In June 1943, Lockheed engineer Clarence “Kelly” Johnson handed the U.S. Army Air Forces a working prototype of America’s first operational jet fighter, the XP-80, just 143 days after his team got the contract. The core was 23 engineers and draftsmen, supported by about 100 shop workers and mechanics. They worked inside a circus tent rented from a traveling carnival, pitched on a patch of Burbank dirt next to a plastics factory whose fumes gave the operation its eventual nickname: Skunk Works. The whole thing should not have worked. It worked so well that Johnson eventually wrote down the operating rules on a single sheet of paper, and those 14 principles became the gold standard for running a small, fast team inside a big, slow company.
The story matters now because every large organization claims to want what Johnson built and almost none of them can produce it.
What actually happened in Burbank
Lockheed had a problem. Britain had supplied the U.S. with its own de Havilland (Halford) H.1B Goblin jet engine, and the Army needed an American airframe built around it before German jets like the Me 262 ran the table over Europe. Johnson promised delivery in 180 days. He came in 37 days early.
The tent was not a publicity stunt. Lockheed’s main facility had no spare floor space, and wartime secrecy meant the project could not be logged into normal building access. Johnson rented the tent, set it up next to the Lockheed plastics plant, and told his engineers to bring their drafting tables. The smell from the plastics factory next door was bad enough that one engineer answered the field telephone by referencing Skonk Works from the Li’l Abner comic strip, where a character brewed something foul in the woods. The name stuck, though the spelling was later cleaned up to Skunk Works.
What the tent did was collapse the distance between the people designing the plane and the people building it. A draftsman could walk ten feet, hand a drawing to a machinist, watch the part get cut, and revise the drawing before lunch. There were no review boards. There was no procurement office. There was Kelly, and there was the airplane.
The 14 rules, in plain English
Johnson’s rules, written later and now archived by Lockheed Martin, are deceptively boring. The headline ones:
One. The Skunk Works manager must be delegated practically complete control of the program in all aspects. Two. Strong but small project offices must be provided. Three. The number of people having any connection with the project must be restricted in an almost vicious manner. Four. A very simple drawing and drawing release system must be provided. Five. There must be a minimum number of reports required, but important work must be recorded thoroughly. Six. There must be a monthly cost review covering not only what has been spent and committed but also projected costs to the conclusion of the program.
The rest covers contractor access, inspection authority, testing rights, specifications written in plain language, funding timeliness, mutual trust between customer and contractor, security, pay tied to performance rather than headcount, and direct access to the customer.
Read them once and they sound like common sense. Read them twice with any experience of corporate life and they sound like a declaration of war on how most companies actually operate.
Why the tent was the point
Modern environmental psychology has caught up with what Johnson seems to have understood instinctively. Recent analysis of physical space and creative thinking concluded that spatial layout, visual complexity, and proximity reliably shape how people think, focus, and solve problems together. The tent was everything essential within arm’s reach, nothing else permitted inside.
Johnson did not call it that. He called it keeping the engineers next to the metal. Same idea. When the people making decisions sit twenty feet from the people executing them, the feedback loop closes in minutes rather than weeks. When they sit in different buildings on different campuses reporting to different vice presidents, the loop closes in quarters.
Tweak Your Biz has covered this elsewhere in its work on why simplicity sells — the same principle that makes a product easy to buy makes a project easy to ship. Remove the layers between intent and action.
The corporate copy problem
Every Fortune 500 has tried to build a Skunk Works. Google had X. Lockheed itself still runs the original. Walmart has Store No. 8. The U.S. Air Force recently sought small businesses for rapid innovation support, asking how to actually do it. Johnson did it in less than five months.
The failure mode is almost always the same. Leadership announces a fast, autonomous team. Then HR insists on standard hiring procedures. Then legal insists on standard contract review. Then finance insists on standard budget cycles. Then the parent organization, sensing something is happening that it cannot see, demands monthly steering committee updates. Within eighteen months, the Skunk Works has a steering committee, a compliance officer, and a roadmap deck with 47 slides. Johnson’s rule three — restrict the number of people connected to the project in an almost vicious manner — gets violated before the first prototype ships.
What the rules actually require from leadership
A recent Forbes Business Council piece on the leadership paradox of innovation made the point that consistent innovation cannot rely on spare time or good intentions — it has to be built into the structure. Johnson’s rules are exactly that structure, stripped down. The hard part is not writing them. The hard part is the executive above the Skunk Works manager genuinely surrendering control.
That is why flat structures work in some companies and collapse in others. As Forbes Tech Council noted on flat organizational design, the form only holds when authority actually moves down with it. A flat org chart with a CEO who still approves every hire is just a tall org chart drawn sideways.
The same dynamic shows up in intrapreneurship programs — they succeed when the parent company protects the small team from its own immune system, and they fail when the immune system wins.
The speed dividend is real and measurable
Compressing development cycles is not a soft benefit. In medical devices, recruiting challenges and innovation gaps can delay critical product development. A small team with full authority and a tight loop sidesteps that problem entirely. There are fewer people to hire, fewer handoffs to coordinate, and fewer specialists whose absence stalls the whole project.
The P-80 went from contract to flying prototype with 23 engineers. A modern fighter program employs thousands. Some of that is real complexity. A lot of it is the cost of consensus.
What the tent teaches a manager today
The lesson is not to literally pitch a tent in the parking lot, although a few startups have tried it for the symbolism. The lesson is that the conditions Johnson created are reproducible in small doses inside almost any organization.
Pick one project. Give one person full authority over it. Cap the team at the smallest number that can possibly do the work. Put them in one room, physical or virtual, with everything they need and nothing they don’t. Cut the reporting requirements to a single monthly cost-and-status review. Refuse, in writing, to add anyone else to the distribution list. Pay the team based on what they shipped, not how many people reported to them.
None of this is comfortable. The person who used to approve their work no longer does. The departments that used to weigh in no longer get to. The parent organization will sense something happening it cannot see, and the pressure to add oversight will be constant. Johnson resisted that pressure for forty years, which is the only reason the U-2 and the SR-71 and the F-117 stealth fighter ever flew.
The tent came down decades ago. The plastics factory is gone. The rules are still pinned to walls in defense contractors, Silicon Valley garages, and the offices of executives who have figured out that the fastest way to build something is to get out of the way of the people building it. Eighty-three years later, that is still the part most companies cannot bring themselves to do.
