How long, really?

No track record yet? Break the work into tasks and give each a best, likely, and worst case. A Monte Carlo PERTA three-point estimating technique: it weights the likely case four times as heavily as the best and worst, so the average leans toward reality but the tails still count. simulation rolls them up into a finish time you can put a number โ€” and a confidence โ€” on.

estimates in
TaskBestThe optimistic case: how long it takes if everything goes right. A floor you'd rarely beat, not your usual estimate.LikelyThe most likely case โ€” what you'd bet on if you had to name one number. PERT weights this one most heavily.WorstThe pessimistic case: how long it drags out if things go wrong. A realistic bad day, not the apocalypse.
10,000 simulations
โ‰ค 21.7days

85% confident all 3 tasks finish within 21.7 days.

P85
10.1days29.5
ConfidenceFinish within
50%confident
โ‰ค 18.1 days
85%confident
โ‰ค 21.7 days
95%confident
โ‰ค 23.7 days
Seeded from your inputs โ€” this link reproduces the exact same forecast.