The ScrumButt Test (5): A prioritized Product Backlog is essential
We started a series on The ScrumButt Test some time ago. This is the fifth explanatory post about the test. To find the others, search above (right). Here is the original post.
The second item in the second section of the ScrumButt Test is this:
“There is a product backlog prioritized by business value”
Seem obvious? Well, maybe you don’t want to say so quickly.
First, if we are doing Scrum, we must have a Product Backlog. A Product Backlog is a list of all the work items the team is expected to work on, or at least someone has suggested they work on. If the Team is doing software, most of them will be features at a high or low level of granularity. In Agile, the growing preference is to put these Product Backlog items in User Story format.
This part of the ScrumButt Test does not require you to use the User Story format. Or any particular format.
“Prioritized”: Ummm. What does that mean?
Well, I think it means that the Product Backlog has been put in order of Business Value. Or perhaps benefits/costs order.
The ScrumButt Test does not give us any hints about what Business Value is. My opinion is that this itself is a complex topic. But I think the implication is that the Product Owner makes at least the final decisions (he or she ‘owns’ the Product Backlog, and makes all final decisions about the Product Backlog). Certainly the PO should listen to others. Amongst these others I think we almost always have some people I call “business stakeholders.” The business stakeholders are chickens, but presumably pretty smart (and often they are) about business value.
Does the test item imply we work the highest business value items first?
This is open to some debate. My view is that one can still pass the test if one also uses “cost” (or story points as a proxy for cost) and dependencies/risks as additional factors (bits of information) in helping the Product Owner to order the work.
Let’s put this some other ways.
The technical team does not have the final decision about what order to do the work. Sometimes it is better to be inefficient and get some high Business Value item(s) out there fast, in production.
Since all PBIs (Product Backlog Items) are not of the same size (let’s call this “amount of work” for now…although one must tread carefully here), and not of the same Business Value, the Product Owner must do cost-benefit analysis to determine (implicitly at least) business value per unit of work for each PBI.
Why do we prioritize by business value? We are always trying to discover and release business value (eg, increase customer satisfaction). Quickly. Only if we use Pareto’s 80-20 rule idea can we do the most important stuff first…can we maximize BV delivered to the customers. Ordering the PBIs puts Pareto’s rule into play.
And I think the ScrumButt Test says the Product Owner cannot cop out and say each PBI has the same amount of Business Value.
Similarly, this part of the test does not prohibit some team members (developers), where they have useful knowledge, from influencing the Product Owner about business value and about the order of the work. (Still, Scrum says the Product Owner gets the final say.)
* * *
Well, that’s a start on the implications from this part of the ScrumButt Test.
Please see our other posts on the ScrumButt Test.
Note: The picture of the story card above was stolen from here: http://www.pragmaticsw.com/newsletters/Newsletter_2008_05_SP.htm