# Structured Problem Solving — Define the Problem, Break It Down, Test Ideas and Find Root Causes

Source: https://skills101.in/learn/professional-skills/structured-problem-solving — Skills101 free tutorial (Professional Skills). Updated 2026-11-21.

In an exam, the problem is given to you clearly. At work it arrives as "the app feels slow", "customers are unhappy" or "sign-ups dropped this week". Nobody tells you which chapter it belongs to. People who can turn a vague, messy problem into clear steps are trusted with bigger responsibility — in engineering, product, data or business roles.

This lesson gives you a five-step method you can use for bugs, projects, placement case questions and even personal decisions.

## 1. Define the problem precisely

Most wasted effort comes from solving the wrong problem. Before doing anything, write the problem in one sentence that is **specific and measurable**:

| Vague | Precise |
|---|---|
| "The app is slow." | "The product page takes 6 seconds to load on mobile since Monday's release; it took about 2 seconds before." |
| "Students don't use the site." | "Of students who sign up, only 1 in 10 finish their first lesson within a week." |
| "My placement prep isn't working." | "I score below 50% in timed aptitude tests, mainly in time-and-work and percentages." |

Useful questions: **What exactly** is happening? **Since when?** **For whom** (all users or some)? **How much** — is there a number? **What does "solved" look like?**

## 2. Break it down with an issue tree

An **issue tree** splits one big problem into smaller parts that can each be checked. The parts should not overlap and, together, should cover the main possibilities.

![An issue tree: the problem "app sign-ups fell this week" splits into three branches — fewer visitors, visitors leave, or the form is broken — and each branch splits into checks such as ads paused, search rank, slow page, new design, error logs and one browser](https://skills101.in/learn-img/professional-skills/issue-tree.svg)

For "sign-ups fell", every lost sign-up must come from one of three places: fewer people arrived, people arrived but left, or people tried but the form failed. Each branch then becomes small questions you can actually check with data.

The same trick works for a coding bug: is the problem in the **input**, the **logic**, or the **output/display**? For a low aptitude score: is it **concepts**, **speed**, or **careless mistakes**?

## 3. Form hypotheses and test the likeliest first

A **hypothesis** is a specific guess you can prove or disprove: "Sign-ups fell because Monday's release broke the form on Safari." Good hypotheses are:

- **Specific** — they point to one cause.
- **Testable** — there is a quick check (logs, a test, a small experiment).

Order them by **likelihood** and **cost to check**. Check cheap, likely causes first: "Did anything change on Monday?" is a five-minute check that solves a surprising number of problems. Do not spend two days on an unlikely theory while an obvious one is untested.

## 4. Find the root cause with the 5 Whys

Fixing a symptom makes the problem come back. The **5 Whys** technique keeps asking "why?" until you reach a cause you can fix properly:

1. **Why** did the site go down? — The server ran out of disk space.
2. **Why** did it run out? — Log files grew to fill the disk.
3. **Why** did logs grow so much? — Old logs are never deleted.
4. **Why** are they never deleted? — Log rotation was never set up on the new server.
5. **Why** was it never set up? — The server setup checklist doesn't include it.

Deleting logs fixes today's outage. Adding log rotation **and** updating the checklist stops it happening again on every server. The number five is a guide, not a rule — stop when you reach something you can change.

## 5. Decide, act and check

Once you know the cause:

1. List 2–3 possible solutions.
2. Compare them on **impact**, **effort** and **risk**. A simple table is enough.
3. Pick one, act, and **measure** whether the original number improved.
4. Write a short note of what happened and what you changed, so the team learns too.

| Option | Impact | Effort | Risk |
|---|---|---|---|
| Roll back Monday's release | High | Low | Low |
| Rewrite the form | High | High | Medium |
| Add a "report a problem" link | Low | Low | Low |

Here, rolling back first and fixing properly later is the sensible order.

## 6. Communicate your thinking

When you report to a manager or answer a case question in an interview, lead with the **answer**, then the reasons:

> "Sign-ups fell because Monday's release broke the form on Safari. Safari users' sign-ups dropped to zero while other browsers were unchanged. I suggest rolling back today and adding a browser test before the next release."

This "answer first" style respects people's time and shows clear thinking.

## Common mistakes

- Jumping to a solution before the problem is defined.
- Testing the most interesting theory instead of the most likely and cheapest one.
- Fixing the symptom (deleting logs) and not the root cause (no log rotation).
- Changing many things at once, so you cannot tell what worked.
- Telling a long story before stating the conclusion.

## Quick recap

- **Define** the problem in one specific, measurable sentence.
- **Break it down** with an issue tree into parts that don't overlap.
- **Test hypotheses**, cheapest and likeliest first.
- Use the **5 Whys** to reach a root cause you can fix.
- **Decide** on impact, effort and risk, then measure — and communicate **answer first**.


Practice quiz and free certificate: https://skills101.in/learn/professional-skills/structured-problem-solving