I've been in some classes recently and one of the lessons we learned about communication was: "don't guess". As technical professionals, we often think we know the answer to the problem someone is describing about halfway through the person's description, and we're quickly moving to the solution. If we don't know the solution, we're much more likely to guess and try things if we are experienced and intelligent in that area. In fact, the more familiar you are with something, the faster you'll guess at a possible solution. Guesses, while a great time saver, can also be very wrong. They can be used as a starting point for investigation, but not as the final answer.
I learned this lesson early on when I was in pilot training. There is a "pre-flight checklist" that after some time you just know by heart. It's a series of steps that guide you through making sure the aircraft is ready to fly. Sometimes you're so familiar with the checklist that you just blast right through it, not even looking down at the paper - but that's a mistake. Many times I've gone down the checklist and sure enough - there's something I would have missed. And missing an element in an aircraft pre-flight is inviting real disaster.
So the next time you're working through a database issue with someone, make sure you listen, quote back what you heard, and then work systematically through the "checklist" of troubleshooting. You'll make less mistakes, cause less stress, and show people that you value their time.