A 2009 article from Seth Godin’s blog recently went viral, Ignore Sunk Costs.
A sunk cost in the context of the sunk cost fallacy implies a certain amount of uncertainty of ever recovering the initial investment. Misspelling a sign is a mistake that needs to be fixed, not a sunk cost. For example, if a Subway franchise owner misspells Subway as two words, it would probably result in diminished sales but the owner knows that fixing the sign will boost sales and rectify the problem, so it’s not a sunk cost but rather just a stupid error and cost of doing business. But investing $2,000 in your uncle’s crazy business scheme that he insists “worked for someone else” but hasn’t made anything in return, is a sunk cost because there is no guarantee it will ever succeed, unlike the franchise, which is demonstrably profitable. Habitually buying lottery tickers in the hope of recovering past losses, having already wasted a bunch of money and committed oneself emotionally to playing, is committing the sunk cost fallacy.