Scanning Technique

Scanning is the name for the systematic eye movement a good solver uses at the start of every puzzle. It doesn't rely on pencil marks, candidate lists, or advanced theory. It's just a way of looking at the grid so that the obvious placements — the ones that almost solve themselves — jump out quickly rather than being missed.

Experienced solvers can clear half of an Easy puzzle in under a minute using scanning alone. It's worth learning before any other technique because it gives you the grid-state that every other technique depends on. Once scanning stops producing placements, the more advanced moves step in.

The Three Directions

Every cell in a Sudoku lives in exactly three regions: a row, a column, and a 3×3 box. Scanning means looking in each of those three directions in turn. A digit cannot appear twice in any one region, so if a row already contains a 7 the other cells in that row are forbidden from holding a 7, and the same goes for the column and the box.

When you scan for a specific digit — say, 4 — you are asking a simple question: in this unit, which cells are still allowed to be 4? If only one cell remains, that cell must be 4.

How to Scan for a Single Digit

  1. Pick a digit. Start with one that already appears several times in the grid. The more copies there are, the more constraints you have and the easier it is to narrow things down.
  2. Mentally cross off forbidden cells. For each row containing that digit, every other cell in that row is forbidden. Do the same for columns and boxes.
  3. Look for a box with only one survivor. In a 3×3 box that doesn't already contain the digit, count how many cells are still allowed. If exactly one cell remains, place the digit there.
  4. Repeat. Move to the next box, then when you run out, switch to a different digit and start again.

Box-by-Box vs. Digit-by-Digit

There are two broad styles of scanning. Neither is strictly better — most solvers use both, depending on the puzzle.

Worked Example

Imagine a grid in which rows 1, 2, and 3 each contain a 3 — three copies spread across the top three rows. Between them, those rows rule out 3s in every cell they touch. The top-left 3×3 box, top-middle 3×3 box, and top-right 3×3 box each already have a 3 somewhere. That locks down a lot of information.

Now look at the middle band (rows 4, 5, 6). If row 4 already has a 3 and column 7 already has a 3, the middle-right 3×3 box has only a couple of cells left that can hold a 3. If only one of them is empty, you can place the 3 there. That's a scan-based placement: no pencil marks, no chain reasoning, just the intersection of three region constraints.

Tips for Faster Scanning

When Scanning Runs Out

Easy puzzles are usually solvable using scanning alone. Medium puzzles get you most of the way, then stall on a few cells that require more careful reasoning. Hard and Expert puzzles will stall much earlier.

When scanning no longer produces placements, the next tools to reach for are naked singles (a cell where all but one candidate has been eliminated) and hidden singles (a digit that can only go in one cell of a unit, even though the cell has other candidates). From there you move into pencil-mark-based techniques like naked pairs, locked candidates, and eventually the advanced patterns like X-Wing.

Practice

The only way to get fast at scanning is to do it. Start with an Easy puzzle and try to solve the first ten cells without pencil marks:

← Back to all strategies

Last reviewed on April 23, 2026.