Example data. Fieldnote is a fictional company used to fill all 61 boards, so the examples tell one consistent story.
Scale-up, the two squads and the future Integrations squad compete for the same quarter; the product managers score four features on cost of delay.
What the team took away
The small activation checklist jumps to the top because it is cheap, then offline release 2; recurring contracts, everyone's favourite, waits a quarter because of its size.
How to run it
Preparation
Invite the product managers of the train and one or two architects. Business owners join for the value column only if they can stay the whole session.
Send the candidate features with a one-line benefit hypothesis each, two days before. No hypothesis, no seat at the table.
Pre-fill the rows and write the Fibonacci scale (1, 2, 3, 5, 8, 13, 20) at the top of each column. Pick the smallest feature as the reference 1 for every column.
Agenda
15 min
Calibrate Read the four definitions aloud: Business value, Time criticality, Risk reduction or opportunity enablement, Job size. Place the reference feature at 1 in every column so everyone scores relative to it.
25 min
Score value Relative estimation, poker style: everyone shows a Fibonacci card at once. Discuss only when the spread is more than two steps. Business owners have the final say.
20 min
Score time criticality Ask for each feature: 'what do we lose per month of delay?' If the answer is 'nothing specific', it is a 1, whatever the stakeholder says.
15 min
Score risk reduction Ask: 'what does this unblock or de-risk?' Platform work usually scores high here and low on value. That is the point of the column.
25 min
Score job size Architects estimate relative size across all teams involved, not just the one asking. Split anything scored 20 before it goes further.
15 min
Compute and read Fill the WSJF column (sum of the first three divided by size). Read the top 10 aloud. Check that at least one enabler made it: if not, revisit risk reduction scores.
5 min
Close Freeze the order for the PI. Note the features whose rank depends on a single high score and flag them for review at the next planning.
Pitfalls
Do not estimate size in absolute days. The whole formula only works with relative numbers on the same scale for every column.
Do not let a feature scored 20 in size stay on the board. Split it: big things always lose to medium things in WSJF, which is intended, but only if the medium things are real.
Do not re-score between plannings to fix an order somebody dislikes. The cost of delay changes only when the market or the date changes.
Afterwards
The ordered list is the program backlog that goes into PI Planning. Enablers ranked high go to the architects first so teams can plan around them. Re-score at the next PI boundary, not before.