Run a pilot program

You should run a pilot program to prove that your model works to yourself and your team. The pilot gives your team some experience; for many of them, it will be the first time they actually understand how it is meant to work and what they are supposed to be doing. If well done, the pilot program also builds credibility with your target community. This makes you well-placed to increase numbers later on.

The pilot program group needs to be the right size:

• It must be small enough for problems to be easily identified and resolved (that is, minimize risk).
• It must be big enough to be viable and to provide a wide enough experience for future development; you can't make generalizations about it if you have only two or three clients in it.
• The size is also determined by the kind of project. The larger and more institutionalized the project and its finances, the more formally your pilot program must be structured and controlled. But in a very small organization with relatively informal structures, your "pilot program" might be to examine specific ad hoc cases that you are already running, and pool your anecdotal evidence.

As you run the pilot and review it afterward, check these hints:

• Look for precedents. These can also be useful for gaining credibility later on and allaying fears of being the first to try something new. You can point to real examples.
• Look for glitches to fix.
• Look for trends and cultural patterns.
• Don't let individual personalities and unique circumstances overly influence your generalizations about how the program works.
• Did you set the right goals and objectives in your plan? You might find that you need to revise them. This involves a difficult judgment call, based on personal temperament:
• Some people want to abandon their plan as soon as something becomes difficult. The plan is actually working, but they are distracted by teething proplems.
• Others want to persevere with the plan even after it has clearly failed. They are so committed to their set of goals that they can't see the implementation.

In your review, assess the project against the planned goals and objectives. Ask everybody what they thought and what needs to be changed. Don't worry if some things didn't work. The point of the pilot is to run the program according to the best information available and find out what does and doesn't work and devise solutions.

If the results are not very encouraging, you might need to run a second pilot project to see if your solutions will work.