Triage Prioritization When Capacity Isn’t the Constraint

MoSCoW leaves you with too many Must Haves. With more delivery capacity, the pressure to say yes gets worse, not better.

Share
Triage Prioritization When Capacity Isn’t the Constraint
Photo by Ozkan Guner on Unsplash

Sometimes PMs get overwhelmed with feature requests and shut down. "It's all must-have," they say. A novel approach to traditional MoSCoW prioritization can help get them unblocked.

Context

PMs are struggling to prioritize their backlog, feeling everything is a must-do. Priorities are needed urgently for a planning session or a presentation and we don't have time for more rigorous methods.

Discussion

Often external pressure is so high that PMs feel they have no options but to do it all.

They're wrong about this.

There are always options; some are just unpleasant and politically difficult. But if PMs are scared of saying 'no' or even 'not yet' to anything, they need a different lens to prioritize.

This play is especially helpful in a chaotic, low-safety, low resource, emotionally led, or low-data environment, because it relies only on intuitive judgment of a small group.

Play

Guide them through a triage prioritization using 'Must Have, Should Have, Should Have, Could Have, Won't Have' (MoSCoW for short). You probably know this approach already, but what's unique in this play is the framing.

The classic problem with MoSCoW prioritization is that you end up with too many Must Have items, and the PM gets stuck, unable to prioritize them. This usually comes from an MVP mindset and a fixation on what makes the product viable. But in most cases, your product doesn't need to be viable right away. Instead, you need to start learning. And often, in an established company, you're learning about risks, your technical constraints, and what's going to block you.

Often the PM doesn't realize that they have the option of learning releases, and I want to get them thinking about this.

So here's how I frame the different buckets for the team doing the exercise:

  • Without 'Must Have' items, we will immediately go out of business. Or the release will simply not function at all.
  • Without 'Should Have' items, we will have problems. We will be embarassed. Users will be disappointed. Promises might not be met. We will have to do real political work to address the gap.
  • Could Have are normal items that we urgently want, and someone needs. If they are delayed, we will need to communicate this, and others will need to adapt their plans to avoid a mess.
  • Won't Have are the rest. If there is time, we will do them.

When we adopt a triage mindset for our prioritization, we can begin to envision a much smaller first increment. One where we can learn somehow, show the system to someone, and get some feedback. That feedback is worth a lot, even if the product isn't viable yet.

In ancient times this was a normal practice that was taught as part of the XP and Scrum methods, but these days I find many people are just sprinting through a fixed scope someone else has given them. Even if the fixed scope is signed off in blood in a requirements document, you can still use triage prioritization to meet a fixed date with something valuable. And once you start meeting fixed dates, you can begin to gain permission from your leadership to pursue more modern ways of working.

Tooling

If the backlog is already in Jira, use the priority field that's already there. Everything defaults to 'normal' priority. First, identify the 'going out of business' items and mark them 'Highest'. Then, identify the 'it can wait' items and mark them 'Lowest'. It's a 5 point scale in Jira - only use the other two options if you start getting hung up in discussion.

If I'm doing this in a board with sticky notes (or in Miro or similar) we can do the sorting in parallel - set up a table with the different options, and have the group silently move them, all at the same time. Set a timer for 3 or 5 minutes. If the group is really able to be silent, they can sort 20-30 items in just a couple of minutes, and then when the timer goes off you can discuss the handful of items where there are disagreements.

Cautions and Caveats

I feel this is a relatively low-risk play in most contexts, so I don't have any cautions. Readers, I'd love to hear from you in the comments about where this play should not be used.