mathematics · beginner · 10 min read · Updated Nov 5, 2026

Matrix Multiplication: Geometry, Dot Products, and Transformations

Master matrix multiplication through row-by-column dot products, dimensional compatibility rules, and interactive visual step tracing.

By Dr. Clara Lindqvist (Head of Applied Mathematics)

Learning Objectives

•
Determine whether two matrices can be multiplied by checking inner dimensions.
•
Compute individual product entries using the row-by-column dot product formula.
•
Understand why matrix multiplication is associative but non-commutative ( $A B \neq = B A$ ).
•
Interpret matrix multiplication geometrically as compositions of linear transformations.

Matrix multiplication is the fundamental engine of modern linear algebra, computer graphics, quantum mechanics, and neural networks. Rather than multiplying element-by-element, matrix multiplication combines rows and columns to represent compositions of linear mappings.

Why This Concept Matters

When you rotate a 3D model in a video game, compute PageRank on the web graph, or train an attention layer in a large language model, you are evaluating billions of matrix multiplications. Understanding how rows interact with columns builds intuition for higher-dimensional geometry.

Key Terminology

Dimension Compatibility

Two matrices

A

and

B

can be multiplied if and only if the number of columns in

A

equals the number of rows in

B

Dot Product

The sum of the products of corresponding elements from a row vector and a column vector.

Non-Commutativity

In general, matrix multiplication order matters:

A B \neq = B A

even when both products are defined.

Identity Matrix (

I

)

A square matrix with ones on the main diagonal and zeros elsewhere; acts as the multiplicative neutral element (

A I = I A = A

1. The Dimensional Compatibility Rule

Before attempting to multiply two matrices, you must check their dimensions. If matrix

A

has dimensions

m \times k

(meaning

m

rows and

k

columns) and matrix

B

has dimensions

k \times n

(

k

rows and

n

columns), their product matrix

C = A B

has dimensions

m \times n

(m \times k) \cdot (k \times n) = (m \times n)

The inner dimensions (k) must match exactly; the outer dimensions (m × n) dictate the resulting matrix size.

(1)

A \in R^{m \times k}

— Matrix A has k columns

(2)

B \in R^{k \times n}

— Matrix B has k rows

(3)

C = A B \in R^{m \times n}

— Inner dimension k cancels out

Common Beginner Mistake

Matrix multiplication is not element-wise multiplication (which is called the Hadamard product

A ⊙ B

). You cannot multiply two

2 \times 3

matrices using standard matrix multiplication because the inner dimensions

3

and

2

do not match.

2. The Row-by-Column Dot Product Formula

Each element

c_{ij}

in row

i

and column

j

of the product matrix

C

is obtained by taking the dot product of row $i$ of matrix $A$ with column $j$ of matrix $B$ .

Definition · Entry-wise Product Definition

For

A \in R^{m \times p}

and

B \in R^{p \times n}

, the entry in row

i

and column

j

C = A B

c_{ij} = a_{i 1} b_{1 j} + a_{i 2} b_{2 j} + \dots + a_{i p} b_{p j}

c_{ij} = k = 1 \sum p a_{ik} b_{kj}

Intuition: Slide your left hand across row i of A while sliding your right hand down column j of B, multiplying paired numbers and adding the products together.

3. Interactive Visual Step Tracing

Use the interactive demonstration below to step through each entry calculation. Notice how row highlights in Matrix

A

and column highlights in Matrix

B

intersect to construct each specific element in product Matrix

C

Matrix Operation

Interactive 2×2 Matrix Multiplication

Interactive

Matrix A (2×2)

Matrix B (2×2)

Product C (2×2)

Calculating Element C[1, 1] — Step 1 of 4

1 × 5 = 5 +2 × 7 = 14 = 19

Row 1 of A dot Column 1 of B: (1 × 5) + (2 × 7) = 5 + 14 = 19

Step 1 of 4

Click any cell in Product Matrix C or use Next Step to trace the exact arithmetic terms contributing to that entry.

4. Worked Example: Step-by-Step Multiplication

Worked Example: Multiplying a 2×3 Matrix by a 3×2 Matrix

Problem:

Given

A = (20 1 - 1 34)

and

B = 13 - 1 201

, compute the product

C = A B

Verify Dimensions

Matrix A is 2×3 and Matrix B is 3×2. The inner dimension is 3 for both, so multiplication is defined. The result will be 2×2.

(2 \times 3) \cdot (3 \times 2) = (2 \times 2)

Compute Entry C[1, 1]

Dot product of Row 1 of A and Column 1 of B:

c_{11} = (2 \times 1) + (1 \times 3) + (3 \times - 1) = 2 + 3 - 3 = 2

Compute Entry C[1, 2]

Dot product of Row 1 of A and Column 2 of B:

c_{12} = (2 \times 2) + (1 \times 0) + (3 \times 1) = 4 + 0 + 3 = 7

Compute Entry C[2, 1]

Dot product of Row 2 of A and Column 1 of B:

c_{21} = (0 \times 1) + (- 1 \times 3) + (4 \times - 1) = 0 - 3 - 4 = - 7

Compute Entry C[2, 2]

Dot product of Row 2 of A and Column 2 of B:

c_{22} = (0 \times 2) + (- 1 \times 0) + (4 \times 1) = 0 + 0 + 4 = 4

Conclusion:

The resulting product matrix is

C = (2 - 7 74)

5. Algebraic Properties of Matrix Multiplication

Associative Law: $A (BC) = (A B) C$ for all compatible matrices.
Distributive Law: $A (B + C) = A B + A C$ and $(A + B) C = A C + BC$ .
Scalar Compatibility: $c (A B) = (c A) B = A (c B)$ for any scalar $c \in R$ .
Transpose Product Reversal: $(A B)^{T} = B^{T} A^{T}$ (the order of factors reverses upon transposition).

6. Check Your Understanding

Check Your Understanding

If matrix

P

has dimensions

4 \times 2

and matrix

Q

has dimensions

2 \times 5

, what are the dimensions of the product matrix

PQ

Flashcard Deck · standard mode

Matrix Multiplication & Linear Transformations

Reinforce inner dimension compatibility, row-by-column dot products, geometric transformations, and non-commutativity with spaced repetition.

Study Deck Quiz Mode

Cards →

Research & Foundations

Academic References & Verification

Foundational research, standard mathematical treatises, and verified peer-reviewed literature supporting this topic.

Retrieval Practice Produces More Learning than Elaborative Studying with Concept Mapping

Jeffrey D. Karpicke & Janell R. Blunt (2011). Science, 331(6018), 772–775.

Demonstrates that retrieval practice outperforms elaborative concept mapping for conceptual retention and inferential problem solving.

Does not imply concept mapping is without value; hierarchical mapping excels at organizing domain schemas, but active retrieval is required to retain them.

doi:10.1126/science.1199327 Open access Primary source

Matrix Multiplication: Geometry, Dot Products, and Transformations

Learning Objectives

Why This Concept Matters

Key Terminology

1. The Dimensional Compatibility Rule

2. The Row-by-Column Dot Product Formula

3. Interactive Visual Step Tracing

Interactive 2×2 Matrix Multiplication

4. Worked Example: Step-by-Step Multiplication

5. Algebraic Properties of Matrix Multiplication

6. Check Your Understanding

Matrix Multiplication & Linear Transformations

Academic References & Verification

Retrieval Practice Produces More Learning than Elaborative Studying with Concept Mapping

Recommended Further Reading

The Derivative: Instantaneous Rate of Change and Tangent Lines

Binary Search: Logarithmic Divide-and-Conquer Search

Learning Objectives

Why This Concept Matters

Key Terminology

1. The Dimensional Compatibility Rule

2. The Row-by-Column Dot Product Formula

3. Interactive Visual Step Tracing

Interactive 2×2 Matrix Multiplication

4. Worked Example: Step-by-Step Multiplication

5. Algebraic Properties of Matrix Multiplication

6. Check Your Understanding

Matrix Multiplication & Linear Transformations

Academic References & Verification

Retrieval Practice Produces More Learning than Elaborative Studying with Concept Mapping

Recommended Further Reading

Related Learning Resources

The Derivative: Instantaneous Rate of Change and Tangent Lines

Binary Search: Logarithmic Divide-and-Conquer Search