Ivo Dominguez Jr recently posted an article on how approach experimentation. I found myself nodding in some agreement with his points, especially when says that the development of anything new should be extensively tested by yourself before sharing it with anyone else. I also disagreed with one point he made:
There is a lot of perfectly good material available that can simply be followed step-by-step and produce great results. We must not place novelty and innovation above what is known to be efficacious...New does not always mean better or for that matter safe or potent.
He's right there is a lot of magical work out there already established that works well. However even if it is efficacious we should never use that as an excuse not to experiment. If anything, we might ask: How could I take and personalize and improve on this technique? And it could be argued that doing so is just reinventing the wheel, but at least it is your wheel at the end of the day. With all that said, I think a magician needs a solid foundation before experimenting, which means making time to learn and practice what has been done until you understand how it works. After that, experiment with it, and start developing your own techniques.
Anything you experiment with should be tested extensively by yourself. Work with it, tinker with it, and really get to know the process you are developing. Once you've spent enough time doing it, then see if others are willing to try out the technique/ Get their input and share you experiences as well. Ivo says the same thing in his article and its sound advice. You can't effectively ask people to experiment with your technique until you know that technique and can share your experiences...at the same time, I wouldn't share the experiences until after the people try working with the technique. Let them have their experiences without having your subjective report in front of them. The reason is simple. If they know what you experienced, it may influence how they experience the technique. The point is to allow them to have their own experience and then share yours. The sharing allows you to verify your technique with their experiences and your own.
I do believe experimentation should be done carefully in the way you'd learn a technique you read in a book. You read about a technique and you carefully practice each step of the technique, with an eye toward understanding how it is affecting you. For example, when I learn a new meditation technique, I'll read about the technique. Then I'll do one step and once I feel I understand it, I'll do another step, etc. The same applies to experimentation. You put together a process and then you work with each step carefully, making sure you understand the process and testing it to make sure it works. Once you've tested it enough you share it with others and see if the results are consistent. If they are you have a new technique that works, but don't sit on your laurels...see if you can improve it.