Be Wary of “XP-Lite”

Sometimes there is no formal resistance from a manager. They simply state
boldly that XP sounds interesting, but that the organization can't afford to try it now.
"Maybe after we transition the wumbubbits to the hinklefarb platform," or some
other future milestone.
At this point, you will be tempted to pick a few XP practices and implement
them, rather than the whole ball of wax. There is nothing inherently wrong with this.
Maybe having more tests, or programming in pairs, or playing the Planning Game
will help flatten the curve you're currently on. Be careful.
Bringing in a single practice that is most likely to have an immediate positive
impact can make people more receptive to the next one. On the other hand, it might
make people think XP is just a bag of practices that can be chosen at random. We
have seen that XP is much more than the sum of its parts. All of the pieces work
together to produce something amazing.
Many have had success introducing XP one practice at a time. We recommend
that you introduce it in "chunks." Introduce at least several practices at a time, and
don't let more than a week or two pass before introducing the next chunk. If you do
it this way, people will recognize how each of the practices supports the others. If
you do it one at a time, your project could be way out of balance.
You should certainly be smart about it. The end of a project might not be the
best time to introduce a full-blown planning game. You probably don't want to pair
your database guru up with someone who can't spell database a week before adeliverable. You certainly don't want to refactor too much without first having a
critical mass of tests around the stuff you are refactoring. But, we can't think of a
time during development where writing a unit test before I write new code would be
a bad thing. We don't know when the customer shouldn't be in charge of setting
priorities.
If you are satisfied with having your process "stink less" and not being as far
from the best as you used to be, introduce a new practice every year. If you want to
play to win, don't play around.