Project Time CalculatorProject Time Calculator

How to Estimate How Long a Project Will Take

Five practical methods for estimated duration, with formulas, worked examples and the mistakes that make a date undefendable.

Quick answer

Estimated duration is planned effort divided by team capacity, rounded up. 320 hours with 2 people at 8 hours a day is 20 working days. If dates are uncertain use PERT; if tasks depend on each other use the critical path, not the sum of every task.

Open the project time calculator

What “how long will it take” actually asks

People asking how long a project will take usually want a finish date, not a pile of hours. Estimated duration is the working time the plan occupies. Effort is the volume of work. Mixing them up is the most common error in a time estimate.

Five methods cover almost every honest estimate you can defend: effort divided by capacity, three-point (PERT) estimates, the critical path through dependencies, a Monte Carlo range (P50, P80, P90), and an analogous figure from a past project. Use more than one when the work is new or the date is a commitment.

The core formula

daily capacity = people × hours per person per day × (availability % ÷ 100)

planned effort = effort × (1 + buffer % ÷ 100)

working days = ⌈ planned effort ÷ daily capacity ⌉

Count that many working days from the start date, skipping weekends and holidays, to get the finish date. The project time calculator does this, including person-days and person-weeks as effort units.

Worked examples

Each example uses numbers you can load into the linked calculator; the results match that tool exactly.

Example 1: effort and capacity

320 hours, 2 people, 8 hours a day, fully available, no buffer, starting Monday 2 November 2026.

  • Capacity = 2 × 8 = 16 hours per day
  • Duration = 320 ÷ 16 = 20 working days (4 weeks)
  • Finish date = Friday 27 November 2026

Load the same numbers in the project time calculator.

Load this example

Example 2: three-point estimate

A task could take 4 days at best, 6 most likely and 14 at worst.

  • Expected time TE = (4 + 4 × 6 + 14) ÷ 6 = 7 days
  • Standard deviation = (14 − 4) ÷ 6 = 1.67 days

TE is longer than the “most likely” 6 days because of the long pessimistic tail. The PERT calculator extends this to many tasks.

Load this example

Method 1: effort divided by capacity

This is the default for a single effort figure. Convert person-days or person-weeks to hours if needed, add a buffer, divide by people × hours per day × availability, and round up. 720 person-days with 5 people is 144 working days (28.8 weeks on a 5-day week). Use the person-days calculator when the estimate is already in person-days. Availability under 100% lengthens the calendar: at 80% those 144 days become 180.

Method 2: three-point estimates and PERT

When a single number feels too confident, collect optimistic, most likely and pessimistic durations per task. Expected time is (O + 4M + P) ÷ 6. Variance is ((P − O) ÷ 6)², and you can read an on-time probability from the total. This is the method behind “time estimation” when the work is uncertain. Open the PERT calculator rather than averaging three guesses by hand for a long list.

Method 3: dependencies and the critical path

Adding every task's duration overstates the project if work can run in parallel, and understates it if you forget a link. The overall duration is the longest chain of dependent tasks. In the default critical-path example the six tasks sum to 23 days, but the critical path is 17 days. If a subproject takes two weeks, the whole project takes at least two weeks and usually more. Use the critical path calculator to find ES, EF, LS, LF and float.

Method 4: Monte Carlo P50, P80 and P90

A simulation draws each task's duration at random between its three-point estimates and reports percentiles. The default three-task model (Design 2/4/8, Build 5/8/17, Test 2/3/4 days, beta-PERT, 10,000 runs, seed 2026) gives P50 16.1 days, P80 18.4 and P90 19.6. Changing the seed or number of runs changes the figures slightly. Commit to P50 and you are late half the time; P80 and P90 are the usual planning and safe commitments. The project duration simulator is the tool for this; it simulates duration, not cost, though the same percentiles apply to a budget you calculate yourself.

Method 5: analogous estimates from past projects

If a similar project took 18 working days with a similar team, start there and adjust for size, skill and known extras. Analogous estimates are fast and often closer than a bottom-up guess on unfamiliar work. They fail when the past project was not actually similar, or when you remember the plan rather than the outcome. Write down what you are analogising from so someone else can challenge it.

Common estimation mistakes

Do not treat effort hours as calendar hours. Do not assume twice the people means half the time: handover and communication eat the extra capacity, and some tasks cannot be split. Do not skip availability: people rarely spend all of a working day on project work; 70–85% is a common planning assumption (rule of thumb from practitioner experience, not a standard). If you have your own history of past projects, use that instead. Do not quote a most-likely figure as a commitment when the pessimistic tail is long. Do not add every task duration and call that the project length when some of them overlap. Each of those mistakes has a matching calculator on this site so you can check the number.

Task lists, PERT and worked problems

A bottom-up list of hours belongs in the task time estimator. Three-point (PERT) estimates belong in the PERT calculator. Textbook numbers (person-days, remaining time, adding hours) are worked on project time word problems. Illustrative project-type networks sit under estimate examples by project type.

A short sequence you can follow

  1. Write the effort in hours, person-days or person-weeks, and the team (people, hours per day, availability).
  2. Divide and round up to working days, then count them on a calendar from the start date.
  3. If the estimate is uncertain, replace the single duration with optimistic, most likely and pessimistic values and compute PERT expected time.
  4. If tasks depend on each other, build a predecessor list and take the critical path, not the sum.
  5. If the date is a commitment, read P80 or P90 from a simulation rather than quoting the mean.
  6. Cross-check with a past project of similar size when you have one.

Assumptions, limits and common mistakes

Assumptions and limits

  • None of these methods replaces a conversation with the people doing the work.
  • Calendars on this site skip weekends and the holidays you type; they do not know a country's public holiday list.
  • Resource limits, skill mix and learning curves are outside the formulas.

Common mistakes

  • Reporting effort as duration. 320 person-hours is not 320 hours on the clock.
  • Ignoring availability. A person billed at 8 hours a day rarely produces 8 hours of project work.
  • Padding every line and then adding a buffer on top. Pick one place for uncertainty, or use three-point estimates.
  • Using P50 as a deadline. That date is missed about half the time in the model.

Frequently asked questions

How do I estimate how long a project will take?

Divide planned effort by team capacity, round up to working days, and count those days from the start date. Use PERT when the numbers are uncertain, and the critical path when tasks depend on each other. The five methods on this page cover those cases.

What is the difference between estimated duration and effort?

Effort is the amount of work, in hours or person-days. Estimated duration is how long the calendar run takes after you divide by the team's daily capacity. The same 320 hours is 40 days for one person and 20 days for two, if the work can be shared.

What is time estimation in project planning?

It is the job of turning a description of the work into a duration or a finish date you can defend. The honest methods are effort over capacity, three-point estimates, the critical path, a simulated percentile, and an analogy with a finished project.

How do I estimate an expected time to finish?

For a single task, use PERT expected time (O + 4M + P) ÷ 6. For a project with a team, convert that expected effort into working days with the project time calculator. For a range, read P80 or P90 from the simulator.

Should I use hours, person-days or a finish date?

Use hours or person-days for the volume of work, then convert to working days and a finish date for anyone who lives on a calendar. Mixing the units in one sentence is how 7:30 gets read as 7.3 hours.

Written by . Formulas checked against the site's test suite on 11 November 2026. Last updated 11 November 2026. Formulas and tests: methodology. Report a calculation error.