Skip to content
Tutorials Center
My learning

Topics

Lesson 5 of 38 ยท Getting Started

A Web Testing Methodology

Learn a repeatable way to test a web app with Burp Suite: scope, map, analyse, probe, confirm and record, so you work systematically instead of randomly.

Opening Burp and "poking around" finds some bugs by luck. A methodology finds them on purpose. It gives you a checklist of areas to cover, an order to work in, and a way to know when you are done. Professional testers follow one even when they do not call it that.

This lesson sketches a simple, tool-friendly methodology you can reuse on every lab and authorised engagement. Later chapters fill in each stage with the matching Burp tools.

The six stages

flowchart LR
  A["1. Agree scope"] --> B["2. Map"]
  B --> C["3. Analyse"]
  C --> D["4. Probe"]
  D --> E["5. Confirm"]
  E --> F["6. Record"]
  F -.->|"new leads"| B

A loop, not a line: findings produce new things to map.

1. Agree scope

Settle who, what, when and how (see Legal and ethical testing). In Burp, mirror it in the target scope. Create a test account (or two) in the application if allowed; two accounts of the same role let you test whether users can see each other's data.

2. Map

Discover everything the app offers: pages, API endpoints, forms, parameters, roles, file uploads. Browse through the proxy as a normal user; use the crawler and content discovery where permitted. The result is the site map. Do not skip areas that "look boring": admin and legacy paths are often the least protected.

3. Analyse

Study what you mapped. Where does user input go? Which parameters look like identifiers (id=1002)? Which responses reveal technology versions or comments? Which features handle money, identity, files or permissions? The output is a list of hypotheses: "if I change this ID, can I read someone else's order?".

4. Probe

Test hypotheses one at a time, starting by hand in Repeater. Change one thing per request so you know what caused the result. Only scale up with Intruder or a scanner after a manual probe shows the idea has promise, and only within the agreed limits.

5. Confirm

A strange response is not yet a finding. Reproduce it, remove accidental causes, and ask: what is the real impact? Can a low-privileged user reach it? Is it a false positive from an automated check? Confirm with the least data needed.

6. Record

Write down the request, the response, the steps and the impact as you go. Burp helps: annotate history items, use Organizer, and save the project. Good notes turn into good reports later.

What to test: a checklist

The areas below are common categories, expressed as questions. They line up with well-known industry lists such as the OWASP Top 10.

Area Question to ask
Access control Can user A reach user B's data by changing an ID? Can a normal user call admin functions?
Authentication Are logins rate-limited? Do reset links expire? Is MFA enforced everywhere?
Session management Are session IDs unpredictable? Are they renewed after login and cleared on logout?
Input handling Is input validated and encoded on output? Do special characters cause errors?
Business logic Can steps be skipped or repeated? Can quantities, prices or coupons be abused?
Data exposure Do errors, comments or responses leak secrets or personal data?
Configuration Are security headers set? Are debug pages and old files removed?
Third-party components Are libraries current? Are API keys exposed in scripts?

Work in small, labelled steps

Good testers keep a log of what they tried and what happened. It prevents repeating work, shows coverage, and protects you if something breaks ("at 10:42 I sent request #318, which was a read-only GET").

Try it: a coverage tracker

A stand-in that models the checklist as data and reports which areas still lack a result, a trivial but effective way to see what is left.

Python Stand-in: a test coverage tracker

Press Run (or Ctrl + Enter) to see the result here.

Program input ยท what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

Try it: one change at a time

A second stand-in shows why changing only one thing per request matters. It generates single-field variants of a baseline request so each result can be traced to one cause.

Python Stand-in: single-variable test cases

Press Run (or Ctrl + Enter) to see the result here.

Program input ยท what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

When to stop

You are done with an area when you have a result for every checklist question that applies, you have followed up every lead, and you have recorded evidence. Time-boxed engagements rarely allow perfection, so prioritise: access control, authentication and anything touching money or personal data come first.

Check your understanding

Quick check ยท 4 questions

of 4 answered You scored / 4 ยท perfect!

  1. 1. Why change only one thing per request when probing?

    Single-variable tests keep cause and effect clear, which also makes findings easier to reproduce and explain.
  2. 2. What is the main output of the 'Analyse' stage?

    Analysis turns the map into concrete questions such as 'can I read another user's order by changing this id?'.
  3. 3. Why test with two accounts of the same role (where allowed)?

    Horizontal access-control flaws are found by using one account to request resources that belong to the other.
  4. 4. When should you scale up with Intruder or a scanner?

    Manual probing first avoids noisy, pointless traffic. Automation then multiplies a result you already understand.

Summary

  • A methodology turns random poking into coverage: scope, map, analyse, probe, confirm, record.
  • Turn what you see into hypotheses, and test them one change at a time.
  • Use a checklist of areas (access control, authentication, sessions, input, logic, exposure, configuration, components).
  • Confirm before reporting and keep notes as you go.
  • Prioritise areas that touch money, identity and personal data.

Next: Installing Burp and the first launch