Five expensive lessons from Mad Men’s most useful failure – and what his second act says about executive recruiting, then and now

The Man Who Was Right About Almost Everything and Still Lost

Everybody who watches Mad Men wants to be Don Draper. Nobody wants to be Duck Phillips. Which is exactly why Duck is the more useful study – most of us will not spend our careers dazzling clients with a carousel. Most of us are the person in the room with the numbers, the forecast, and the uncomfortable opinion about where the business is heading.

Herman “Duck” Phillips joins Sterling Cooper as Head of Account Services in late 1960, hired by Don himself. He comes over from London, he is a recovering alcoholic, and he is – on nearly every big strategic question – correct. He wants younger copywriters who understand the youth market. He wants the agency to get serious about television. He wants scale, because he can see that independent shops are about to get squeezed. He is right, right, and right. He also loses the client, the merger, the job, the sobriety, and the dog. That gap between being right and being effective is where the actual lessons live. Here are five of them, with the receipts.

Lesson 1: Never Fire a Paying Client to Chase a Prospect

Duck’s signature disaster: he persuades Roger Sterling that Sterling Cooper should drop Mohawk Airlines – a small, regional, paying client – so the agency can be conflict-free to pitch American Airlines after the Flight 1 crash in March 1962. American never signs. Sterling Cooper ends up with neither airline. Duck calls it the price of ambition. It is actually the price of arithmetic he never did.

The math has only gotten more brutal since 1962. Bain & Company’s landmark retention research, popularized through the Harvard Business Review, found that a 5% increase in customer retention lifts profits by 25% to 95%, and that acquiring a new customer costs somewhere between 5 and 25 times what it costs to keep an existing one. Marketing Metrics puts your odds of selling to an existing customer at 60–70%, versus 5–20% for a brand-new prospect. Duck traded a 65% probability for a 12% probability and called it strategy.

Pitching is not free, either. The ANA and 4As found in their Cost of the Pitch work that a formal agency review costs the client an average of $408,500 per pitch – and that is just the client’s side of the table. Agencies burn staff hours, morale, and opportunity cost on top of it. Every new-business chase you start is a bill you have already agreed to pay before anyone says yes.

Lesson 2: One Contact Is Not a Relationship. It’s a Single Point of Failure.

The reason the American Airlines pitch was dead on arrival is that Duck’s entire play rested on one man: his inside contact at the airline, who gets fired before the meeting happens. The agency presents anyway, which is the corporate equivalent of showing up to a wedding after the couple has called it off.

This is not a period-piece problem. Spencer Stuart’s tenure tracking, reported by Adweek, puts average CMO tenure at S&P 500 companies at 4.1 years in 2025, down from 4.3 in 2024 – the shortest run of any core C-suite seat except COO (3.3 years), against 4.7 for CFOs and 7.6 for CEOs. Spencer Stuart also found that 71% of Fortune 500 CMOs were doing the job for the first time, and that 22% of business-to-consumer marketing chiefs had been in their chair a year or less.

Translation: if your account depends on one champion, you should assume that champion has roughly a four-year shelf life and a one-in-five chance of being brand new. Map the org chart. Build three relationships deep on every account you care about. Duck had one, and one is a rounding error away from zero.

Lesson 3: Being Early Is Not the Same as Being Persuasive

Duck pushes Sterling Cooper toward youth copywriters and television when Don is still defending experience and print. Duck is reading the market correctly and everyone treats him like a nuisance.

Look at what he was actually looking at. Madison Avenue’s total television billings ran $12.3 million in 1949, $40.8 million in 1950, and $128 million in 1951 – roughly a tenfold jump in two years. Television penetration of American households went from about 9% in 1950 to 65% in 1955 to over 90% by the early 1960s. At BBDO, billings rose from $118 million in 1952 (radio and TV making up 34%) to $248 million in 1963, with radio and TV now 47% of the total. The Smithsonian’s account of the era notes that gross advertising spending in the 1950s climbed 75% – outpacing GNP, personal income, and every other economic index on the board.

Duck had all that wind at his back and still couldn’t move the room, because he pitched his conclusion instead of his evidence, and because he was always visibly building a case for Duck. Being early only pays if somebody follows you. A forecast nobody acts on is a diary entry.

Lesson 4: Know What You Are Actually Selling

Duck’s masterstroke is quietly steering Putnam, Powell and Lowe into buying Sterling Cooper, with himself installed as president. It collapses for one reason: the buyers discover that Don Draper has no employment contract. The agency’s crown jewel could walk out the door on a Tuesday. Duck had sold a balance sheet. The buyers were shopping for a person.

Every marketer makes this mistake in miniature. You sell the feature set when the customer is buying confidence. You sell the deliverable when the client is buying somebody to blame. You inventory your assets and forget that the most valuable one has legs, opinions, and a recruiter’s number in their phone.

Lesson 5: The Dog

In one of the show’s quietest scenes, Duck’s kids drop off the family dog, Chauncey, at his office. Duck wants a drink and can’t take one in front of the animal. So he slips the leash off, walks the dog out to the sidewalk, and goes back upstairs alone.

Nobody sees it. It never comes up in a performance review. And it tells you everything about how he will behave under pressure with your account, your team, and your reputation. Brand is not the deck. Brand is what a person or a company does when the meeting is over and nobody is taking notes. Duck’s collapse was visible in that hallway years before it hit the P&L.

Executive Recruitment: Then vs. Now

Here is the delicious part. When Duck washes out of the agency world, he reinvents himself as a headhunter – and he is finally good at something. By the sixth season he is placing executives in advertising, digging up dirt on Bob Benson for Pete Campbell, and eventually walking Lou Avery into SCDP as Don’s replacement. In the final season he is the one steering Pete Campbell toward the Learjet job. The man who couldn’t manage his own career turned out to be excellent at managing other people’s.

Then: The Rolodex Was the Product

Duck fell into a profession that was, at that exact moment, being invented. Heidrick & Struggles was founded in 1953, Spencer Stuart in 1956, and the Association of Executive Search Consultants – the body that wrote the ethics rules – in 1959. Korn Ferry and Russell Reynolds both opened their doors in 1969. Executive search grew out of management consulting, where firms like McKinsey and Booz Allen kept needing to find the leaders who could actually execute the strategies they had just sold.

The business model was pure information asymmetry. The recruiter knew who was unhappy, who had been passed over, who drank at lunch, and which company was quietly about to lose its president. You could not look any of that up. You bought it one lunch at a time from a man with a Rolodex and a very good memory.

Now: Everybody Has the Database

LinkedIn is past 1.2 billion members, and something close to 90% of talent acquisition teams use it as a primary weekly sourcing channel. Gem’s 2026 recruiting benchmarks – built on 165 million applicants and 1.2 million hires – report that 46% of sourced hires now come from candidates already sitting in the company’s own database, up from 26% in 2021. Korn Ferry reports that 84% of talent leaders plan to use AI in some capacity in 2026 and 52% plan to deploy autonomous AI agents for recruiting specifically.

The volume is genuinely absurd. Gem clocks the average hiring process at 13 interviews per hire, up 42% in three years, while applications per posting surged 93% in 2025 on the back of AI-assisted apply tools. LinkedIn recorded a peak of 14,200 applications per minute in February 2025. SHRM’s 2026 benchmarking puts median executive time-to-fill at 45 days (down from 60 in 2022) while median executive cost-per-hire climbed to $15,000 from $10,600 the year before. Faster and pricier at once.

What Hasn’t Changed Since 1962

Almost everything that mattered.

• The good jobs are still invisible. Research consistently puts 70–80% of roles as never publicly advertised, and recruiters put that figure at 80% or higher for VP-and-above positions. The hidden job market Duck worked in 1966 is still where the C-suite gets filled in 2026.
• Warm still beats cold. ERIN’s 2025 referral analysis found referred candidates hire at a 30% rate versus 7% from other sources, stay 45% longer, and reach offer in 29 days instead of 55. Referrals are about 2% of applicants and 11% of hires.
• The failure rate is stubbornly, embarrassingly high. Heidrick & Struggles analyzed 20,000 of its own placements and found 40% of senior executives are pushed out, fail, or quit within 18 months. Leadership IQ pegs new-hire failure at 46% inside 18 months – and found 89% of those failures were attitudinal, not a skills gap.
• It costs a fortune to get wrong. SHRM benchmarks put executive replacement at 200–213% of annual salary. Broader estimates from TopGrading and Chief Executive run far higher once you count strategic misdirection and the good people who quit on the way out.
• Humans still want a human deciding. Pew found 71% of Americans oppose letting AI make the final hiring call. The tooling has changed; the trust requirement has not.

The Lou Avery Test

Duck places Lou Avery as Don’s replacement. On paper Lou is a perfectly qualified creative director. In practice he is a culture bomb who alienates the department and produces nothing anyone remembers – the 89% statistic in a gray cardigan. The man could do the job and could not do it there.

Which is why retained search survived the internet. Anybody can find the candidate now; the scarce skill is knowing whether the candidate survives the room. Retained search still holds roughly 63% of a $58 billion global executive search market, per Mordor Intelligence, with C-suite work half of it. Nobody pays that for a list of names.

The Scoreboard: Turning Duck Into a Checklist

1. Audit your revenue concentration this quarter. If any one client is more than 20–25% of billings, you are running Mohawk risk whether you fired anyone or not.
2. Count your contacts per account. Fewer than three named relationships and you are one 4.1-year CMO tenure away from an unexplained loss.
3. Price the chase before you start it. If a $408,500 pitch is the client-side benchmark, your own new-business hours have a number too. Write it down.
4. Sell evidence, not conclusions. Duck’s television read was right and unheard. Bring the 9%-to-90% curve, not the verdict.
5. Contract your crown jewels. The Sterling Cooper deal died over one missing signature. Know which of your assets can walk.
6. Hire for the room, not just the résumé. Eighty-nine percent of new-hire failures are attitudinal. Interview for that, or pay 200%+ of salary to learn it the slow way.

The Redemption Nobody Talks About

Duck Phillips is remembered as a punchline – the guy who lost the airline, blew the merger, and did something unspeakable in the wrong executive’s office. But the last thing the show does with him is quietly let him win. He finds the work his actual talent fits: reading people, working a network, matching a person to a seat. He gets Pete Campbell the Learjet job, and Pete gets his family and his life back.

That is the lesson nobody puts on a slide. Duck didn’t fail for lack of ability. He failed because he spent years selling the wrong thing – including himself. The moment he stopped pitching the Duck he wished he were and started trading on what he was actually great at, he became indispensable. If you have been losing pitches lately, ask whether the problem is your deck or your product. Sometimes the product is you, and the fix is a different job, not a better slide.

Sources

•  Retention economics: Bain & Company / Harvard Business Review; Marketing Metrics close-rate data.

•  Agency economics: ANA & 4As, “Cost of the Pitch” (2023, 2024) and “Client-Agency AOR Relationship Tenure” (April 30, 2025).

•  Executive tenure: Spencer Stuart CMO Tenure Study, reported by Adweek (Jan. 2026) and Marketing Dive.

•  1950s–60s ad industry: Smithsonian National Museum of American History, “Madison Avenue, 1940s–1960s”; A Wealth of Common Sense, “A Short History of Advertising”; Business History Conference (BBDO billings).

•  Search industry history and size: Eurogalenus, “Headhunting: A Brief History of a Profession”; AESC / Alder Koten records; Mordor Intelligence, Executive Search Market 2025–2031.

•  Modern recruiting benchmarks: Gem 2026 Recruiting Benchmarks; LinkedIn Talent Solutions (2024–2025); Korn Ferry 2026 talent outlook; Pew Research Center; SHRM 2026 recruiting benchmarking.

•  Hiring failure and referrals: Heidrick & Struggles 20,000-placement analysis; Leadership IQ; ERIN 2025 employee referral analysis; published hidden-job-market estimates.

•  Mad Men (AMC, 2007–2015), created by Matthew Weiner – Seasons 1–2 (“For Those Who Think Young,” “Flight 1,” “Three Sundays,” “Maidenform”), Season 4, and Seasons 6–7.