There are things I don't really love doing like ghostwriting company blogs but if the project is short and well-defined (and is something I can actually do a good job of fairly quickly), I'll take a few $K from you. I'd be much more hesitant to commit to something more open-ended even if the income stream were good if it were something I was doing strictly for the money.
The other thing I wanted to mention, aside from opportunity cost, is risk. I'm not suggesting never leave your sweet spot. That would be stupid on my part :)
However, as you leave your sweet spot, be aware (read: beware) of what you don't know. Be aware of the fact that bad decisions are very similar to trust...one slip and you've lost 10x more than you anticipated. If you're going to leave your sweet spot be sure you're prepared for the worse. It can and does happen.
Put in real terms, over the years I've worked as a contractor for a number of marketing / web dev agencies. The more successful ones didn't get there from yes, they got there because of no. No to projects beyond them. No to clients with expectations out of proportion to budget. No for the win, so to speak.
I've also seen those same agencies start to drink their own Kool Aid. They get in a financial pinch and take on work / clients they're not tuned for. For example, they don't have the in house expertise in the technology / solution required. Eventually the project costs more than it brings in *and* the team is miserable, has lost faith, leaves the conpany, etc.
If you're leaving your sweet spot and you're not concerned, you're probably doing it all wrong.
It’s a reflection of your perceived opportunity cost.
But you are right to suggest time-bounding your engagement up front to avoid getting stuck in a local maxima.
Or, at a certain scale, have a segregated set of people doing these requests where possible to limit and help govern the distraction.