Select Scenarios in the left navigation to test what-if ideas across every initiative you can see. The page has two tabs:
Scenarios: ask Bo, the Portfolio Agent, for alternatives to the plan across budget, resources, delivery dates and value. Every result is saved so you can compare them side by side.
Monte Carlo: the Scenario Lab simulations described below.
A portfolio's Scenarios tab offers the same two tabs for that portfolio.
Plan scenarios with Bo
On the Scenarios tab:
Bo's panel opens with Where your initiatives stand: a short read of budget, projected late initiatives and committed FTE, with a note when data is sparse.
Select a suggested question, such as Where should the org reserve go?, Rebalance toward the highest-value initiatives or Which late initiatives can funding recover?, or type your own and select Ask.
The Plan of record shows Budget allocated, Unallocated, Resources, On target and Value on time, with a table of each initiative's budget, resources, delivery date and value score.
Use Plan by to plan across Initiatives or Programs.
Untick In plan on a row to take that initiative's budget out of the plan.
Unallocated money is shown only when a budget is set. Set the portfolio's Budget, or the Org reserve in Configuration → Organization defaults, to plan against it.
How a Monte Carlo scenario works
As the New scenario panel explains, a scenario is a what-if hypothesis on top of your real portfolio. You pick initiatives and attach uncertainty distributions to their cost, duration and completion. The Monte Carlo engine then runs thousands of possible futures to surface P50, P80 and P95 outcomes, plus what's driving the variance:
P50: the middle outcome. Half the simulated futures come in below it.
P80: a cautious estimate. Four in five futures come in below it.
P95: a near worst case.
Create a Monte Carlo scenario
Select Scenarios in the left navigation, open the Monte Carlo tab, then select New scenario. You can also start from a portfolio's Scenarios tab.
Under Basics, enter a Name (required; a short, distinct label you'll pick from a list when comparing) and an optional Description of the story behind the assumptions.
Under Simulation settings, set:
Simulation horizon: the simulated window. Cost roll-ups and completion checks apply to it.
Iterations: how many random futures to simulate. The panel recommends 10,000; the range is 100–200,000.
Seed (optional): pins the random source for reproducible runs. Leave it blank for fresh randomness on every run.
Under Initiative assumptions, select Add initiative for each initiative in the hypothesis. For each one, pick the Initiative and set:
Budget multiplier: applied to the planned budget. 1.0 is on plan, 1.5 is 50% over.
Duration multiplier: applied to planned duration. 1.0 is on time, 1.3 is a 30% slip.
Completion probability: the probability the initiative finishes within the horizon (85% by default).
Select Create scenario. A scenario needs at least one initiative assumption to be runnable.
Save the scenario first, then run it to compute results.
Choose a distribution
Each multiplier uses a distribution. PERT (min, mode, max) - typical for PM is preselected.
Distribution | When to use it |
Fixed (no uncertainty) | The value is locked in, such as a signed contract price |
Uniform (min, max) | You have no idea where in the range it will land |
Triangular (min, mode, max) | A quick three-point estimate |
PERT (min, mode, max) | A three-point estimate for project cost and duration |
Normal (mean, stdev) | Uncertainty is equally likely either way |
Log-normal (μ, σ) | Values can grow large but can't go below zero, such as cost overruns |
For example, as the panel explains, PERT (0.9 / 1.05 / 1.4) means best case 10% under, most likely 5% over, and worst case 40% over.
Tips
Start from PERT for most cost and schedule estimates. Use Fixed only for numbers that are truly locked.
Write the hypothesis in the description, so you and your teammates know later why the scenario exists.
Plan to P80, report P50 - P80 gives a sensible buffer for commitments, and P95 shows your exposure if things go badly.