IBP Delivery Analytics

Util = hours logged ÷ capacity planned, for the week, per person. Weekly rather than per day because people log in batches — the same person can read 62% on Tuesday and 238% on Wednesday for work that was even across the week. The daily figure is still in each cell’s hover text.

Only days that have already passed count. For the week in progress the denominator stops at today, so a Monday morning does not read as 20% of the week. Days off are 0 in ADO’s capacity and drop out of both sides.

n/a means no capacity was recorded, not zero effort. Only about half the people who log hours are on a team that runs sprint capacity in ADO; for the rest there is nothing to divide by. The Capacity column shows the denominator, so an unusual percentage can always be traced to it.

0% means not logging yet, not idle. People with capacity but no hours are shown in grey, because leaving them out would divide the week by only those who did log — the week of 17 Aug reads 60% that way and about 16% with everyone included. 7pace adoption is still partial, so plenty of 0% rows are expected. Untick the box above to see only people who logged.

Over 100% is normal and worth reading. Capacity is what was planned per team, and people routinely work more than their planned share — or on more than one team’s worth of work.

Filtered to a product. Capacity is narrowed to match: a person’s allocation on the teams that own the selected product(s), not their whole day. Someone who worked on a product they hold no allocation in shows hours against a —, which is the finding, not a gap.

Filtered to a work item type. Capacity in ADO is not split by type, so it cannot be narrowed the same way. Against a whole capacity the percentage reads as share of capacity spent on this type.

They are left out of every figure on every tab, not just this table, so the charts and the timesheet cannot disagree with it. The rule lives in the v_worklog_analytics view.