If they do a lot of business, and buy more trucks and get more employees - and if not them, the competition after people discover that this is a viable business model - the 50% might lower with economies of scale giving them more room to take a reasonable profit.

Depends on how fast they can keep those trucks moving. I'd start investing in some math and software implementing that math in order to optimize those routes. You could tie it right into the auction website and give all the drivers webapps feeding them their next pickup/dropoff points. You would be able to charge based on distance/size calculated on the fly. I see room for making this really cheap. If you wrote good software for it, you could license the platform and not even actually do the business - this is a natural franchise.

Sorry, just rambling. I think this idea is great if done well.

Daylight robbery of whatever you're willing to put on a truck and blindly sell to the highest bidder in the next 90 minutes?

If you have something of value, chances are you're not going to be using the above.

Presumably people can or will be able to set up an RSS feed for items they're looking for (like a couch, bed, TV, etc.), and the truck company will standardize on terms to make it easy to see.

50% isn't that bad (or a whole lot) when a lot of the items won't be going for much anyway.

I'd see an upsell existing where you could pay $5 or $10 (or $1/lb) to extend the auction by, say, 24hrs.

The biggest problem I see is with a lot of leases ending at the change of the calendar month, there will be a surplus of items at the end of the month (as people have to leave), and a lot of demand at the beginning of the month (as people move in).

Furniture is very illiquid because of the transportation hassles. I bought three desks that had retailed for $500 each, for $33 each. That person would have been happy to get 50%.