NCTs at Keeper

General planning process steps

  1. CEO creates business-level NCT kickoff document outlining business priorities and a top-down drafted proposal for major initiatives
  2. #staff-leads reviews and debates NCT kickoff document and assigns shepherds for each Narrative
  3. NCT kickoff document shared with rest of the company - comments, questions, concerns are encouraged
  4. Shepherds create drafts of their narratives & share in #goals-and-strategy. Initial draft should include story narrative and bullet-list of tasks. No need to group tasks into commitments yet.
  5. Shepherds work with team members to generate, spec, and prioritize big list of tasks. 1:1 brainstorm meetings between the shepherd and each team member are encouraged, but group brainstorm meetings are also OK. ****The end result should be a prioritized long list of tasks. There should be at least double the number of tasks listed vs the amount that the team will likely be able to accomplish. The top 2/3 of that list should be spec’d, meaning there should be mockups or at least a document linked that provides visibility into the work that will actually need to be done.
  6. Shepherds share & defend their list prioritization of tasks with team members. Async document / Figma comments + a team-wide meeting to resolve the bigger ones is encouraged. The goal here is NOT to assign the tasks, it’s just to develop a shared understanding of what’s most important to do and how much work it’ll entail.
  7. Team members select Tasks from the list. Items higher on the list are encouraged but not mandatory. Roughly speaking, the top 5 items on the list are mandatory, and items 5-15 are encouraged. Claimed tasks should have a 70% completion rate by the end of the Tertile. Pick carefully and spend time to think through the work itself to make sure you don’t under or overcommit.
  8. Shepherds + managers review selected Tasks, and try to convince team members if critical items are unclaimed. Ultimately, it’s the team member’s choice what they take on. If there is disagreement, this is natural and healthy and should be discussed but the buck stops with the team member’s decision.
  9. Shepherds group claimed tasks into Commitments, set metrics for each commitment, and share in #goals-and-strategy. Commitments are simply groupings of tasks. If certain tasks don’t fall into a clean commitment bucket, it’s okay to create a misc bucket or to not have a corresponding commitment whatsoever so long as the Tasks without commitments are less than 20% of total work.
  10. Last round of comments and feedback on NCTs. This final round of comments shouldn’t result in major changes, but might call out cross-functional gaps or misunderstandings.
  11. Tertile kicks off, and shepherds share finalized NCTs in all-hands. Simple slide with commitments stoplight-colored.

Why set Narrative-based goals vs team-based goals?

Implicit in the above process is an emphasis on cross-functional collaboration. This makes inherently cross-functional projects easier to keep track of (e.g. implementing a rebrand), at the cost of making harder for each individual team to have a single-place of record for all of their tasks. We believe that this tradeoff is worth it, because people are inherently good about understanding their own areas of ownership rather than thinking across org lines.

Let’s do an example. Let’s say that a company has 3 cross-functional initiatives that need to be done in the upcoming quarter — a rebrand, shipping a new product line, and major push towards customer support automation. Getting this done involves getting tasks prioritized on the backlogs of dozens of different teams. In a matrixed planning system, that means each of those teams needs to remember to add related tasks to their backlogs. To make that happen, a lot of responsibility is placed on the team lead to coordinate and be aware of each of those cross-functional projects and their timelines. The NCT process does two critical things to fix this: (1) it gives business-level visibility and emphasis to the project, reducing the odds of inter-team politics due to misaligned incentives and (2) instead of expecting every team lead to have all the context to be able to prioritize cross-functional tasks, it puts that work on an explicit “shepherd” who can project manage more effectively.