I've never been happy with the standard EXP rewards system in D&D. Some time ago I saw an alternate system posted in the House Rules Forum and decided to manipulate it to my own liking. The results have been positive; less focus on killing things, more focus on character development.
The system works by awarding points for various aspects of play. Two macro categories, Group and Individual Rewards, are subdivided into various other categories, which are further broken down into specific rewards and their Point Value. The main categories are as follows:
----------------------------------------------------------
GROUP REWARDS
Combat (per combat, based on the encounters difficulty)
Information (Learning new information; unravelling or 'solving' some piece of plot or a mystery)
Goals (From short term / minor goals to major campaign goals)
Planning (Self explanatory - I like to make my players brainstorm!)
INDIVIDUAL REWARDS
Sacrifices & Challenges (facing greater challenges or putting yourself at risk; heroism)
Spellcasting (Effective and innovative use of spells
Thinking & Planning (solving some riddle or conundrum, or masterminding an elaborate plan)
Skills (making good use of character skills)
Character (exhibiting character quirks & flaws)
Goals (accomplishing minor or major individual goals)
Social (interacting with NPC's (or sometimes PC's) in character)
OOG (contributing OOG material to the game (painting miniatures, useful knowledge, etc.))
----------------------------------------------------------
After awarding points, I apply them to the following formula:
EXP = Points x (40 + (10 x PC level))
I calculate the Group Rewards and the Individual Rewards separately, using the average party level for the Group calculation. This allows the lower level PC's to move up faster, but doesn't completely stifle the advancement of the upper level PC's. This creates a nice progression, never too fast, never too slow, and the formula is easy enough to manipulate so that DM's can customize it to their preferred rate of advancement.
And, of course, I built a spread sheet to do all the calculating for me.
If anyone is interested in seeing more, let me know. I'll post whatever you like.
Dr. Dan