Technique guide
Grid method for portrait drawing: a structure-first approach
The grid method breaks an intimidating portrait into small, checkable units. It is one of the oldest accuracy tools in art — and it works best when you pair it with an understanding of the head as a three-dimensional form, not as a flat pattern of cells.
What the grid method actually does
Place a grid of equal squares over your reference, draw the same grid lightly on your paper, and copy the image cell by cell. Inside a single square, the corner of an eye stops being “an eye” — a symbol your brain wants to draw from memory — and becomes a simple dark shape crossing a line at a measurable point. That shift from symbols to shapes is why grids improve accuracy so quickly, especially for likeness-critical work like portraits.
Five hundred years of gridded references
Proportion grids are far older than photography. Ancient Egyptian workshops laid out figures on wall grids to keep a consistent canon of proportions. In 1435, Leon Battista Alberti described the velo — a translucent gridded veil the painter looks through to map what they see onto a gridded panel. Albrecht Dürer published woodcuts of his drawing machines in 1525, showing artists sighting a subject through a gridded frame. The same idea persisted through academic ateliers into every modern art classroom: reduce a complex image to units small enough that your eye cannot fool you.
Grid method vs. Loomis and Asaro
The grid method and construction methods like Loomis and Asaro are often presented as rivals. They are actually the same instinct — decompose the complex head into manageable units — applied to two different problems:
Grid method — 2D mapping
Answers “where exactly does this shape sit?” Perfect for transferring a specific reference with a faithful likeness. It does not teach you why the features sit where they do.
Loomis / Asaro — 3D construction
Answers “why is this shape here, and where would it go if the head turned?” Perfect for drawing from imagination — but easy to get subtly wrong without a ground-truth check.
Used together, each covers the other’s blind spot: the construction tells you what to expect, and the grid verifies what is actually there. When your gridded observation contradicts your Loomis expectation, you have found exactly the spot where your head model needs correcting — the fastest kind of feedback an artist can get.
A structure-first grid workflow with FaceRef
- 1
Generate the exact reference you need
Instead of hunting for a photo that roughly matches, open the FaceRef workspace and set the precise head angle and light direction you want to practice. Export the Loomis or Asaro structure sheet — you now have a reference that already encodes the 3D logic of the pose.
- 2
Overlay a proportion grid
Run the exported image through the free grid maker. For a portrait, start with 6×6, keep the lines thin and around 60–80% opacity, and enable diagonals if you struggle with the tilt of the eye line.
- 3
Grid your paper and block in
Draw the matching grid lightly on your paper — scale it up or down freely, as long as the cell count matches. Block in the big structural landmarks first: skull mass, brow line, center line. Only then commit to features, checking each against its cell.
- 4
Wean yourself off the grid
As the structure becomes familiar, drop to a 4×4 grid, then to none — but keep checking your freehand studies against a fresh FaceRef reference at new angles. The goal is to internalize the head, not to depend on the scaffolding.
Frequently asked questions
What is the grid method in drawing?
The grid method is a transfer and proportion technique: you place a grid of equal squares over a reference image, draw a matching grid lightly on your paper, and copy the image one cell at a time. Because each cell contains only a small, abstract fragment of the picture, it is far easier to judge shapes and placements accurately than when looking at the whole face at once.
Is using the grid method cheating?
No. Artists have used mechanical proportion aids for centuries — Alberti described a gridded veil in 1435 and Dürer published designs for drawing machines in 1525. The grid trains your eye to compare positions and shapes precisely; most artists rely on it less as their observational accuracy improves. It only becomes a crutch if you never study the underlying structure of what you draw.
What grid size is best for a portrait?
A 6×6 or 8×8 grid suits most portrait work. Use fewer, larger cells to train your eye harder, or a denser grid when accuracy matters more than practice value — for example on a commissioned likeness. Diagonal guides inside each cell help pin down the exact tilt of the eyes and mouth.
How is the grid method different from the Loomis method?
They solve different problems. The grid method is a 2D mapping tool — it helps you copy what you see with correct proportions. The Loomis and Asaro methods are 3D construction systems — they help you understand the head as a volume so you can draw it from any angle without a reference. The strongest workflow combines them: construct to understand, grid to verify and transfer.
Can I combine the grid method with an AI or 3D reference?
Yes, and it is a powerful combination. Generate a 3D head reference at the exact camera angle and lighting you want, export it, then overlay a proportion grid before transferring it to paper. You get full control of the pose and a reliable accuracy check in one workflow.