Project Management Central

Please login or join to subscribe to this thread

Topics: Risk Management
In risk management, how can I determine the probability that a risk can materialize?

Im at the planning of a new project and we are generating a risk plan. In the PMBOK there is a glimpse of how to define the probability and the impact of a risk. The impact is kind of clear but the probability is little fuzzy and subjective due the percentages. How can I establish a criteria to those percentage? any suggestion is welcome?

Sort By:

In a few fields like safety and quality, statistical analysis can derive a numerical probability. For most purposes, that's not required. A scale of 1 to 3 or 5 is often used (low to high, or very low to very high), and the term likelihood is sometimes used instead.

You can create your own criteria for what is a 1 verses a 5, but the absolute value is not really that important. When we have discussions over a risk, some might want to call it a 3 and others a 4, but generally everyone will come up with a fairly close score unless they have different information.

The most important part is not the score, but the conversation or thought process as how you chose that score. Knowing the rationale behind why you scored it as you did is where the real value lies.

Thanks Keith I agree with you, as PM I was asked to came up with a draft of how the impact and probability should be used to issue a severity in the risks. The first draft I showed was 1 to 3, I’ve read the PMBOK and was expressed in percentages but I think I will involve my team to determine a”score / criteria”

Hi Daniel. You need to go back to PMBOK 11.3 were you will find the Tools and Technics to determine probability! In very plain language it all depends on your (and your organization) experience and information!
That is my experience you will have to gather as much information and background on each risk as possible and then make your assumptions!

Something that was created inside the CMU SEI when I was there and I used for years and works for me is the following table:The key iis: when you ask if you received the "Uncertainty Statement" as an aswer then you have to assign:
Probability Uncertainty Statement Evaluation
80% Almost certainly, highly likely 5
60-80 Probable, likely, probably, we believe 4
40-60 We doubt, improbable, better than even 3
20-40 Unlikely, probably not 2
1-20 Highly unlikely, change are sight 1

Ask your team. In most cases, they'll have more insight than you do because they're doing the work. Review your risk register with them on occasion, because their insights will change as the project progresses.

On most projects, high-level probabilities (such as 1-5) will suffice. In most cases, these are just guesses, anyway, and you're merely categorizing so you know which risks require your attention. Don't spend too much time trying to get these perfect ("Perfect" is the enemy of "Good"), because we expect them to change over time as 'unknowns' become known.

Daniel -

The challenge with assessing probability of project risk realization is that no two projects are the same and hence even if you had hundreds of similar projects to draw data from that might not help you on your current project.

In general, expert judgment and historical data are used for determining probability, but using a scale such as the one provided by Sergio can help to get all the SMEs on the same page.

You might consider using techniques such as Delphi Method to overcome any initial biases if determining probability within a team meeting.


Please login or join to reply

Content ID:

"No opera plot can be sensible, for in sensible situations people do not sing."

- W.H. Auden



Vendor Events

See all Vendor Events