Disney's early years, and when talking first is the right move
A rough cut of Snow White persuaded a banker where a pitch could not. Disneyland, though, began with years of planning.
The way to get started is to quit talking and begin doing.
Walt Disney said something very close to this in 1957, talking to a newspaper columnist about the hardest part of his early years, which was getting anyone to lend him money. His recorded words were a memory of his own start, that he had found the way to get started was to quit talking and begin doing it; the version on posters drops the personal frame and turns it into a general rule.
The memory fits his record. His first studio, Laugh-O-Gram in Kansas City, went bankrupt in 1923. He moved to Los Angeles, set up again with his brother Roy, and spent the next fifteen years borrowing against work that was not yet finished. The clearest example came with Snow White and the Seven Dwarfs, which ran far over budget. The studio needed another loan, and Bank of America’s Joseph Rosenberg approved it after being shown a rough cut of the unfinished film. A pitch had not been enough, and a partly built film was.
That is the practical content of the line. People with money, customers and possible hires all respond more reliably to a thing than to a description of one. A founder who has spent three months refining a deck might learn more in three weeks of building a crude version and putting it in front of the people the deck was meant for. Doing also generates the specific questions that talk cannot, such as what breaks first and what it really costs.
Talk has a particular trap for founders, which is that it feels like work. Naming the company, arguing over a logo and debating the pricing page of a product that does not exist all give a sense of progress while producing nothing a customer could use. Disney’s studio in the 1920s had no room for that habit, because each finished short had to be sold before the next one could be paid for.
The line fails when doing is expensive to reverse. Disneyland is the counterexample inside Disney’s own career. Before the park opened in 1955 there were years of planning, and the Stanford Research Institute was hired to study where to put it and how many people would come; the Anaheim site came out of that study. Building first would have meant choosing land by instinct and discovering the mistake after the concrete was poured.
The same split applies to a startup. Software, content and small experiments are cheap to start and cheap to undo, and there the advice is almost always right. Leases, hardware runs, large hires and borrowed money are hard to undo, and a week of talking, with numbers, is cheaper than a year of recovery. Disney himself did both, in the order that suited what was at stake.