Estimating Work
Estimation Approaches
Story Points
Relative complexity, not time.
| Points | Complexity |
|---|---|
| 1 | Trivial, well understood |
| 2 | Simple, minor unknowns |
| 3 | Moderate complexity |
| 5 | Complex, some unknowns |
| 8 | Very complex, significant unknowns |
| 13 | Extremely complex, many unknowns |
| 21+ | Too big, needs decomposition |
T-Shirt Sizing
For high-level estimates.
| Size | Relative Effort |
|---|---|
| XS | Hours |
| S | 1-2 days |
| M | 3-5 days |
| L | 1-2 weeks |
| XL | 2-4 weeks |
Estimation Factors
Consider:
- •Complexity: How difficult is the problem?
- •Uncertainty: How much is unknown?
- •Effort: How much work is involved?
- •Risk: What could go wrong?
Estimation Techniques
Planning Poker
- •Present the task
- •Everyone selects estimate privately
- •Reveal simultaneously
- •Discuss outliers
- •Re-estimate if needed
Three-Point Estimation
code
Expected = (Optimistic + 4×Likely + Pessimistic) / 6
Reference Stories
Keep calibration stories:
- •"This 3-point story took 2 days"
- •"This 8-point story took a week"
Common Pitfalls
- •Anchoring: First estimate biases others
- •Optimism: Underestimating unknowns
- •Scope Creep: Original estimate doesn't match final scope
- •Ignoring Overhead: Code review, testing, deployment
Tips
- •Estimate in ranges, not points
- •Include buffer for unknowns
- •Track actual vs. estimated
- •Re-estimate when scope changes
- •Don't estimate in hours (use relative sizing)