Cheat sheet
Delivery metrics & estimation cheat sheet
The arithmetic of getting work done predictably — and why “just work harder” rarely helps.

Flow
- Little's law
- Cycle time = WIP ÷ Throughput. 20 items ÷ 5/week = 4 weeks.
- Flow efficiency
- Active time ÷ Elapsed time. Often under 20%.
- WIP limit
- Stop starting, start finishing.
- Bottleneck
- The slowest step sets the throughput of the whole system.
Estimation
- PERT
- (Optimistic + 4 × Likely + Pessimistic) ÷ 6.
- Reference class
- How long did similar projects actually take?
- Iron triangle
- Scope, time, cost: fix two, flex one.
- Critical path
- Longest chain of dependent tasks; sets the end date.
DORA metrics
- Deployment frequency
- How often you ship to production.
- Lead time for changes
- Commit to running in production.
- Change failure rate
- Deploys causing a failure ÷ Total deploys.
- Time to restore
- How long to recover from a failure.
Capacity and reliability
- Queue wait
- Grows like u ÷ (1 − u): 80% busy ≈ 4×, 90% ≈ 9×.
- Plan for
- 70–80% utilization, leaving room for the unplanned.
- Error budget
- (1 − SLO) × Period. 99.9% over 30 days ≈ 43 minutes.
- Brooks's law
- Adding people to a late project makes it later.
Try it — Utilization & queues
Queue wait factor
4.0×
Slack
20%
The wait factor is ρ ÷ (1 − ρ): how long work waits relative to how long it takes to do. From 80% to 90% utilization it more than doubles.
1.0×50%
1.5×60%
2×70%
4×80%
6×85%
9×90%
19×95%
Wait factor at each utilization level. The curve is why a team planned at 100% is slow to respond to anything unplanned.
Test yourself
Questions from the lessons behind this sheet, explained when you check them.
Learn the ideas behind it
- Flow, WIP & cycle time →To finish faster, start less. The arithmetic is not optional.
- Estimation & planning →An estimate is a forecast, not a promise — and a single number hides the uncertainty.
- Capacity & utilization →A team that is 100% busy is slow at everything. Slack buys responsiveness.
