Tom Angove and I gave a talk today on what we think a great product owner is.
We were at the IIBA “Festival of Business Analysis,” so there were quite a few people who knew what a Product Owner was.
I would love to say that the talk consisted or erudite guidance from the presenters and passionate, persuasive recommendations that caused the audience to take immediate action on leaving the presentation. Unfortunately I don’t have much evidence to support that claim.
What I can say is that we got no where near the end of the material we could have gone through because there were a lot of questions from the audience and also several points that members of the audience wanted to share. This made it more of a discussion than a lecture, which was great.
I particularly liked a discussion between Tom and one of the audience (Terry I think it was) when they discussed making clear to the team how valuable they were, not in generic terms, but in terms of how much it costs to have the team and why what the team does is worth more than the money invested in them. That discussion alone would make for a great talk or workshop.
However there was one answer I gave that I think was incomplete. Someone asked us “how do you measure the value of a product owner.
That is a good question
It is a very relevant question, because Tom and I had been saying that companies should get a positive return on the money they spend on a PO. We had also been saying that a good PO should be able to help teams to measure things well.
So if we claim that a PO should be good at measuring things and that we should be able to measure the value we get from having a PO, then it would make sense that we should be able to identify, demonstrate and measure the return we are getting for investing in a PO.
I gave a quick answer
I said that the problem was never measuring the value but rather defining it.
I think that is true. It is hard to measure intangible things (contribution to the mojo of the team) but easy to measure concrete things (stories completed per sprint, cars sold per month). So when we say it is hard to measure the return we get on our investment on a PO, the challenge is to clearly define what the PO contributes.
The more tangible we can be the easier it will be. But even if we end up with something intangible like “team members know what the team’s priorities are” then we can rate it roughly, by asking the team what they think.
I am glad I did not say the wrong thing
I could have given a worse answer. I could have given a specific set of measures (like these dodgy ones or some measures I have seen in performance agreements). I have used these effectively but they only really work in context.
I believe that it is easy to come up with measures too quickly, turning them into goals in their own right (See Goodhart’s Law).
“when a measure becomes a target, it ceases to be a good measure.”
Goodhart’s law
Instead, I think we need to give the PO room to define what their goals are and how they can achieve them, but understanding their particular context and the most important things that they can do for the particular team they work with.
A better answer will take a lot longer
I would say generically, that a PO is successful if:
- The product they are working on is creating value for the right people.
- The decision makers that the PO works with are making better decisions because the PO is there
- The team(s) that are better able to create value (enhance the product, pivot to something new, kill a bad product off) because the PO is there.
There are many factors that dictate whether the product is successful. I think the PO can add a lot of value by helping make these factors apparent, so the team and its stakeholders know the value of the product and, to be blunt, if the effort of building and supporting the product is worth committing.
However, a good PO can be trapped with a bad product (I have been and it took a while to get people to understand) AND a mediocre PO could walk into an environment where the product is incredibly successful in spite of what the PO is doing (teams I have worked with have been successful without me actually being the reason at all).
So that leaves us with some simple goals, which are only examples:
- Better decisions
- Easier, more efficient pursuit of value.
If those are the goals, then the next step is to ask some questions:
- Who cares if this goal is achieved? Whose opinion really matters the most here?
- Based on those answers, what questions should we ask on their behalf to understand if the goal is being achieved?
For example
Has the PO helped this stakeholder to make better decisions? How have they contributed to the quality of decisions being made?
Me trying to come up with a measure
Then the last step is to challenge ourselves to find:
- The answers to the questions that we identified;
- The evidence that supports our answers; and
- The confidence we have in answering those questions.
This is still hard, but it is now achievable.
- Maybe we can see a reduced cycle time in the team and the reason is quicker decisions, smaller stories and better coffee supplied by the PO.
- Maybe we can see that decisions are more often based on external customer data (4/10 of our decisions now), the stakeholders all have a voice now, or we have created tests to validate assumptions that caused us to make small mistakes and not giant ones.
So my answer is still “the challenge is defining what we want from the PO, not actually the difficulty of creating measures once we know that.”
There was a lot more in the talk and I would love to have captured it all. Maybe you should run a conference and ask Tom and I to speak again so we can revisit the discussions.
James King 