The oft-repeated wise saying is that you should only give someone X% of your business if they will increase its value by more than X%.
This is a bit of a trap, though, because it does not convey just how scrupulous you should be about doing that maths. "John is a great guy, I'm sure having him on board will increase the business's value by at least 5%, probably much more!" is a likely computation in the mind of the inexperienced founder.
One common trait between experienced entrepreneurs, as compared to the rookies, is that they treat equity like the blood, the lifeforce of the business. They only give it away grudgingly, to those who can prove beyond the shadow of a doubt (with either a large pile of cash, or measurable results), and in the smallest amount possible.
So, when hearing that piece of wisdom, what it really means is that you should only give X% of your business to someone after they have demonstrated beyond any doubt that they increase your business by more than that - not based on a guess.
Another interesting note is that one of the big reasons why freelancers do not want to be paid in equity is that any founder that has not yet learned to keep all their equity to themselves is unlikely to amount to anything (at least in this particular business venture).
Of course, the VC game described in this article changes things slightly. There, the input of the investor is measurable, and there are rules to optimise that process and tricks used to mess with valuations. Keep your wits about you, and don't fall for thinking that the sorts of equity stakes thrown around in the entertainment show "Dragon's Den" are anything to do with reality in the startup world.
If you read this far, you should follow my RSS feed here.