Which date can you defend when the client pushes?
Adding up the likely durations gives a date you will beat fewer than 10 times in 100. Simulate the same ten tasks and you get the date you can defend, plus the news that two workstreams take turns driving it.
Words on this sheet
- Predecessors: The tasks that have to finish before this one can start. Put their IDs here, separated by commas, and leave it blank when nothing has to come first.
- F statistic: The variation between the groups divided by the variation inside them.
Work Intermediate Schedule Risk Pro engine
After you install, this is the model to open.
Can I Promise That Date to the Client?
- In your spreadsheet, click the Sortia icon in the strip of icons down the right-hand edge. No strip? Click the arrow at the bottom-right to open it. You can also use Extensions, then Sortia, then Open Sortia.
- Click Start from a template and put that name in the search box.
- Pick the card with that name and click Load this template. It arrives on a new tab with real numbers already in it.
This one runs on a Pro engine, and every free install includes five full-quality runs on your own numbers, shared across all five Pro engines rather than five for each. After that, Pro is $199/year.
The answer
- The date you can defend
- 50 days the P80 of 20,000 simulated finishes
- The plan the meeting hears
- 41 days the likely durations added up; beaten in 4.7% of runs
- Median finish
- 47.0 days the P90 is 52
- Two chains take turns
- 62% / 38% of trials driven by the data chain vs the interviews
Ten tasks on one client engagement, each with a best case, a likely case and a worst case in working days. Add the likely durations down the longest chain and the plan totals 41 days, which is the number that gets said out loud in the meeting. Click Run on the 20,000 trials the template loads with. The mean finish is 47.0 days, the P50 is 46.8, the P80 is 50.1 and the P10 is 42.2, so the 41-day plan is beaten in fewer than 10 of every 100 runs.
The date you can commit to and be right 80 times in 100 is 50 working days. That nine-day gap is not padding and it should not be hidden inside individual tasks. It is schedule contingency, it belongs to the engagement rather than to any one person, and naming it as contingency is far easier to defend to a client than quietly adding two days to every line.
The criticality table is the second half of the answer and it is the more useful half here. Kickoff, the workshop, the draft, the client review and the final issue are on the driving path in every single trial, because everything funnels through them. The interesting number is the split between the two streams: the data path drives the finish in 62 of every 100 runs (62.3%) and the interview path in 38.
Neither one owns the date. A plan whose likely durations are one day apart does not have a critical path, it has two paths that take turns, and management attention spent on only one of them is a coin flip. That is what a single deterministic run cannot tell you, and it is the reason to run this before promising anything. The widest input is the client data extract at 5 to 20 days, and that gives you the most useful sentence you can say in the meeting: the date depends on when your team gives us the extract, and here is what each week of delay does to it.
Second run: change that task to 5, 8 and 12, which is what a client with a named data owner and a written specification looks like, keep the seed at 7 so the comparison is clean, and rerun. The mean comes in to 46.0 and the P80 to 48.9, and the criticality split moves to 53.8% against 46.2%, which tells you the interview path is next in line for attention.
What this cannot tell you is anything about resource clashes: the simulation schedules every task at its earliest start with no regard for who is doing it, so run Resource Load on the same team before you trust the date. The sibling template "What Finish Date Can I Commit To?" runs the same method on an internal system cutover, for dates defended to a steering committee rather than a client.
To adapt it, replace the tasks and predecessors with your own, and set the three estimates from your own history rather than from feeling. If you have a list of real task durations, run a histogram and a distribution fit on it first: the fit tells you the shape and the spread of what your tasks really do, and if it comes back right-skewed then your pessimistic leg should sit much further from the likely than your optimistic one does, which is exactly what this table looks like.
The model
It arrives on a tab called Template: Can I Promise That Date, carrying these columns:
- ID
- Task
- Predecessors
- Optimistic (days)
- Likely (days)
- Pessimistic (days)
Once it is in your sheet
- The model arrives with real numbers in it and runs as it stands, so you can press the button first and understand it second.
- Change the numbers to yours. The sheet marks which cells are inputs and which hold formulas, and most labels carry a note explaining the row.
- Press the run button at the bottom of the panel. It is labeled for the tool you are in, and the result lands on its own tab, with a written reading of it beside the figures.
Never used Google Sheets? Start here goes the whole way, in seven steps, and assumes nothing.
Next question
- What is this comp package really worth?
- What is the engagement worth once scope grows?Take the Engagement or Pass?
- Which day can you promise and still be right nine times in ten?What deadline gives 90% odds of finishing?
- Which bid is low enough to win and high enough to pay for the job?What bid wins with 70% odds and still makes money?
- Which job offer really pays more?Two Job Offers: Price the Tradeoffs
- Whose calendar is the real critical path?Is Anyone Double-Booked?
Every model like this one, and the method behind them: Schedule risk analysis.