Home/ Techniques/ Junior/ Error Guessing
Junior · Experience-Based Technique

Error Guessing

Anticipate where the system is likely to break, based on experience and knowledge of common failure patterns. It's not random — it's formalised intuition.

Junior ISTQB CTFL 4.4.1 — K2 Understand ~12 min read + exercise

1 The Hook — The "Zero Dollar" Bug

A tester at a NZ e-commerce site was testing the cart. They used EP/BVA to test $1, $10, and $100. Everything passed. Then they used their gut.

They thought: "What happens if I try to buy something that costs exactly $0.00?" (like a free sample). They found that the payment gateway crashed because it couldn't handle a "null" payment amount. EP/BVA covers the requirements; Error Guessing covers the "What if?" scenarios that requirements often forget.

2 The Rule — The Fault List

Build a "Fault List" based on your experience of what usually breaks in similar systems.

Error guessing is about anticipating failures. Common "Guesses" include:

  • Empty inputs: Leaving fields blank.
  • Special characters: Using emoji, SQL strings, or punctuation.
  • Numeric extremes: 0, -1, or very large numbers.
  • Timing: Clicking "Submit" twice rapidly.

3 The Analogy — The Wellington Wind

Analogy

Building for the Wellington Wind.

If you build a house in Wellington, the "Requirements" might say it needs to be waterproof. But an experienced builder guesses that the wind will blow the rain sideways and upwards into the vents. They don't just follow the code; they use their experience of the local weather to add extra seals where they know the wind will hit. Error Guessing is adding extra seals to your software where you know the "wind" (users) will hit it hardest.

4 Watch Me Do It — The NZ Postcode Trap

Scenario: You're testing a 4-digit NZ Postcode field. EP/BVA says test 1010, 9999, etc.

  • The Experience Guess: "I remember a bug where postcodes starting with '0' were treated as octal numbers or dropped the leading zero."
  • The Test Case: 0612 (Waitakere, Auckland).
  • The Result: The system stores it as 612 (3 digits). Bug Found!
  • The Second Guess: "What about the highest possible 4-digit number?"
  • The Test Case: 9999 (Special postcode).
  • The Result: System crashes because it only expected up to 9800. Bug Found!

5 Decision Tool — Intuition vs. Random Guessing

How do you tell the difference between a "Pro" guess and a "Random" click?

✅ Professional Error Guessing

  • Based on a Fault List.
  • Targeted at High-Risk areas.
  • Uses knowledge of similar bugs.

--ink-2 Ad-hoc Testing

  • Random clicking.
  • No documentation.
  • No clear goal.

6 Common Mistakes

🚫 Only testing the "Obvious" stuff

I used to think: I'll just try an empty field and I'm done.
Actually: Error guessing should go deeper. Try things like "Paste a 2MB image into a text field" or "Disconnect the internet while the form is submitting."

🚫 Not keeping a Fault List

I used to think: I'll remember the bugs I've seen before.
Actually: Your memory will fail you. Keep a physical (or digital) list of "Bugs I've found in the past." This becomes your most valuable asset as a tester.

7 Now You Try — Build Your Fault List

🎯 Interactive Exercise

Scenario: You're testing a NZ phone number field (e.g., 021 123 4567).

List three "Guesses" you would try to break this field:

  • 1. Letters (e.g. 021-ABC)
  • 2. Leading +64 instead of 0
  • 3. Spaces and Brackets (021) 123...

1. Too many digits: 021 123 456789 (Buffer overflow check).

2. Only the prefix: "021" (Partial input check).

3. Special chars: 021/123/4567 (Delimiter handling).

8 Self-Check

Q1. Is Error Guessing a formal or informal technique?

It is considered an Informal (or Experience-based) technique. It doesn't have strict mathematical rules like BVA, but it is highly effective when done by experienced testers.

Q2. When should you perform Error Guessing?

After you've finished your scripted tests (EP/BVA). Use your formal tests to cover the specs, then use Error Guessing to "hunt" for the remaining bugs.

9 Interview Prep

"How do you approach a system with no documentation?"

Answer: "I combine Exploratory Testing with Error Guessing. I explore the system to learn the flows, and simultaneously use my 'Fault List' to guess where the weak spots might be — such as input validation, session timeouts, and error handling for external dependencies."