If you buy keyword specific domains, you're really buying the type in traffic.  I use the URL builder and redirect through that URL so you can see how much traffic your getting from the keyword domain.  There seems to be no rythme or reason to what keyword domains deliver traffic and what don't.  By tracking traffic with the Google URL builder you get a feel for what names are giving you traffic and which are not. ie. the plural, the singular, two words, three words, the possessive, etc.
Hi I totally agree with you Domaineer42. My partner and I are web developers and are seeing a lot of ‘hype’ selling and not enough ‘real value’ selling based on actual search engine criteria. Recently brokers have knocked back selling domains with great potential value based on some pie-in-the-sky logic, defying even Estibot values we’re told to use. The domains were not our best but with four figure value, according to Estibot, I thought it was a great start as a newbie to domaining. In an industry which should flourish, some of these domain ‘fashionistas’ are making rods for their own backs by devaluing domains that have REAL potential value based on customer search. I don’t understand what is going on, but I have decided to opt out for awhile. We have quite a few domains in our repository, with data backed search criteria, trending upwards. Will bring them out of the closet when the industry is ready to get real.
In theory, you could technically acquire a domain for $10,000 and be able to sell it for $100,000. You could also acquire a domain for $10 and be able to sell it for $100,000, $10,000… or $100, you get the picture. Because of this, it’s hard to recommend a perfect starting budget because your strategy, experience, and the pace at which you learn will hugely affect your buying and selling decisions.