Larry Maccherone, Rally’s Director of Analytics started a blog series on the seven deadly sins of Agile measurement. He asked me to write about the last sin that discusses probability and risk, and I of course accepted his hospitality. Our journey on getting all Agile teams to consider probability and risk when forecasting continues, and opportunities like this are golden to getting the message to a wider audience. Thanks Larry!
The full article is here – http://www.rallydev.com/community/premise/seven-deadly-sins-agile-measurement-sin-7
The keypoints at the conclusion of the article were –
When forecasting a project, spend some time determining what can go wrong, the impact it would have, and how long that would take to resolve. Typically, the impact of the first three risks on a project will be bigger than the variation of the known work. This is a good point to ponder when setting your developers on the task of detailed work breakdown and estimation.
- All forecasts have multiple possible outcomes, some more likely than others.
- An outcome at the pessimistic or optimistic end of the spectrum is least likely.
- Use probabilistic forecasting techniques like Monte Carlo simulation to identify which results are more likely than others.
- Risk factors probably pose a bigger chance of impact than the uncertainty of any one story estimate , or even the cumulative effect of all story point estimates. Spend more time asking your teams, “What can go wrong?” and less time asking them, “How long will it take to …”
If you’re interested in reading more about this topic, here are some good books to get you started: Flaw of Averages by Sam L Savage, and How to Measure Anything and Failure of Risk Management by Douglas Hubbard.
– See more at: http://www.rallydev.com/community/premise/seven-deadly-sins-agile-measurement-sin-7#sthash.329vlTpl.dpuf