/qa-manual-istqb

Test Estimation (ISTQB Foundation Level)

Part of the qa-manual-istqb skill. See SKILL.md for full context.

Test estimation predicts the effort, duration, and resources needed for testing activities. Accurate estimates are crucial for planning, budgeting, and managing stakeholder expectations.

Why Estimation Is Difficult

• Requirements evolve during the project
• Defect counts are unpredictable
• Test environments may not be available
• Team experience varies
• External dependencies introduce uncertainty

Estimation Factors

Consider these factors when estimating test effort:

Estimation Techniques

1. Expert Judgment / Wideband Delphi

How it works:

1. Present the scope to multiple experts
2. Each expert estimates independently
3. Compare estimates and discuss differences
4. Repeat until consensus is reached

Best for: Early estimates, unique projects, when historical data is lacking

Tips:

• Use 3-5 experts with diverse perspectives
• Document assumptions behind estimates
• Track actual vs. estimated for future calibration

2. Historical Data / Analogous Estimation

How it works:

1. Find similar past projects
2. Adjust for differences (size, complexity, team)
3. Use actual effort from past as baseline

Best for: Projects similar to past work, organizations with good metrics

Example:

Past project: 500 test cases, 200 hours execution
New project: 750 test cases (1.5x)
Estimate: 200 × 1.5 = 300 hours
Adjust for: new technology (+20%), experienced team (-10%)
Final: 300 × 1.1 = 330 hours

3. Ratio-Based Estimation

How it works:

• Use industry or organizational ratios
• Apply to project parameters (LOC, function points, story points)

Common ratios (vary by context):

4. Work Breakdown Structure (WBS)

How it works:

1. Decompose testing into granular tasks
2. Estimate each task individually
3. Sum up all task estimates
4. Add buffer for coordination, dependencies, risks

Test activity breakdown:

Example WBS estimate:

5. Three-Point Estimation (PERT)

How it works:

1. Estimate optimistic (O), most likely (M), pessimistic (P) for each task
2. Calculate expected value: E = (O + 4M + P) / 6
3. Calculate standard deviation: σ = (P - O) / 6

Example:

6. Test Point Analysis (TPA)

How it works:

1. Identify test points from requirements (similar to function points)
2. Adjust for complexity, risk, and test type
3. Apply productivity factor (hours per test point)

Simplified formula:

Test Effort = Σ(Test Points × Complexity Factor) × Productivity

Estimation Worksheet Template

## Test Estimation - [Project/Feature Name]

### Scope

- Features in scope: [list]
- Test levels: [component, integration, system, acceptance]
- Test types: [functional, performance, security, etc.]

### Assumptions

- [ ] Requirements are stable (80%+ complete)
- [ ] Test environment available by [date]
- [ ] Team of [N] testers available
- [ ] [Other assumptions]

### Effort Breakdown

| Activity              | Hours | Notes |
| --------------------- | ----: | ----- |
| Test planning         |       |       |
| Test analysis         |       |       |
| Test design           |       |       |
| Test implementation   |       |       |
| Test execution        |       |       |
| Defect management     |       |       |
| Test closure          |       |       |
| **Subtotal**          |       |       |
| Contingency (\_\_\_%) |       |       |
| **Total**             |       |       |

### Duration

- Start date:
- End date:
- Calendar days:
- Working days:
- Parallel with development: [yes/no]

### Risks to Estimate

| Risk | Impact on Estimate | Mitigation |
| ---- | ------------------ | ---------- |
|      |                    |            |

Contingency Guidelines

Tracking and Refining Estimates

1. Track actuals: Record actual effort for each activity
2. Compare regularly: Earned Value, Burn-down charts
3. Re-estimate: Update estimates as scope clarifies
4. Post-mortem: Analyze variance and improve future estimates

Tracking metrics:

Common Estimation Mistakes