How do I actually get better at drawing?
Drawing a lot is not the same as practising. Improvement comes from working on one identified weakness at a time with a way of telling whether it worked.
- Difficulty
- beginner
- Time
- 20 min
- Read
- 2 min
Short answer
Pick one specific thing you cannot do yet — proportion, edges, values, ellipses — and work only on that for a session. Compare against the subject rather than against how you feel about the drawing. Repetition without a target reinforces what you already do, including the errors.
Most people who draw regularly for years plateau, and it is not for lack of hours. The hours go into producing finished pictures, which exercises what you can already do. Getting better means spending some of that time on the specific thing that is currently wrong, which is less enjoyable and far more effective.
Step by step
- Find the actual weakness rather than guessing.Put a drawing next to its subject, or hold it to a mirror. The mirror is unreasonably effective: it removes familiarity and errors in proportion and symmetry become obvious immediately.
- Work on one thing per session.Proportion, or edges, or value, or ellipses — not all of them. A session with one target gives you something to judge; a finished picture gives you an impression.
- Use short studies rather than finished pieces.Many quick attempts at the same problem teach more than one long attempt, because the feedback comes sooner and you get more repetitions of the decision you are practising.
- Measure rather than trust your eye, while learning.Comparative measurement — holding a pencil at arm's length to compare one length against another — feels mechanical and is how the eye is trained. It stops being necessary once the errors shrink.
- Draw from life as well as from photographs.A photograph has already flattened the subject and made decisions about value and edge. Life makes you make those decisions yourself, which is the skill being built.
- Keep the work, dated.Improvement is invisible week to week and obvious across six months. Keeping studies is the only reliable way to see it, and it is also how you find recurring errors.
- Copy deliberately, and analyse what you copied.Copying a drawing you admire is useful when you are asking why a particular edge or value choice works. It teaches nothing done passively.
- Accept that studies are allowed to be bad.A study exists to find out something. Judging it as a picture is the fastest way to stop doing them and go back to drawing what you can already draw.
Questions people ask
How long should a session be?
Short and frequent beats long and occasional. Twenty focused minutes on one problem several times a week does more than one long weekend session.
Should I use references?
Yes, and knowing what a reference has already decided for you is part of the skill. Photographs compress values and sharpen edges in ways that mislead if copied uncritically.