A companion piece to Episode 4 of the BMD marketing case-study series. The video tells the story in minutes. This is the full autopsy: what actually happened, why the popular $100 million story is the wrong lesson, and the order smart companies use before they touch an icon.
Contents
A logo that lived six days
Gap put a new logo on gap.com on Monday 4 October 2010 and announced the revert on the evening of 11 October. Newsrooms called it a week. The six-day count is the public-notice window: word circulating 6 October, press confirm 12 October. Either way, a 20-year mark came off a homepage without a story, and came back because the people who already wore it said no.
The popular sentence that follows is that the experiment cost about 100 million dollars. That number is widely cited. It is not a figure Gap disclosed, and it does not sit in any 2010 filing we opened. A website swap that never reached the stores cannot carry a nine-figure rollout bill on its face.
This is a lesson for anyone who wants to "modernise" a mark people already treat as theirs. We have watched that impulse get expensive across 40+ marketing departments. The file is rarely the problem. The missing rehearsal is.
The blue box people already owned
For more than twenty years the Gap wordmark was white letters inside a dark blue square. Not the most loved design in American fashion. The one on the shopping bag, the hoodie, and the storefront. Marka Hansen, then president of Gap North America, called it a logo the company had had for more than 20 years. CNN, the BBC and the Guardian used the same span. The public did not need a brand book to recognise it.
By 2010 the clothes behind that box were already moving. Hansen named the 1969 denim line and newer stores. That pressure is backdrop. It is not an invoice from a homepage that had not gone live yet.
Hansen's own explanation, written for the Huffington Post and quoted by CNN on 7 October, was that the brand and the clothes were changing, and the logo had to follow: "more contemporary and current," honouring "heritage through the blue box while still taking it forward." Louise Callagy, a Gap spokesperson, told Vanity Fair the shift was from "classic, American design" to "modern, sexy, cool."
A 20-year box is not a file you still own outright. The people who wear it own a share.
The decision: a homepage, no story
On Monday 4 October the blue box disappeared from gap.com. In its place: the word Gap in Helvetica, black, with a small gradient blue square parked on the p. The BBC named the typeface. James Yu, who built a parody generator that week, named it too. Laird & Partners in New York had worked the mark with Gap. Trey Laird had no comment.
Callagy told Forbes the introduction was quiet, "with minimal fanfare," on the website first. The plan, she said, was to use the new logo online and in holiday ads "before rolling out fully in 2011." The BBC noted that British customers would not have noticed, because the change had not been implemented in the UK. Stores were still the old box.
No advance campaign. No story in the window. No sample of the people who already wore the hoodie. A homepage, then a hope that "contemporary" would explain itself.
The week the internet drew 14,000 Gaps
The reaction did not wait for a holiday campaign. Within hours a parody Twitter account, @GapLogo, was speaking in the voice of the new mark. The Atlantic recorded thousands of followers in the first days and reported that Gap's press office told Ad Age it was "tracking" the account. Callagy told Forbes the company had nothing to do with it. The Guardian, on 12 October, put the protest account at nearly 5,000 followers.
Then the public started drawing. Yu launched Make Your Own Gap Logo on 6 October. By the 7th, Forbes had him at close to 5,000 generated marks. By the 9th, Yu himself wrote that the app had seen about 100,000 page views and almost 10,000 logos. A second generator, Crap Logo Yourself, ran in parallel. On 12 October the Guardian wrote that the "Make your own Gap logo" site had prompted nearly 14,000 parody versions.
Gap's first answer, on Facebook around 6 October, tried to rename the storm as a plan: the company loved its version, and would like to see other ideas. Designers told Forbes not to donate spec work. On the evening of 11 October, about 19:30 Eastern, Gap posted a surrender: they had heard, loud and clear, and the blue box was coming back that night.
Hansen's statement, carried by the Guardian, the New York Times and CNN, is the company's own autopsy: "we did not go about this in the right way." They had missed the chance to engage the online community. Crowdsourcing, she said, was not the right project at the right time. "There may be a time to evolve our logo, but if and when that time comes, we'll handle it in a different way."
The scoreboard
| Metric | Figure |
|---|---|
| Icon before | White GAP in a dark blue box, more than 20 years (~1990) |
| The swap | 4 Oct 2010 on gap.com: Helvetica wordmark, small blue square, no store rollout |
| Icon after | Revert announced 11 Oct 2010, US evening; blue box back "across all channels" |
| Time alive | 6 days / one week |
| Parody | @GapLogo, thousands / ~5,000 followers; ~14,000 generated logos |
The six-day window is a website swap that never reached stores. The 14,000 is a public parody count, not a Gap metric. About 100 million dollars is widely cited in later blogs and is not a company number; it stays off this table.
The autopsy: the typeface is the alibi
"Gap spent 100 million dollars on an ugly logo." You will hear that sentence in decks, and it is the half-true version that makes the case useless. The swap happened. The parody storm happened. The climb-down is in Gap's own words. Fold those facts into a taste argument about Helvetica and you will copy the failure the next time the typeface is better and the sequence is the same.
Look at what the numbers actually measure. The 14,000 logos are what a joke website produced in a week, not a research sample Gap commissioned. The 5,000 followers sat on a parody account the company did not run. The 100 million dollars does not appear in the Guardian, the BBC, CNN, the New York Times or Forbes from that week. Later branding blogs repeat "estimates" and point at each other. A mark that never reached the stockroom cannot have spent a rollout budget it never printed.
They call it a design problem. Look closer and it is a change-management problem wearing a wordmark. Hansen already had a reason: clothes were moving, the box felt old, younger shoppers were the target Callagy named. The reason can be real and the sequence still backwards. A 20-year icon is partly owned by the people who walk into the store. Drop a new file on them with no introduction, no story, and no rehearsal, and they will finish the design in public.
The later Facebook invitation, "share your designs," tried to invent a process after the file was already live. Designers read it as unpaid spec work. Customers read it as a company that had not decided. Crowdsourcing is not a rehearsal you start once the homepage has already changed.
Quibi spent 1.75 billion dollars on a product that had no meaning in people's lives. New Coke nearly broke a product whose meaning was larger than its taste. Bud Light spent a "beer for everyone" promise on a side. Gap is the fourth face of the same coin: a company that treated meaning as a file it could replace overnight. The typeface is what people pointed at. The missing step is that they were never asked to come along.
Bring people before you move the box
The lesson that sits in a six-day homepage is older than Twitter. Before you touch an iconic asset, treat it as something the public already holds. Identity work is a change-management job, held inside the marketing function, before anyone publishes a new file.
A marketing function, the kind we build, holds three answers before the designer exports: what the current mark means to the people who already buy, why this change is happening in language they would recognise, and where the first version will be tested so the core can reject it in private. If those three answers are not written down, the next homepage is a surprise with a logo on it.
That is why the portable object is not "never change a logo" and not "always crowdsource." Companies refresh marks on purpose. The ones that survive already know who will stay, and they have shown those people the story. The ones that guess, then apologise in a Facebook comment, pay twice: once in ridicule, once in the permission they just spent.
The right order: name, rehearse, then ship
None of this means a 20-year mark is frozen. Gap's own problem, named inside the company that week, was a brand that felt dated and a need for a younger shopper. The need can be real. The sequence was backwards. The right order has three steps.
Name the asset first. Write down, in language the core would use, what the current mark stands for and who already feels they own a piece of it. If the answer is "it looks old to us in the office," you do not have an answer yet. You have a mood.
Rehearse the change as a decision, not as a file drop. Show a sample of the paying base the new mark in the places they will actually meet it: a bag, a staff shirt, a single city, a landing page that still carries the old box in the corner. Watch whether they lean in or start drawing jokes. A landing page is cheap. A week of being the internet's sport is not.
Hold the gate, then pick the door. No homepage swap until the core has heard the story. If the flagship cannot carry a new face without looking like a different company, put the new face on a side door first: a product line, a pop-up, a capsule. Gap already had a 1969 denim line in the same conversation. The error was asking the 20-year box to absorb a new personality overnight.
The trap repeats at local scale every week: a cafe in Riyadh that changes its 12-year sign over a Thursday closure and is surprised when regulars walk past on Friday; a restaurant that drops its handwritten mark because a deck said "modern"; an app that ships a new icon on a Sunday and spends Monday explaining to users that it is still the same product. The budgets are smaller. The missing rehearsal is identical.
The takeaways
A mark people already wear is partly theirs. Changing it without them is a change of ownership, whether the brief called it a refresh or not.
The popular story will always name the typeface. The expensive story is the unpaid work you skipped: what the icon means, why it is moving, who gets to see it first.
Numbers that circulate after a fiasco are not a cost. About 100 million dollars is a later blog habit. Write "widely cited" on the slide. The invoice can wait for a filing.
Crowdsourcing after the file is live is not engagement. It is a retrofit. The core hears a company that had not decided.
Sequence beats taste. Name the asset, rehearse the change in the core's life, then decide whether the flagship or a side door carries the new face.
Frequently asked questions
Why did Gap change its logo in 2010?
Hansen said the brand and the clothes were changing and the 20-year box had to evolve with them: more contemporary, still carrying a blue square. Callagy tied the refresh to a younger target and to the 1969 line. The reason can be real. The sequence still failed.
How long did the new Gap logo last?
The file went on gap.com on 4 October 2010 and the revert was announced on the evening of 11 October. Contemporaneous rooms called it a week. The series counts six days. Stores and the UK never ran it.
Did the Gap logo redesign cost 100 million dollars?
No confirmed figure says so. The 2010 newsrooms we opened (Guardian, BBC, CNN, New York Times, Forbes) do not print 100 million dollars. Later blogs call it an estimate and cite each other. Gap did not disclose a cost. A website-only mark cannot carry a rollout invoice it never printed.
What did the 2010 Gap logo look like?
Helvetica, black, the word Gap rather than GAP, and a small gradient blue square on the p. Laird & Partners worked it with Gap. The old mark was white letters in a dark blue box.
What should a marketing director take from the Gap logo case?
Write down what the current icon means to the people who already buy, before anyone exports a new file. Rehearse that file in their life, not only in the office. If the flagship cannot carry the new face without looking like a different company, put the new face on a side door first.
Is this an argument against ever changing a logo?
No. It is an argument against changing one in the dark. If the core has heard the story and still wants the move, a refresh can be precise. If the core is "we in the meeting room," a refresh is a homepage with a comment thread attached.
This article is part of BMD's marketing case-study series. Episode 5 is an ad that lived 48 hours, then came down. That is the story of Pepsi.
Sources and further reading
Forbes Mike Isaac 7 Oct 2010; The Atlantic Alexis Madrigal 7 Oct 2010; James Yu viral app ~10,000 marks by 9 Oct 2010; Guardian Josh Halliday 12 Oct 2010; BBC Magazine Tom Geoghegan 12 Oct 2010; CNN John D. Sutter 12 Oct 2010; NYT Media Decoder Stuart Elliott 12 Oct 2010; Vanity Fair 12 Oct 2010.
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.