I'm a big fan of Eric Ries's MVP notion. I firmly believe that it's exceedingly rare that a startup or a product fails because it's missing that nth feature.
Feature creep is tempting in the early stages when the product is the one area of a business that the founders have close to full control on. And since founders tend to be very passionate and hard working, they feel that adding that one extra feature will differentiate their product and help them in going to market.
Optimizing features by applying the Pareto Principle will lead to a more effective use of resources. Nivi has a great recent post on this topic with some great ideas and examples. My favorites:
“If Apple can launch a smartphone without Find or Cut-and-Paste, what can you cut out of your product requirements?” – Sramana Mitra
“The first version of Gmail was literally written in a day.”