Monte Carlo forecasting, explained without the maths

Ask a team when something will ship and you'll get a date. Ask how sure they are and you'll get a shrug. Monte Carlo forecasting replaces that shrug with a number.

Start with what really happened

Instead of estimates, we use how many items your team finished each week over the last few months. That history already includes meetings, bugs, sick days and all the other things estimates forget.

Roll the dice ten thousand times

The simulation picks a random week from your history, again and again, until the remaining work is done. Repeat that ten thousand times and you get ten thousand possible finish dates.

A forecast should tell you how likely a date is, not just what the date is.

Read the range

Sort the results and you can say: there's an 85% chance we finish by 14 November. That is a promise you can make to customers.