The Scrum Guide doesn’t talk about the role of managers. So does that mean that one day we have managers, and then as soon as we begin an agile transformation, they suddenly disappear? Of course not. So what do we do with them? Many organizations use managers to direct the work of their reports so it could seem reasonable to assign these managers as Product Owners. I can see the logic in this, but a younger me would have opposed this with extreme prejudice. A younger me would use all my persuasive might emphasizing how this isn’t the right thing to do. But why? What’s wrong with it?
Some might point to the Scrum Guide reminding the org that the Guide only talks of three roles: Product Owner, Scrum Master, and the Development Team. However, I don’t find that argument very convincing. It’s too dogmatic, and it’s clear that there could be value in the manager as Product Owner. However, it comes at a cost.
Manager as Product Owner
|Provides clear line of sight on all reports to appreciate their caliber of work...||...but at the cost of the team candor because of the invisible gun effect. It can also potentially foster competition instead of collaboration as reports wish to be viewed more favorably than others.|
|Streamlines conversations because the PO and manager are the same...||...but at the cost of confusion when a team member is unsure which hat is being worn. Also often leads to single-discipline or skill-deficient component teams instead of cross-discipline feature teams.|
|Clarity in accountability because the manager knows who to talk to if something goes wrong...||...but at the cost of self-management.|
|Greater influence over the growth of the people and growth of the product simultaneously...||...but at the cost of divided attention since each is usually a full-time role.|
Does that mean a manager made Product Owner is doomed to fail? No. In fact, I remember a team where all team members reported to the Product Owner. It remains as one of the most effective and self-organized teams I’ve worked with, and this manager as Product Owner’s rule was a simple one:
Unless we’re at our one on one or you’re throwing a punch at someone, I’m always your Product Owner.
Unfortunately, most managers feel they can put their egos aside and not tromp on the team’s ability to manage its own work. They believe the costs I’ve outlined above don’t apply to them. Believe it or not, they still do, and it takes a great deal of time and effort to mitigate those costs. Further, I recently asked the great people at the Agile and Lean Development Group their thoughts on this topic. I like how Paul Oldfield puts it:
They [managers] genuinely do believe that they know better.
I agree with Paul. So what happens when we flip this on its head? What are the costs and benefits of a separation of Product Owner and manager?
Separation of Product Owner and Manager
|Allows for self-sufficient, cross-functional feature teams...||...but managers lose a direct line of sight to an individual's performance.|
|Clarity in role since everyone wears exactly one hat...||...but at the cost of having to involve managers when such situations arise. (But is that really a bad thing?)|
|Clarity in focus since the Product Owner can focus solely on the product and manager solely on the people...||...but at the cost of divided responsibilities, agendas, and perspectives.|
|Multiple layers of coaching and mentoring by Product Owner, Scrum Master, and manager...||...but at the cost of time, indirection, and confusion as to who is responsible for what.|
I think we can see why a traditional organization might value manager as Product Owner and agilists value the latter. So if not Product Owner, what should managers do? Should they be Scrum Masters? I don’t believe so. Read my post about Managers in an Agile Shop for my ideas.
Next time, I plan to tackle how managers with reports distributed across several teams can still effectively mentor their people so stay tuned.