We just hired a new person onto our Learning & Development team. The shift from three to four people caused me to have a serious think about improving our methods for prioritization. The expectation of increased workload may have prompted new thoughts. Maybe it was the beginning of a new year with new goals.

Other factors may also have entered in—for instance, the broader subject matter requests our team received. And as additional requests flooded in, I realized that we really ought to consider prioritization more seriously.

Previously, we prioritized requests based on our understanding of business needs and training demand from external customers. We created cards in the backlog (we refer to requests in our backlog as “options”) and when we had time to work on those cards, we pulled the cards into our work-in-progress lanes on our Kanban board. Granted, some work was born-in-doing, where work bypasses both the backlog and the prioritized “top 5” column to begin its life in the Doing column (or the Implementation lane as we call it).

I wondered: Was the work born-in-doing a signal that we weren’t anticipating business demand correctly? or that we weren’t communicating clearly? It could have been a result of our ability to react and adapt quickly, as is the nature of a successful startup. Still, I wondered: How do we improve? Prioritization seems key.

A look back at our Kanban board data helped me see the work type distribution over the last six months: 40% training, 16% learning/research, 5% writing, 10% speaking, 24% general (general is a misc. bucket for everything else, similar to slurm).

An assortment of the various types of work that our team does helps us balance our work demand with our capacity to meet that demand. It is quite useful for the team to have a short prioritized list to pull from when considering what to work on next.

The approach for creating a short prioritized list from our long list of options includes:

1. Balance both high and low priority items. Lower priority items can be temporarily delayed if expedited requests surface. For our team, this could be a new integration opportunity that leadership wants, or a request to fill in for a Solutions engineer during a call with a new big customer.
2. Balance training & coaching, writing, speaking, learning/research, and general work based on demand for the work type allocations.

Because the demand for writing has increased, we believe we should always have some writing in progress, so I bumped up writing from 5% to 15%. I didn’t realize just how much time we spent on general work. I hoped that admin tasks, recruiting and expense reporting had not consumed 24% of our time!

The quantity of cards in a category does not equate to the amount of time worked in that category, I pulled up the cycle times for the general work. There, I discovered that we had been using the general work card type for everything from board design policies to strategizing. No longer puzzled by the amount of time spent on all those tasks, and knowing that our board design has since stabilized and our recruiting diminished, I decided to reduce the general work allocation from 24% to 15%.

Now, what about slack time? Previously hidden in other work types, slack time was not formally allocated as part of our balancing act. (No, I’m not talking about our favorite chat tool.) I’m talking about slack time that allows our team to a) be available when needed without dropping the ball on previous commitments, b) perform maintenance activities, and c) discover ways to improve. Our work type allocation plan for 2016 now includes slack.

Because of added slack time, learning time appears decreased, but it’s not really. When I have slack time, it tends to go toward learning: reading books, watching TED talks and conference videos, doing research, etc. We are a learning organization after all!

Learning time and slack time considered together effectively generate 20% slack time for us. Time that if needed, can be repurposed for expedited work or those out-of-nowhere special requests that drift in to take priority over other work-in-progress.

That said, I’m not sure about representing slack time using Kanban. It seems weird to put “slack” cards on the board. I ended up scheduling slack as a one-hour daily meeting with myself, but in my attempt to optimize 90 min Ultradiun rhythms, I later split it out into two 30 min sessions – one in the morning and the other in the afternoon.

However, slack is a valid priority. Maybe it’s the nomenclature… Maybe if we rename the term to “discovery” it will acquire status as a significant priority. After all, meetings with oneself are a form of discovery as are responses to unforeseen needs. Maybe an allocation of slack time is really an acknowledgment that slack is not a “nothing-to-do-here” mentality, but a way to open up to increased learning and capacity for responsiveness—especially responsiveness to the reality of work born-in-doing.