“You only have to be right once.”
Drew Houston delivered this line in his 2013 commencement address at MIT. He described how his first company failed before Dropbox became the opportunity that changed his life. School can train people to treat every mistake as a permanent subtraction from a perfect score, but entrepreneurship works differently. Many attempts can fail without determining the final outcome if the founder continues learning and eventually finds a problem and solution with exceptional fit. Houston’s sentence offers relief from the fantasy of an unbroken record while preserving the obligation to keep searching.
Being right once is not the same as guessing once. Useful attempts create information about customers, technology, timing, distribution, and the founder’s own capabilities. Teams can limit the cost of each test, record what changed, and carry valid learning into the next version. They should choose questions that could genuinely alter a decision rather than stage demonstrations designed to confirm enthusiasm. A portfolio of disciplined experiments increases the chance that one success rests on understanding instead of luck and can survive beyond the first moment of attention.
The quote can be abused if it encourages reckless bets or ignores who bears the failures. Employees, customers, and investors do not have unlimited tolerance for a founder’s education. Safety, truth, consent, and solvency impose boundaries. A company also needs repeatable execution after the breakthrough; being right about one product insight does not guarantee durable operations, fair culture, or responsible scale. Failure deserves perspective, but it still requires accountability, honest communication, and closure when evidence says the next attempt cannot justify its cost.
The lesson is particularly helpful in markets filled with cheap AI prototypes. Teams can produce dozens of plausible demos, yet only a few solve a recurring problem well enough to change behavior. Leaders should define what “right” means through retained use, customer outcome, willingness to pay, reliability, and acceptable risk. Houston’s line encourages persistence without demanding perfection. The business task is to learn from many imperfect efforts, recognize the rare combination that truly works, and then build the systems needed to make that success trustworthy and repeatable. An experiment ledger can prevent the phrase from becoming a celebration of random betting. Each attempt records the question, exposure limit, result, insight, and decision it changed. Leaders review how learning carried forward and whether repeated failures share a cause that the team has avoided. When a success appears, the ledger helps distinguish the decisive insight from surrounding luck. That history supports the operational work needed to reproduce value after the exciting moment of being right.
Houston used the line in his 2013 MIT commencement address after describing an early company that failed.
He contrasted school’s perfect-score mindset with a career in which one consequential success can outweigh many wrong turns.
Disciplined experiments make failure informative by limiting cost and preserving learning for the next attempt.
Customers and employees cannot be disposable test material, and one breakthrough still needs responsible, repeatable execution.
AI makes demos cheap, but few products create a recurring outcome that changes customer behavior.
Retention, willingness to pay, reliability, and acceptable risk define whether a promising prototype is actually right.
Explore more "Quotes of The Day"
Discover more notable quotes from influential voices across politics, science, business, technology, sports, and culture. Each quote offers insight into how ideas, beliefs, and decisions shape the world around us.