Sudoku Grid Explained

A classic Sudoku grid contains 81 cells arranged in nine rows and nine columns. Those same cells are also divided into nine 3×3 boxes, so every cell belongs to three overlapping groups at once: one row, one column, and one box.

That overlap is the structure behind almost every Sudoku deduction. A digit can be legal in a cell only when it does not conflict with the cell's row, column, or box. As solving becomes more advanced, the terminology grows — houses, peers, bands, stacks, coordinates — but the underlying grid never changes.

This guide is the reference for that vocabulary.

How is a Sudoku grid structured?

A standard classic Sudoku grid is a 9×9 square containing:

• 81 cells;
• 9 rows;
• 9 columns;
• 9 boxes of 3×3 cells;
• 27 row/column/box groups in total.

Each row, column, and box contains nine cells. In a completed classic Sudoku, each of those groups contains the digits 1–9 exactly once.

The visible thicker lines usually separate the nine boxes. They are not decoration: boxes are one of the three structural constraints of classic Sudoku.

Cells

A cell is one square in the grid.

At any point in a solve, a cell can be:

• a given or clue supplied by the puzzle;
• an unsolved cell;
• a solved cell filled by the player or solver.

An unsolved cell may also display candidates or pencil marks — the digits that remain possible there according to the current grid.

A cell is the smallest unit of the board, but no cell is logically isolated. Every cell belongs simultaneously to a row, a column, and a box.

Rows

A row is a horizontal group of nine cells.

There are nine rows in a classic Sudoku. In a completed puzzle, each row contains every digit from 1 to 9 exactly once.

Rows are normally numbered from top to bottom:

r1, r2, r3 ... r9

If row 4 already contains a solved 7, no other cell in row 4 can contain 7.

Columns

A column is a vertical group of nine cells.

There are nine columns, and each completed column also contains the digits 1–9 exactly once.

Columns are normally numbered from left to right:

c1, c2, c3 ... c9

If column 7 already contains a solved 5, every other unsolved cell in column 7 can rule out 5.

Boxes, blocks, and regions

A box is one of the nine 3×3 areas marked by stronger grid lines.

Each box contains nine cells and must contain the digits 1–9 exactly once when the puzzle is solved.

Different Sudoku resources use different words for the same classic 3×3 structure:

• box;
• block;
• 3×3 region.

VeyraPlay uses box as the default term for classic Sudoku, while recognizing block and region as common aliases.

In variant Sudoku, the word region can sometimes describe a non-rectangular area, so context matters. On a standard classic grid, however, box, block, and 3×3 region normally refer to the same thing.

Units and houses

A unit is a collective term for any complete row, column, or box.

Many advanced Sudoku resources use house in the same way. In that terminology, a classic grid has:

• 9 row houses;
• 9 column houses;
• 9 box houses;
• 27 houses in total.

For VeyraPlay, unit is often clearer in beginner material, while house becomes useful when reading advanced technique descriptions and external Sudoku literature.

Instead of repeatedly saying:

this deduction works in a row, column, or box

an advanced explanation can say:

this deduction works in a house

The important point is the constraint: a digit may appear only once in any one of those groups.

Peers and cells that “see” each other

Two cells are peers when they share a row, column, or box.

Advanced explanations may also say that two cells see each other. The meaning is the same: because the two cells share a house, they cannot both contain the same solved digit.

If r4c7 contains a solved 8, every unsolved peer of r4c7 can eliminate 8 from its candidates.

This idea becomes increasingly important in candidate eliminations, Wings, Coloring, Chains, and many other techniques. When a guide says that a target cell “sees both endpoints,” it is describing the peer relationship created by the grid.

How many peers does each cell have?

In classic 9×9 Sudoku, every cell has 20 distinct peers.

A cell shares:

• its row with 8 other cells;
• its column with 8 other cells;
• its box with 8 other cells.

Simply adding those gives 24, but four cells have been counted twice: two share both the same row and box, and two share both the same column and box. After removing those overlaps, 20 unique cells remain.

You do not need to memorize the number 20 to solve Sudoku. What matters is being able to identify which cells share a constraint with the cell you are studying.

Givens, clues, and solved cells

A given is a digit supplied in the starting puzzle. The word clue is also common.

VeyraPlay prefers given in technical explanations because clue can also mean a hint in other puzzle contexts. Both terms are valid when the meaning is clear.

A useful distinction is:

• given / clue — part of the original puzzle;
• solved cell — a value deduced during play;
• unsolved cell — a cell whose final value is not yet known.

Once a digit is correctly solved, it participates in row, column, and box constraints in exactly the same way as a given.

Sudoku coordinates and r1c1 notation

Coordinates let a guide identify a cell precisely without relying only on a picture.

VeyraPlay uses row-column notation:

• r1c1 = row 1, column 1;
• r4c7 = row 4, column 7;
• r9c9 = row 9, column 9.

The notation scales to groups as well. Some advanced sources compress repeated coordinates — for example, r57c2 can mean r5c2 and r7c2 — but VeyraPlay can spell cells out when clarity matters more than compact notation.

A beginner never needs to memorize coordinate notation merely to play. It becomes valuable in worked examples because the highlighted board can show the pattern while the text names the exact cells involved.

How are Sudoku boxes numbered?

When box numbers are useful, the usual convention is to number them from left to right and top to bottom:

• boxes 1–3 across the top;
• boxes 4–6 across the middle;
• boxes 7–9 across the bottom.

So the central 3×3 box is box 5.

Not every explanation needs box numbers. A visual highlight or phrases such as top-left box are often easier for beginners, while numbered boxes help in compact technical notation.

Bands and stacks

Sudoku solvers sometimes group three adjacent boxes together.

A band is a horizontal strip of three boxes covering three complete rows. A stack is the corresponding vertical strip of three boxes covering three complete columns.

That gives:

• 3 horizontal bands;
• 3 vertical stacks.

Some specialist sources use additional terms such as chute, floor, or tower for related three-box structures. These terms are not required for ordinary solving, but bands and stacks become useful when discussing grid transformations, puzzle generation, and some advanced patterns.

How rows, columns, and boxes intersect

The power of the Sudoku grid comes from overlap.

Take an empty cell at r4c7. To decide whether digit 5 is currently legal there, check:

1. row 4;
2. column 7;
3. the box containing r4c7.

If any one of those three houses already contains a solved 5, then 5 is not a candidate for r4c7.

The same intersections also create deductions involving where a digit can still appear inside a house. For example, if every possible 6 in a box lies on the same row, that interaction can produce Locked Candidates.

So the grid creates logic in two directions:

• cell-first: which digits can go in this cell?
• digit-first: where can this digit go in this row, column, or box?

Both views come from the same overlapping structure.

Why the grid structure creates Sudoku logic

Every Sudoku technique ultimately exploits constraints imposed by the grid.

Singles prove a value must occupy one cell. Locked Candidates use an intersection between a box and a line. Subsets reserve several digits for several cells inside one house. Fish compare candidate positions across rows and columns. Links and Chains follow relationships created when candidates share cells or houses.

The techniques become more sophisticated, but they never escape the basic anatomy of the board.

Understanding cells, houses, peers, and intersections therefore gives you more than vocabulary. It gives you the coordinate system for every later Guide.

Sudoku grid terminology at a glance

FAQ

How many cells are in a Sudoku grid?

A standard 9×9 Sudoku contains 81 cells.

How many rows, columns, and boxes are in Sudoku?

Classic Sudoku has 9 rows, 9 columns, and 9 boxes. Together those form 27 row/column/box houses.

Is a box the same as a block?

In classic Sudoku, usually yes. Box, block, and 3×3 region commonly describe the same 3×3 area. VeyraPlay uses box by default.

What is a house in Sudoku?

A house is a row, column, or box — any group in which a digit may appear only once. The term is especially common in advanced Sudoku explanations.

What is a peer in Sudoku?

A peer is a cell that shares a row, column, or box with another cell. Every standard 9×9 Sudoku cell has 20 distinct peers.

Are the thick lines part of the rules?

They visually mark the boundaries of the nine 3×3 boxes. Those boxes are part of the classic Sudoku constraints.

What are bands and stacks in Sudoku?

A band is a horizontal group of three boxes, while a stack is a vertical group of three boxes. They are useful structural terms but are not additional Sudoku rules.

What to learn next

Read Sudoku Rules if you have not already. Then continue to Sudoku Candidates to see how rows, columns, boxes, and peers determine which digits remain possible in every unsolved cell.