Independent resource / treukhov.com
A guide to the small visual study.
A practical sequence for turning a visual question into a set of versions you can compare. Use it with the tools and medium that fit the question.
1. Begin with an observation
Choose an image, object, arrangement or piece of your own work to examine. Describe the visible parts and the relationship that interests you before writing a judgement about quality.
If you use reference material, record its source and keep it clearly identified. Your study should remain distinguishable from the original work and from your description of it.
2. Write one question
Make the question narrow enough to inspect. You might explore how spacing changes hierarchy, how scale shifts balance or how a repeated element creates rhythm.
Avoid a brief as broad as making the whole piece better. Name the relationship you want to understand and what would count as a useful observation at the end.
3. Choose a constraint
Set the frame for the study: a format, palette, set of elements, type system or other relevant boundary. Choose a constraint because it focuses the question rather than because it sounds demanding.
Record what will stay stable across the versions. If you later change the constraint, note that decision so the comparison remains understandable.
4. Make a baseline
Build a simple version that contains the relationships you need to inspect. Keep enough context for the question to make sense and avoid spending the whole session on unrelated detail.
Save the baseline with a clear name. Add a short note about what you expect a change might reveal; treat that expectation as something to examine, not as a conclusion.
5. Make a meaningful variation
Change a relevant relationship and save the version beside the baseline. Where several elements need to change together, record the group of changes rather than pretending only one variable moved.
Keep the versions visible under comparable conditions. Describe the difference you can actually point to in the work, including effects you did not expect.
6. Review with a clear request
Ask a reviewer a question that matches the stage of the study. Invite them to describe where attention goes, how the parts relate or what feels ambiguous, then ask which visual features support that reading.
Keep their response as a perspective on the work. Compare it with your own observation and decide which point deserves another version.
7. Archive the learning
Save the question, constraint, versions and a brief review note together. Identify which material is your own and which comes from a reference. A useful record should make sense when you return to it later.
End by naming the next question. It may come from an unresolved relationship, a surprising variation or a change in context, rather than a need to keep polishing the same piece.