Experimenting is enjoyable, but it becomes more useful when you can tell what caused the difference between two pictures. If you change the viewpoint, exposure, focus and edit at once, you may produce something interesting without learning which decision mattered.
I like to give a short experiment one clear question. It leaves plenty of room for play, but gives the results something to be compared against. You can always follow an unexpected discovery after you have a reference.
Choose a question you can answer with pictures
“How can I become more creative?” is too broad for one session. “What changes when this familiar subject falls out of focus?” is manageable. So is “Does a lower viewpoint make the foreground useful?” or “How does a moving subject read at different shutter speeds?”
Write the question before making the variations. You are not committing to a scientific paper; you are giving yourself a reminder when an interesting scene tempts you to change direction halfway through.
Make a reference before pushing the idea
Take a straightforward frame that shows the subject clearly. Keep it alongside the experiments, even if it will never be published. It helps you identify what the variation added or removed.
For an abstract-light exercise, the reference might establish where the lights sit in the frame. Later versions can explore blur, movement or colour treatment. Those are separate questions. Trying each deliberately gives you more information than combining all three immediately.
Vary one thing where practical
Make a small sequence in which one chosen decision changes. Keep the framing similar when exploring focus. Keep the basic edit similar when exploring crop. Label the variants in a way that lets you remember the difference without reconstructing the whole session.
Real scenes do not hold still for your convenience. Light shifts, subjects move and people enter the frame. Note those changes rather than pretending every pair is a perfect comparison. If the conditions alter substantially, make a fresh reference and continue from there.
A simple exercise could use five frames: a clear reference, two modest variations, one exaggerated variation and a final attempt informed by what you noticed. The number is a convenient limit, not an optimum. The useful part is making each frame answer the same question.
Judge the effect at the right scale
Review the set as small images first. What changes about the overall structure, mood or readability? Then inspect a larger version for faults that affect the intended use. An effect that only becomes visible at an extreme zoom may contribute little to a small final image.
Ask what information disappears as well as what appears. Defocus may turn signs and buildings into colour and shape. Motion may make direction more obvious while hiding detail. Those can be strengths when they serve the idea, or losses when the viewer needs the missing information.
Keep a useful failure
Save at least one attempt that explains the limit you found. It might show the point at which blur makes the subject unreadable or a crop removes the relationship that held the image together. A polished final alone does not preserve that lesson.
Write a short note: the question, the strongest variation, the failure and what you would change next time. Avoid declaring a permanent rule from one scene. You have learned something about this arrangement under these conditions, which is enough to guide another attempt.
When you next make a photograph with a similar problem, you have a decision you can use deliberately. That is the practical value of experimentation: a larger set of choices you understand, rather than a folder of effects you happened to like.