Skip to main content

Reasoning

Deductive ReasoningInductive Reasoning
Valid or InvalidStrong or Weak
100% true or falseProbable and not probable
very sure as long as premises are true.cant make strong claims just general direction.
sound argument as long as all premises are true otherwise unsound argument.cogent vs not cogent.

Premise = Core Fundamental True Facts.

Frameworks​

IF → THEN → THEREFORE

  • IF these conditions are true
  • THEN this must happen
  • THEREFORE we conclude/implement this

Example:

  • IF an employee clocks more than 8 hours
  • THEN overtime rules apply
  • THEREFORE the system must calculate overtime pay

MECE (Mutually Exclusive, Collectively Exhaustive)​

When breaking down requirements, always ask:

  • Did I split everything into non-overlapping parts?
  • Did I include all possible cases?

Example (attendance rules):

  • Present
  • Sick leave
  • Vacation
  • Missing punch

If something overlaps (e.g., “late + absent”), your logic is not clean yet.

Exceptions​

Every requirement has two layers:

  1. Happy path (normal case)
  2. Exceptions (what breaks it)

Example:

  • Happy path: employee works scheduled shift → normal pay
  • Exception: overtime, absence, shift swap, public holiday

If you can’t list exceptions, you don’t fully understand the requirement yet.

4. The “Explain it to a 12-year-old” filter​

After you understand something, force yourself to restate it:

  • No jargon
  • No abbreviations
  • One idea per sentence

If it becomes unclear when simplified, that means your understanding is still incomplete.

5. The “decomposition ladder”​

Break problems down in levels:

  1. System goal
  2. Process
  3. Rule
  4. Data input/output
  5. Edge cases

Example:

  • Goal: calculate payroll correctly
  • Process: collect time data
  • Rule: apply overtime logic
  • Data: hours worked, contract type
  • Edge cases: missing punches, holidays

This is extremely useful in ERP / SAP / ASES-type environments.

6. Communication trick: “headline → detail → example”​

When explaining:

  • Headline: what it is
  • Detail: how it works
  • Example: concrete case

{% hint style="info" icon="eye" %} aggressively looking for what must be true, and what is missing. {% endhint %}