Could the INVEST criteria Bill Wake came up with for evaluating User Stories help us come up with effective PI Objectives in SAFe as well?
I think a good PI Objective should be:
– Independent – meaning ideally it could be delivered and evaluated on its own without any dependency on other PI Objectives. And if a team is able to own a PI Objective and delivery it on their own – it’s a sign that it’s a cross-functional autonomous team.
– Negotiable – meaning that we keep the details of the HOW open. The team commits to the WHAT and will figure out the HOW along the way based on what they learn. This relates to “Assume Variability, Preserve Options” which is a crucial but sometimes ignored principle in SAFe.
– Valuable – in the eyes of the business owners. At a minimum, a good PI Objective should be UNDERSTOOD by business owners of the Agile Release Train. Even if it is about enabling a future outcome rather than delivering it directly.
– Estimatable – In order to figure out if the objective is realistic and can be delivered in the upcoming PI, the team should be able to estimate what it might entail
– Small (Or Sized appropriately) – In this context Small means sized to fit into a PI
– Testable – can we test that we actually achieved this objective? How would we know what to give it as an “Actual Business Value”? This is where things like “Key Results” (Think OKRs) can help us make a PI Objective more concrete.
So what do you think? Is the INVEST criteria useful for PI Objectives?