How we count, so you can check it

• We count every plan. No sample. Every Environment Plan on the register when we took our copy on 21 July 2026: 720 entries, less seven that were not real filings, 713 plans. The copy was frozen on 27 July 2026, so every figure can be reproduced.
• We measure elapsed time, and we don't divide it. Calendar days from submission to decision. Every day counts: the regulator's review, the titleholder's revisions, public comment and any wait between them. The register doesn't show how those days divide between NOPSEMA and the titleholder, so we don't divide them.
• We group by the year a plan was decided. Grouping by filing year leaves out slow plans that are still open, and makes recent years look faster than they are. Grouping by decision year counts every plan once it has an answer.
• We report the slow plans, not only the typical one. The typical plan is the middle one: half took longer, half took less. The slow plans are the slowest one in ten. Your schedule is lost to the slow plans.
• We don't report a group under 20 plans. Below 20, one unusual plan can move the answer too far. Where a thinner figure is printed, it is marked, and we say what moves it.
• We check against NOPSEMA's own numbers. Our counts of decided plans match NOPSEMA's published assessment-completion figures for every year from 2017 to 2023. Against NOPSEMA's annual reports, acceptances agree exactly in seven of nine financial years, and refusals in all nine. Filing counts differ, because NOPSEMA counts assessments and we count plans as the register lists them.
• What we don't do. No causes: the register shows what happened, not why. No forecasts: no decision date for any plan, no probability of acceptance. No titleholder named in the Census; names appear only in a Market Comparison Report. No plan text quoted.

ProveINTEL is not affiliated with, or endorsed by, NOPSEMA. Material on the register is subject to NOPSEMA's copyright and terms of use.