law, who is doing the same thing – costs the agency the Vicks account outright. Personal behavior is not a personal matter when you are the relationship. It is a line item.

THE SCORECARD

1. Read the client’s own numbers before you commission research. Admiral’s opportunity was in Admiral’s sales report.

2. Stage the uncomfortable insight. Internal air cover first, client second. The right idea in the wrong room is a lost idea.

3. Know your 71%. Track revenue concentration as obsessively as you track growth, because concentration looks exactly like success until it doesn’t.

4. When you are forced to give something up, trade it. Do not just absorb the loss.

5. Audit the all-nighter. Past 55 hours you are burning people to produce nothing, and the trophy account may turn out to be a Vega.

6. Make the eighth call. Ninety-two percent of your competitors have already stopped.

Pete gets a reasonably happy ending, incidentally. A Learjet, a job at Wichita, and Trudy back. The show’s meanest joke is that the guy nobody wanted to be is the one who ends up fine.

SOURCES

Mad Men, AMC. Episodes referenced: “Shoot” (S1E9), “Nixon vs. Kennedy” (S1E13), “The Fog” (S3E5), “Public Relations” (S4E1), “The Rejected” (S4E4), “Chinese Wall” (S4E11), “For Immediate Release” (S6E6), “The Crash” (S6E8), “Favors” (S6E11), “The Strategy” (S7E6). Episode detail cross-checked against the Duke University Rubenstein Library “Mad Men Mondays” recaps and the Mad Men Wiki.

Nielsen, “The Black Influence: How Black Culture and Identity Drive the Market,” January 19, 2026; Nielsen, “Engaging Black Audiences,” 2025; buying-power projections from the Selig Center for Economic Growth, University of Georgia.

Reichheld, F. and Sasser, W.E., “Zero Defections,” Harvard Business Review, 1990; Reichheld, The Loyalty Effect (Bain & Company, 1996); ratios as restated by HBR, 2014.

World Health Organization and International Labour Organization, “Long working hours increasing deaths from heart disease and stroke,” joint news release, May 17, 2021, and the underlying WHO/ILO Joint Estimates published in Environment International, 2021.

Pencavel, J., “The Productivity of Working Hours,” The Economic Journal, vol. 125, 2015.

Gallup, “Employee Burnout: Causes and Cures” and related Gallup workplace analytics. International Game Developers Association, Developer Satisfaction Survey, 2015 press release; IGDA overtime compensation data as reported by LBB Online, 2018.

Chevrolet Vega history and recall figures: Ate Up With Motor, “Falling Star: The Checkered History of the Chevrolet Vega”; Consumer Guide. Sales follow-up figures: The Brevet Group (2014) and persistence-decay data attributed to the National Sales Executive Association – widely repeated benchmarks with imperfect provenance, cited here as directional.