[jaggederest]

I wrote a blog post ( http://illicittech.blogspot.com/2007/05/sunk-cost-fallacy-and-dilemma.html ) about this, but basically, people are irrationally attached to things they've invested in. If it's (objectively) not working, put it aside and try something else.

[davidw]

Yes, but "not working" might be difficult to gauge if it's something that grows according to a curve. You might risk bailing out before the inflection point.

[jaggederest]

'put it aside' != bail out

I'm not saying you throw it out, just switch the majority of your effort elsewhere. Keep your demo/beta or whatever running, and if people start using it, you'll see that and can move back to it. Further investment is all you should worry about.