I’ve heard dozens of good reasons for time-boxing iterations, or Sprints, since I started reading up on agile methodologies, but found some real food for thought in Craig Larman’s book Agile & Iterative Development.
Parkinson’s law states that “Work expands so as to fill the time available for it” … something I’ve definitely experienced myself very recently! The flip side is that where there is a specific timebox, work can and will fit in (up to a point, obviously!). Personally, I know this is true for me – I work faster and am more motivated if my work has deadlines.
Expanding work wasn’t such a new concept for me, but the second point was something I’d never considered before: “People remember slipped dates, not slipped features”. Deliver on time with 75% of the features, and the project will be remembered as a success. Deliver a month late with 100% of the features, and it will be considered a failure. Strange quirk … but I can identify with it!