I was called into a meeting with a team here in the office a couple weeks ago because they told me they had a “question” about the estimations that they were doing. As we started talking, it became immediately apparent what the problem was, they were getting into arguments about whether their estimates were “too big!” Apparently, someone had told them that they “couldn’t” have any stories that were above a certain value, or at least that’s how they took the directions they were given. I stopped them for a minute and had a quick discussion about the reasons why we estimate stories, and why it’s incredibly important for the story points to reflect the size the team thinks the work is, regardless of what other people “want” them to do. I walked away to leave them to their work, and was entirely unsurprised when I saw some 20-pointers land on the backlog. Far too many teams suffer from some malady similar to that of this team — they forget why we’re asking them to estimate, so they start to engage in anti-patterns that undercut the very purpose for which estimation exists. In a follow-up conversation with another member of our Product Team, I started to think about how to describe Story Points as something other than “estimates” — and I came up with the idea of them as a “signalling tool”…
I’ve been working on B2B solutions for a very long time (dating almost all the way back to the turn of the millennia), and in that time I’ve come to realize that far too many applications try to be everything to everyone, and as a result really wind up serving nobody at all. You can see this in many product designs that try to capture all of the possible things that you could do at a given point in time, rather than leading you through a logical path, or showing you the most likely things that you may want to do. As much heat as I give the “ribbon” change that Microsoft introduced in Office a few (many) years back — conceptually, it was the right thing. It focuses you on the specific things that you need to do in some contextual space, without requiring you to remember which specific menu item someone decided to hide that option under. While the rollout was challenging, in my opinion, the approach really encapsulates a concept that I like to call “build for the novice, enable the expert”.
There’s a strong trend in Product Management circles to insist that a good Product Manager must be strongly technical in addition to having strong marketing and communication skills. And while this approach is well-meaning, it often results in a weak Product Management role that merely supports Development rather than challenge it.
Now, that’s not to say that a Product Manager can be successful without some basic level of technical competency — in order to have honest discussions with development teams, and to build the trust and respect of those teams, you must have at least a passing familiarity with the technologies that are being used by that team. You have to at least know what the terminology means – not knowing the difference between MySQL and NoSQL at a very high level, for example, can and will negatively affect your ability to write effective user stories.