A companion piece to Episode 15 of the BMD marketing case-study series, and the last of the brand-versus-growth half. Episode 14 was about a campaign that announced a promise for twenty years without ever being bound by it. This one is the opposite case and the harder one, because the campaign worked: Nike put the most divisive man in American sport at the centre of its thirtieth anniversary, took three per cent off its own share price in a day, and set a record high nine days later. This is the full autopsy of the four numbers that story is told with, none of which Nike has ever published.
Contents
Nine words and a three day window
On Monday 3 September 2018, Colin Kaepernick posted a black and white photograph of his own face. Across it ran nine words: believe in something, even if it means sacrificing everything.
He had not played a professional down in twenty months. He was in the middle of a legal case against the league that had frozen him out, and Nike, which had just made him the face of the thirtieth anniversary of Just Do It, was that league's uniform supplier. Within a day people were filming themselves burning shoes they had already paid for. On the Tuesday the stock closed down 3.2 per cent.
Then the part everybody quotes. The share price recovered inside two weeks and set an all-time high. Online sales rose 31 per cent. The campaign added 6 billion dollars of market value and 163 million dollars of earned media. The film won an Emmy. It is taught as the bravest marketing decision of the decade and as the proof that taking a side pays.
Every one of those numbers is real, and not one of them is a Nike figure. The 31 per cent is three days long and one of those three days ran before the ad existed. The 6 billion dollars is a share price. The 163 million is a model. And the quarterly results everybody points to as confirmation cover a period that closed two days before the campaign launched. This is the autopsy: what each number counts, what Nike knew about its own buyers before it ran any of it, and the single thing the point of view turned out to forbid.
What Nike already had before the first post
Nike did not discover anything in September 2018. It went into that month with 36.4 billion dollars of annual revenue and a share price up around 36 per cent on the year, the best performing stock in the Dow. Whatever the campaign did, it did it to a business that was already compounding hard and was already being rewarded for it.
It also went in with the customer data. NPD Group's read was that nearly two thirds of the people wearing Nike in the United States were under 35, and that the buyer base ran ahead of the national population on the measures that mattered to this particular decision: roughly 18 per cent of Nike buyers were Black against 13 per cent of Americans, and roughly 19 per cent were Hispanic against 16 per cent. Nike knew the shape of its own base before it picked the face of the campaign, not after.
And Kaepernick was not a signing. Nike had signed him in 2011, seven years earlier. He was already on the endorsement roster through the 2016 protests, through two seasons without a team, and through the whole period in which he became the most argued about figure in American sport. The company did not go out and acquire a controversy. It already owned one and had been paying for it quietly.
Six months before the campaign, in March 2018, Nike extended its NFL agreement: official supplier of uniforms and sideline apparel through the 2028 season, an eight year extension. So at the moment the ad ran, Nike was the uniform supplier to the league that Kaepernick was suing, under a contract it had just locked in for a decade.
That is the machine. A compounding business, a measured buyer base, an athlete already under contract, and a league relationship freshly secured. The campaign is usually described as the thing that created the result. It arrived on top of all four.
The decision that was not the stand
The version taught in business schools is a boardroom choosing principle over profit and accepting the consequences.
What was reported at the time is close to the reverse. The New York Times account, carried by CNBC that September, is that in the summer of 2017 Nike's marketers decided to cancel Kaepernick's contract. He had no team, he was not generating product sales, and keeping him on the roster was seen as a risk to the NFL relationship that Nike would go on to extend the following March.
What reversed it was the agency. Wieden and Kennedy, which had written Just Do It in 1988, argued Nike out of the cancellation and into making him the centre of the anniversary campaign. The contract that was about to lapse was renegotiated into a multi-year deal instead.
So the famous act of conviction began as a near cancellation that an outside partner talked the company out of. That is not a smaller story. It is a more useful one.
If the lesson is that Nike found its principles, there is nothing in it for anybody else, because you cannot instruct a company to have a conviction. If the lesson is that a company nearly dropped an asset and a partner showed it what that asset was worth against customer data the company already held, that is a decision a marketing department can actually reproduce.
What online sales rose 31 per cent counts
The 31 per cent comes from Edison Trends, a firm that estimates purchasing by reading anonymised and aggregated e-receipts, in this case from a panel of more than three million American consumers. Nike has never published the figure, before or since.
The window is three days. Sunday 2 September to Tuesday 4 September 2018, compared against the same three calendar days in 2017.
The first ad went up on the Monday afternoon, 3 September. Sunday the 2nd, a full third of the measurement window, ran before the campaign existed.
The same three days in 2017 grew 17 per cent by the same method, with no campaign attached to them at all. So the portion of the 31 that the campaign could conceivably account for is around fourteen points across a long weekend, and even that assumes nothing else in two years had changed.
The window is also the American Labor Day weekend, which is one of the two heaviest discount weekends in the United States retail calendar. It is the period a sportswear company would expect its online sales to grow in whether or not it had run anything.
None of this makes the number wrong. It makes it a three day estimate of online orders, produced by a third party panel, across a discount weekend that began before the advertisement did. That is a different sentence from sales rose 31 per cent after the Kaepernick ad, and the different sentence is the one in every case study.
The scoreboard
| Metric | Figure |
|---|---|
| The campaign | Dream Crazy, the thirtieth anniversary of Just Do It, built by Wieden and Kennedy. Announced 3 September 2018 through Kaepernick's own account, with the two minute film following that week |
| The famous sales figure | Online sales up 31 per cent, estimated by Edison Trends from anonymised e-receipts of more than 3 million United States consumers. Never published by Nike |
| The window it covers | Three days: Sunday 2 September to Tuesday 4 September 2018, against the same three days of 2017 |
| What preceded the ad inside that window | Sunday 2 September. The first advertisement went up on the Monday afternoon, so one of the three measured days ran before the campaign existed |
| The same window without a campaign | The equivalent three days in 2017 grew 17 per cent by the same method, on the same Labor Day discount weekend |
| The share price on the day | Closed down 3.2 per cent at 79.60 dollars on Tuesday 4 September 2018, about 3.75 billion dollars of market value |
| What the competition did that day | Puma and Adidas opened to similar losses and the wider market was down. Neither had run a comparable campaign |
| The record high | An all-time high of about 83.90 dollars on 13 September 2018, nine days after the fall |
| The famous 6 billion dollars | The market capitalisation gain from the Labor Day announcement to that high, a move of roughly 5 per cent, on a stock already up about 36 per cent for the year and leading the Dow |
| The famous 163 million dollars | An earned media exposure estimate by Apex Marketing Group one week after launch, up from about 43 million in the first 24 hours. A model of what the coverage would have cost to buy, not money received |
| The quarter everybody credits | Fiscal first quarter 2019, revenue up 10 per cent to 9.9 billion dollars and EPS up 18 per cent, reported 25 September 2018. That quarter ended 31 August 2018, two days before the advertisement ran |
| The first quarter that contained the campaign | Fiscal second quarter 2019, ended 30 November 2018 and reported 20 December 2018: revenue up 10 per cent to 9.4 billion dollars, up 14 per cent currency neutral, North America digital up more than 30 per cent. Nike's release did not mention the campaign |
Every headline figure in this case is an outside estimate or a market price, and the table labels each one as such. The 31 per cent is a third party panel estimate of online orders over three days and has never been confirmed by Nike. The 6 billion dollars and the 3.75 billion dollars are market capitalisation moves, which is to say prices on a date rather than money earned or lost by the business. The 163 million dollars is an exposure valuation model from a media analytics firm. Only the quarterly revenue and earnings figures are audited company disclosures, and they are included here specifically to show which periods they cover. The customer composition figures are NPD Group panel research as reported in the trade press of the period. Several primary pages are blocked by the egress proxy, including `sec.gov`, `forbes.com`, `techcrunch.com`, `inc.com` and `adage.com`, so Nike's investor releases, the Edison Trends analysis and the Apex Marketing estimate were confirmed through search results and through outlets quoting those documents rather than by opening them directly.
The autopsy: four famous numbers, none of them Nike's
The second number is the 6 billion dollars. It is a market capitalisation delta, measured from the Labor Day announcement price to the all-time high of about 83.90 dollars on 13 September 2018, which is a move of roughly five per cent. It is not revenue and not profit. It is the market's opinion of Nike over nine days, on a stock that was already up about 36 per cent for the year and leading the Dow before anybody had seen the ad.
The third is the 163 million dollars of earned media, estimated by Apex Marketing Group a week after launch, having been about 43 million in the first 24 hours. Exposure value is a model: it prices the coverage a company received as though the company had bought it at rate card. No money changed hands in either direction. It is a useful way to size attention and a bad way to size a result.
The fourth is the 3.2 per cent fall on 4 September, quoted as the market punishing the stand. On that same day Puma and Adidas opened to similar losses and the wider market was down. Neither of them had run a Kaepernick campaign, or any campaign. The most cited evidence that Nike paid a price for the ad is largely the sector moving together.
Then the date order, which is the cleanest this series has had since Old Spice. On 25 September 2018 Nike reported revenue up 10 per cent to 9.9 billion dollars and earnings per share up 18 per cent, and the coverage that day tied the result to the campaign. Those are the figures for fiscal first quarter 2019, and that quarter ended on 31 August 2018. Two days before the advertisement ran.
The first quarter that actually contained the campaign was reported on 20 December 2018: revenue up 10 per cent to 9.4 billion dollars, up 14 per cent currency neutral, with North America digital up more than 30 per cent. Those are good numbers. They are also almost exactly the same growth rate as the quarter that ended before the campaign started, which is the most honest single fact in the whole case.
Nike's own results statements did not mention the campaign. Not in September, not in December. The company credited with the most successful values campaign in modern marketing has never attached a figure to it in a results release, and the reason is probably that it could not separate one from everything else the Consumer Direct Offense was doing at the time.
So the shape is this. Four famous numbers: a three day panel estimate, a share price, an exposure model, and a one day market move that the competition shared. None of them disclosed by the company. And the only audited figures in the story cover a period that ended before the story began.
The control group, and the question that settles it
This episode was handed a control group in two forms, which is rare.
The first is the same day comparison. Puma and Adidas fell alongside Nike on 4 September without having taken any position on anything, which tells you most of what the three per cent was.
The second is that nobody else took the stand. Under Armour, Adidas and Puma sell into substantially overlapping audiences, the same argument was available to all of them, and none of them went near it. When a tactic is celebrated as the decade's best idea and no competitor copies it, that is worth reading carefully: either they were slower, or the move was specific to Nike's particular base in a way that does not generalise. The buyer data above says the second.
Which brings the episode to the question Dove raised one episode ago. Announcing a point of view is cheap. Being bound by one is expensive. So what did this point of view actually forbid?
Once, it forbade a product. In July 2019 Nike recalled the Air Max 1 carrying the Betsy Ross flag after Kaepernick told the company that the symbol was tied to the era of slavery. The shoe had already shipped to retailers for the Fourth of July. Arizona's governor then withdrew incentives for a 185 million dollar plant and 500 jobs in the state.
That is a real cost, paid in public, and it makes Nike's case genuinely stronger than Dove's. Dove's platform forbade nothing in twenty years. Nike's forbade a shoe inside twelve months and took a state level consequence for it.
But the contract the stand never touched was the NFL one. Nike remained the league's uniform supplier throughout, under the extension it had signed in March 2018 and running to 2028. Kaepernick settled his collusion case against the league in February 2019, reported at under 10 million dollars across two players, and never played again. Within days of that settlement Nike released a Kaepernick Icon jersey.
So the position was binding on a product line and was never binding on the relationship that created the grievance in the first place. That is not hypocrisy and it is not a gotcha. It is the actual shape of a commercial values position, and it is the part the case studies leave out, which is precisely why companies that copy the surface of this campaign are surprised by what it costs them.
The right order: read the base, then take the stand
Step one is to measure who buys from you before you decide who to speak to. Nike had the age profile and the buyer mix of its own customers in hand before it chose the face of the campaign. The stand was aimed at people already in the base, which is why the boycott was so loud and so financially irrelevant: the people burning the shoes had already bought them, and in most cases were not repeat buyers to begin with.
Step two is to write down what the position forbids before you announce it. Nike's answer, revealed a year later, was one shoe and not the league contract. Knowing that in advance is the entire difference between a position and a liability, because the moment somebody asks you to act on your stated values, the answer you give is the position, not the advertisement.
Step three is to decide what would count as proof before the campaign runs, and to make it something visible in your own numbers rather than someone else's panel. Nike never set that measure publicly, which is why a campaign this famous is still argued over with a three day e-receipt estimate and a share price chart eight years later.
At Saudi scale the three steps get cheaper, not harder. A Riyadh company turning over 40 million riyals does not need NPD, because it already owns the data NPD sells: point of sale records, the CRM, the delivery addresses, the repeat rate by segment. An afternoon with that will tell you who your base actually is with more precision than a national panel, and it is the same afternoon that tells you whether a public position would be aimed at your customers or past them. The local wrinkle is concentration. In a market where a large share of revenue can sit with a small number of family owned accounts, a position that reads as precision in a fragmented consumer market can read as an exit from a key account instead, so step two matters more here than it does in Portland.
The order is the point. Polarisation is a targeting decision, and it works the way targeting works: it concentrates attention among people who already lean your way. It confirms a base. It does not build one. Budget it as reach against customers you already have, and the campaign becomes a reasonable bet. Budget it as acquisition and you are buying an argument you cannot win.
The takeaways
A number quoted without its window is not evidence. The 31 per cent is three days long, one of those days preceded the ad, and the same window grew 17 per cent the year before with no campaign at all.
Check the date order before the argument. The quarterly results used to prove this campaign worked cover a quarter that ended two days before the campaign launched.
A share price is a price. The 6 billion dollars is nine days of market opinion on a stock already up 36 per cent for the year, and the 3.2 per cent fall was shared by two competitors who had taken no position.
The base data comes first. Nike knew the age and the composition of its buyers before it chose the face. That sequence is what turned a polarising decision into a targeting decision.
Ask what the position forbids. Nike gave up a shoe and a state incentive package, and never gave up the NFL contract. A point of view with nothing behind it is a slogan, and a point of view you have not priced is a liability.
Polarisation confirms a base and does not convert one. The people burning the shoes had already bought them. Nothing in this case shows a single customer switching to Nike over the stand.
Frequently asked questions
Did the Kaepernick ad really increase Nike's sales by 31 per cent?
No, and the figure is not a sales figure in the sense people use it. Edison Trends estimated that Nike's online sales across three days, Sunday 2 to Tuesday 4 September 2018, were 31 per cent higher than the same three days in 2017, using a panel of e-receipts from more than three million United States consumers. The first advertisement appeared on the Monday afternoon, so one of the three measured days ran before the campaign. The same three days in 2017 had grown 17 per cent with no campaign, and the window is the Labor Day discount weekend. Nike has never published the number.
Did Nike lose money because of the boycott?
There is no evidence that it did. The clearest reading is that the people filming themselves burning Nike products had already bought them, so the destruction was of goods Nike had already been paid for. The 3.2 per cent share price fall on 4 September 2018 is the figure usually cited as the cost, but Puma and Adidas opened to similar losses the same day without having run any comparable campaign, and the wider market was down.
Where does the 6 billion dollar figure come from?
It is a market capitalisation delta, not revenue or profit. Nike's shares fell 3.2 per cent on 4 September 2018 and then reached an all-time high of about 83.90 dollars on 13 September. Measured from the Labor Day announcement, that run is roughly a five per cent move and adds up to about 6 billion dollars of market value. The context that is usually dropped is that Nike stock was already up around 36 per cent for the year at that point and was the best performer in the Dow.
Did Nike's earnings prove the campaign worked?
The earnings usually quoted cannot prove anything about the campaign, because of the dates. Nike reported revenue up 10 per cent to 9.9 billion dollars on 25 September 2018, and that is fiscal first quarter 2019, which ended on 31 August 2018, two days before the advertisement ran. The first quarter that actually contained the campaign was reported on 20 December 2018, showing revenue up 10 per cent to 9.4 billion dollars, which is essentially the same growth rate as the quarter before the campaign. Nike's own releases did not mention the campaign in either quarter.
Was putting Kaepernick at the centre of the campaign a risk?
Less of one than the retelling suggests. Kaepernick had been under Nike endorsement contract since 2011, so he was an existing asset rather than a new signing. Nike held research showing that nearly two thirds of its United States wearers were under 35 and that its buyer base was more racially diverse than the national population. And reporting at the time indicated that Nike's marketers had decided to cancel Kaepernick's contract in 2017, before its agency argued them out of it. The decision looks less like a leap of conscience and more like a targeting decision taken against data the company already held.
Should a mid sized company take a public stand on a social issue?
Only after two steps, and in this order. First, measure who actually buys from you, because a stand aimed at people who are not your customers costs you the ones who are. Second, write down what the position forbids before you announce it, because the moment someone asks you to act on it, the answer you give is the position. Nike's answer turned out to be that it would withdraw a shoe and would not touch its NFL contract. A stand you have not priced is a liability, and a stand aimed at a base you have not measured is a guess.
This article is part of BMD's marketing case-study series. Episode 15 closes the brand and growth half of the season. Episode 16 opens the Gulf run with two Saudi companies racing to own how the Kingdom pays, one of them valued at 4.5 billion dollars and heading for a listing. That is Tabby and Tamara.
Sources and further reading
Nike's own investor releases for the quarterly figures and their period end dates, specifically the fiscal 2019 first quarter results of 25 September 2018 for the quarter ended 31 August 2018 and the fiscal 2019 second quarter results of 20 December 2018 for the quarter ended 30 November 2018. Edison Trends for the 31 per cent online sales estimate and its three day Sunday to Tuesday window, together with the 17 per cent comparison for the same days in 2017, as reported at the time by Time, NBC News, Marketing Dive and ABC News. Apex Marketing Group for the earned media exposure estimates of about 43 million dollars in the first 24 hours and more than 163 million dollars after one week. Contemporaneous market reporting from CNBC, CBS News, Money and CBC for the 3.2 per cent close at 79.60 dollars on 4 September 2018, the comparable Puma and Adidas moves that day, the all-time high of about 83.90 dollars on 13 September and the roughly 6 billion dollar market value gain. The New York Times reporting on Nike's 2017 decision to cancel the Kaepernick contract and Wieden and Kennedy's role in reversing it, as carried by CNBC on 27 September 2018. NPD Group research on the age and composition of Nike's United States buyer base as reported in the trade and business press of September 2018. Nike's March 2018 extension of its NFL uniform agreement through the 2028 season. NPR, CNBC and NBC News for the July 2019 withdrawal of the Betsy Ross Air Max 1 and the Arizona incentive decision that followed. NPR and CBS News for the February 2019 settlement of Kaepernick's collusion case against the NFL. Deadline and the Television Academy for the 2019 Creative Arts Emmy for outstanding commercial. Several of these publishers are blocked by the egress proxy used to research this piece, so the figures they carry were confirmed through search results quoting them rather than by opening the pages, and that is noted here so any figure can be spot checked.
About BMD
Most companies don't have a marketing problem. They have a marketing department that was never built. BMD is a boutique consultancy that installs structured, measurable marketing departments inside mid-market companies across the GCC. We don't run your campaigns, and we don't hand you a strategy deck and leave. We build the operating system: the structure, the measurement, and the ownership that turn marketing into a function leadership can rely on. The method is the BUILD framework, published and practiced: a book, an online program, a community of Gulf founders and marketers applying it, and diagnostics that replace assumptions with measurement. Delivered in Arabic and English, founder-led.
Redha Alayesh
A marketer with a software engineer's discipline and a scientist's mindset. Across 40+ marketing departments in the GCC, he built the BUILD framework to solve the problem he kept finding: capable marketers trapped inside companies that never built them a department.