Welton's rule: for each bit of trite business advice, there is an equal and opposite bit of trite business advice.
In fact, fail often fail quickly, if done effectively, will lessen the number of years to break even (IMO).
> Three years. Sometimes a bit less, sometimes a bit more.
3 years is pretty quick in terms of success, in terms of failure it is an eternity. It all depends on what you perceive the trend to be. If the trend 'flatlines' then you can draw a negative conclusion much earlier. After a year or less you can probably predict quite accurately what the future will look like.
Three years is pretty quick on the scale of a viable business.
In the early days, it is an eternity in terms of strategy when that strategy is failing.
In other words, three years may be an appropriate timeline in which to view overall sunk costs, but that doesn't mean that the business shouldn't be prepared to pivot several times within that period.
However, how does this fit into the "fail often, fail quickly" advice I have seen so often in this forum? Does quickly mean 3 years? Just a thought...