There is a moment most painters know well. You step back from a piece you have worked on for hours, and something is wrong. The drawing is accurate. The colours are pleasant. But the whole thing looks flat — like a decorated surface rather than a window into space. Nine times out of ten, the problem is value.
If you are still working out what value is, start with our guide to what value means in art. This article assumes you know the concept and picks up at the harder question: how do you actually check your values while you work, before the painting is too far gone to rescue?
Why your eyes lie to you about value
Human vision did not evolve to measure brightness. It evolved to identify objects, and it does that by aggressively correcting for context. A white shirt in deep shadow may reflect less light than a black shirt in sunlight, yet you will confidently perceive one as white and the other as black. Your brain is telling you what things are, not how light they appear.
This is enormously useful for walking around a room and catastrophic for painting. You put down a colour that matches what you know about the object rather than what your eye actually receives, and the result is a painting where every shadow is too light and every highlight is too dark. The range collapses, and the image goes flat.
Every technique below is a way of getting around this correction — of forcing yourself to see brightness rather than identity.
Method 1: Convert your reference to greyscale
The single most reliable check is to strip the colour out entirely. Once hue is gone, there is nothing left to distract you, and value relationships that were invisible in the colour version become obvious. Shadows you thought were dark turn out to be mid-tones. Two areas you painted differently turn out to be identical.
The trick is to look at the greyscale version side by side with the colour original, not instead of it. What you are training is the ability to look at the colour image and see the greyscale one underneath.
The free SketchKit Value Checker puts your reference photo and its greyscale version side by side in the browser. Nothing is uploaded — the conversion happens on your own device.
Method 2: Work from a limited value scale
Professional painters rarely think in a continuous gradient. They think in steps — usually somewhere between three and nine. A common approach is a nine-step scale where 1 is pure white and 9 is pure black, but for most work a five-step scale is enough: highlight, light, mid-tone, shadow, and dark accent.
Limiting yourself to discrete steps forces decisions. Instead of asking “is this a bit lighter or a bit darker,” you ask “is this a 3 or a 4?” That is a question you can actually answer, and it produces paintings with clear structure rather than muddy transitions.
Make a physical value scale on card, punch a small hole in the centre of each step, and hold it up against your reference. Look through the hole. Whichever step disappears against the photo is the value you want.
Method 3: Thumbnail in three values
Before committing to a full painting, spend five minutes on a thumbnail no larger than a matchbox, using only three values: light, mid, dark. No detail, no edges, no colour.
If a composition works at three values and two inches across, it will work at full size. If it does not, no amount of rendering will save it — and you have learned that for the cost of five minutes instead of five hours. This is the highest-return habit in representational painting, and almost nobody does it consistently.
Our image simplifier does the digital equivalent: it reduces a reference photo to a small number of flat value zones so you can see the underlying structure without the noise.
Method 4: Squint — but squint properly
Squinting is the oldest trick in the studio, and it works because narrowing your eyelids reduces the light entering the eye and blurs fine detail, collapsing the image toward its major value masses.
Most people squint too gently. You want your eyes almost fully closed, to the point where you can only just make out the shapes. If you can still read detail, you are not squinting hard enough. Squint at your reference, then immediately at your painting, and compare the pattern of light and dark masses rather than any individual passage.
Method 5: Photograph your work in progress
Take a photo of your painting with your phone and convert it to black and white. This does two things at once: it shows you your own values without colour, and the reduction in size collapses detail the same way a thumbnail does.
Photographing also breaks the familiarity trap. After two hours with a painting you stop seeing it — you see your intention for it. A photo on a screen is a different enough object that your critical eye switches back on. Flipping the image horizontally has the same effect, which is why so many artists work with a mirror at their side.
The four mistakes that flatten a painting
Shadows that are too light. By far the most common. Beginners are nervous about committing to dark, so every shadow lands in the safe middle. The fix is to find the darkest note in your reference, put it down at full strength early, and judge everything else against it.
Too many mid-tones. If most of your picture sits between values 4 and 6, it will read as grey mush no matter how accurate the drawing. Strong images usually have a clear majority value with smaller areas of contrast.
Ignoring value when choosing colour. A saturated colour can be any value. Painters often swap a colour for one they prefer without checking whether it sits at the same value, and the structure quietly falls apart.
Equal contrast everywhere. If every edge in the painting has the same value contrast, the eye has nowhere to rest and nothing to focus on. Save your strongest light-to-dark jump for the focal point.
A fifteen-minute practice routine
Do this three times a week for a month and your value judgement will change measurably:
- Minutes 0–3. Pick a reference photo. Make a three-value thumbnail from memory, without checking.
- Minutes 3–5. Convert the reference to greyscale and compare. Note every place you were wrong — and by how much.
- Minutes 5–12. Redo the thumbnail, now with five values, correcting what you missed.
- Minutes 12–15. Write one sentence about the error you keep repeating. Most people have exactly one, and naming it is most of the cure.
Value is not a talent you either have or do not. It is a measurement skill, and measurement skills respond to feedback loops. The tools here exist to shorten that loop — to tell you within seconds whether the value you just judged was right, instead of leaving you to find out three hours later when the painting has already gone flat.