How to Solve It: Pólya's Problem-Solving Framework

The Book That Changed Problem Solving

In 1945, the Hungarian-American mathematician George Pólya published How to Solve It, a slim book that became one of the most influential works on mathematical problem solving ever written. Over a million copies have been sold, and its ideas have shaped how mathematics is taught worldwide.

Pólya's central insight was that problem solving is a skill that can be taught and learned, not just an innate talent. He distilled the process into four steps, each with a set of guiding questions.

This post explains Pólya's framework in detail and shows how to apply it to real mathematical problems.

---

The Four Steps

These steps seem obvious, but the power lies in the specific questions Pólya associates with each step. Most students fail not because they lack knowledge, but because they skip Steps 1 and 4.

---

Step 1: Understand the Problem

This is the most underrated step. Many students read a problem once and immediately start calculating. Pólya insists that you must first understand the problem completely.

Key Questions

• What is the unknown? What are you asked to find or prove?
• What are the data? What information is given?
• What are the conditions? What relationships connect the unknown to the data?
• Is it possible to satisfy all the conditions simultaneously?
• Can you restate the problem in your own words?
• Can you draw a figure or diagram?

Example

Problem. Prove that if $a_1, a_2, \ldots, a_n$ are positive real numbers with $a_1 a_2 \cdots a_n = 1$, then $a_1 + a_2 + \cdots + a_n \ge n$.

Understanding the problem:

• Unknown: We need to prove an inequality.
• Data: $n$ positive real numbers whose product is 1.
• Condition: The product constraint $a_1 a_2 \cdots a_n = 1$.
• Restatement: The arithmetic mean of positive numbers with product 1 is at least 1.
• Connection: This is the AM-GM inequality.

Already, just by restating the problem, we have identified what tool to use.

---

Step 2: Devise a Plan

This is the creative step. Pólya lists many strategies, and an experienced problem solver has all of them available:

Strategy 1: Look for a Pattern

Try small cases. If the problem involves $n$, try $n = 1, 2, 3$ and look for a pattern.

For our example: when $n = 2$, we need $a + b \ge 2$ given $ab = 1$. We can write $b = 1/a$ and check that $a + 1/a \ge 2$ for $a > 0$, which is equivalent to $(a - 1)^2 \ge 0$. This suggests the general proof might use a similar algebraic trick — or induction.

Strategy 2: Use a Related Problem

Have you seen a similar problem before? Can you use its result or method?

The AM-GM inequality states:

$$\frac{a_1 + a_2 + \cdots + a_n}{n} \ge \sqrt[n]{a_1 a_2 \cdots a_n}.$$

Since the product is 1, the right side equals 1, and we immediately get $a_1 + a_2 + \cdots + a_n \ge n$.

Strategy 3: Work Backwards

Start from the conclusion and ask: what would I need to prove this?

Strategy 4: Introduce Auxiliary Elements

Sometimes adding a new variable, function, or construction reveals the solution.

Strategy 5: Decompose and Recombine

Break the problem into smaller subproblems. Solve each one, then combine the results.

Strategy 6: Specialize, then Generalize

Solve a simpler version of the problem first, then adapt the method to the general case.

Strategy 7: Draw a Picture

In geometry, analysis, and many other areas, a good diagram can immediately suggest the proof.

Strategy 8: Consider Extreme Cases

What happens when a variable is very large, very small, zero, or equal to another variable?

---

Step 3: Carry Out the Plan

Once you have a plan, execute it carefully. Check each step.

Key Advice

• If you get stuck, go back to Step 2 — your plan might need revision.
• Write clearly. Even in scratch work, sloppy notation leads to errors.
• Check that each step follows logically from the previous one.

Example Continued

Let us carry out the AM-GM approach.

Proof. By the AM-GM inequality,

$$\frac{a_1 + a_2 + \cdots + a_n}{n} \ge \sqrt[n]{a_1 a_2 \cdots a_n} = \sqrt[n]{1} = 1.$$

Multiplying both sides by $n$ gives $a_1 + a_2 + \cdots + a_n \ge n$. $\square$

Alternatively, we can prove this by induction, which is more elementary.

Proof by induction. Base case: $n = 1$. If $a_1 = 1$, then $a_1 \ge 1$. ✓

Inductive step: Assume the result holds for $n - 1$ positive reals. Let $a_1, \ldots, a_n$ be positive reals with product 1. Without loss of generality, assume $a_1 \le 1 \le a_2$ (such a pair must exist since the product is 1 and not all terms can be strictly greater than 1 or strictly less than 1).

Define $b_1 = a_1 a_2$ and keep $b_i = a_i$ for $i = 3, \ldots, n$. Then $b_1 b_3 \cdots b_n = 1$ and by the inductive hypothesis:

$$b_1 + b_3 + \cdots + b_n \ge n - 1.$$

So

$$a_1 + a_2 + \cdots + a_n = a_1 + a_2 + (a_3 + \cdots + a_n).$$

We need to show that $a_1 + a_2 \ge 1 + a_1 a_2$. This is equivalent to $(1 - a_1)(a_2 - 1) \ge 0$, which holds because $a_1 \le 1 \le a_2$.

Thus $a_1 + a_2 + \cdots + a_n \ge 1 + a_1 a_2 + a_3 + \cdots + a_n = 1 + b_1 + b_3 + \cdots + b_n \ge 1 + (n-1) = n$. $\square$

---

Step 4: Look Back

This step is what separates good problem solvers from great ones. After solving the problem:

Check the Result

• Does the answer make sense?
• Can you verify it with a specific example?
• Can you derive the result differently?

Reflect on the Method

• Can you solve it a different way?
• Can you generalize the result?
• What was the key idea?
• Could you use this method on other problems?

For Our Example

• Check: For $n = 2$, $a = 2, b = 1/2$: sum is $2.5 \ge 2$. ✓
• Generalize: The result is the AM-GM inequality, which generalizes to weighted means.
• Key idea: The constraint "product equals 1" combined with the inequality between arithmetic and geometric means.

---

Pólya's Heuristics in Practice

Example: A Geometry Problem

Problem. Prove that in any triangle, the sum of any two sides is greater than the third side.

Step 1: Understand. We need to show $a + b > c$ for any sides $a, b, c$ of a triangle.

Step 2: Plan. This is the triangle inequality. A direct proof uses the fact that the shortest path between two points is a straight line. Alternatively, use the definition of a triangle: three non-collinear points.

Step 3: Execute. In a triangle with vertices $A, B, C$, the side $AB$ is a straight line segment of length $c$. The path from $A$ to $C$ to $B$ has length $a + b$. Since a straight line is the shortest path, $a + b \ge c$. Equality holds only if $A$, $C$, $B$ are collinear, which contradicts the triangle assumption. So $a + b > c$.

Step 4: Look back. This proves one instance; by symmetry, $a + c > b$ and $b + c > a$ as well. The key idea is the shortest-path property.

Example: A Number Theory Problem

Problem. Show that every integer greater than 1 has a prime divisor.

Step 1: Understand. For any $n > 1$, there exists a prime $p$ with $p \mid n$.

Step 2: Plan. Try strong induction. If $n$ is prime, we are done. If $n$ is composite, write $n = ab$ with $1 < a < n$, and apply the hypothesis to $a$.

Step 3: Execute.

Proof by strong induction. Base case: $n = 2$ is prime, so it is its own prime divisor.

Inductive step: Assume every integer $k$ with $2 \le k < n$ has a prime divisor. If $n$ is prime, then $n$ itself is a prime divisor of $n$. If $n$ is not prime, then $n = ab$ with $1 < a, b < n$. By the inductive hypothesis, $a$ has a prime divisor $p$. Then $p \mid a$ and $a \mid n$, so $p \mid n$. $\square$

Step 4: Look back. This is the first step toward proving the fundamental theorem of arithmetic. The key idea is reducing a composite number to a smaller factor.

---

Beyond Pólya: Modern Problem-Solving Resources

Pólya's framework has been expanded by many authors:

• Paul Zeitz, The Art and Craft of Problem Solving, emphasizes tactical and strategic thinking.
• Arthur Engel, Problem-Solving Strategies, catalogs hundreds of techniques with examples from mathematical olympiads.
• Terence Tao has written extensively on problem-solving strategies in his blog and in Solving Mathematical Problems: A Personal Perspective.

These books build on Pólya's foundation with more advanced techniques and harder problems.

---

Why Pólya's Method Works

The method works because it imposes structure on a process that otherwise feels chaotic. When you are stuck, instead of staring at the page, you have a checklist of specific actions to try.

Over time, the four steps become automatic. You will find yourself naturally asking "What is the unknown?" and "Have I seen something like this before?" without consciously thinking about Pólya.

That is the goal: to internalize the method so deeply that it becomes a natural part of how you think about mathematics.

---

Summary

---

References

• George Pólya, How to Solve It, Princeton University Press, 1945 (reprinted 2014).
• George Pólya, Mathematics and Plausible Reasoning (2 volumes), Princeton University Press, 1954.
• Paul Zeitz, The Art and Craft of Problem Solving, Wiley, 2017.
• Arthur Engel, Problem-Solving Strategies, Springer, 1998.
• Terence Tao, Solving Mathematical Problems: A Personal Perspective, Oxford University Press, 2006.