Every Environment Plan on the register. Measured.
A census, not a sample. All 720+ plans NOPSEMA has published, 9,000+ extracted sections, every decision reason in the delegate's own words. How long assessments take by year, by region and by activity type, what has since happened to every plan filed, and a plain chapter on what the public record cannot answer.
Or request a custom report and we will scope it against the same register.
Start with the free edition. It costs nothing and it is the whole register.
The Environment Plan Census 2026 is a census, not a sample: every plan NOPSEMA has published, measured. Each report below is described in full on its own page: every metric, every chapter, and what it deliberately will not do. See every report, free to commissioned.
- Five findings on the register, each stated with its own population
- Filing volume by year, and what has since happened to every plan
- The assessment clock by decision year: medians and the 90th percentile tail
- Refusals against cancellations, and which activity types carry them
- The methodology, and a plain list of what the register cannot answer
- Assessment duration by region, activity type and lifecycle stage — 2017–2020 against 2023–2025
- All fifteen activity types by volume, median duration and cancellation
- Every titleholder with ten or more filings — filing and cancellation counts
- What returned plans actually failed, from NOPSEMA's published decision notices
- A data appendix reproducing every table, and a chapter on what the register cannot answer
- Your assessment durations beside every other titleholder's, by name, on the same era basis, post-2023
- n and a confidence interval printed on every figure
- Your own trajectory by decision year, and your open pipeline exposure
- Your enforcement history and spill envelope across the full published record
- Every figure re-derives from a pinned snapshot, so any number can be checked
The free edition needs an email address and nothing else. For the full edition, order a copy and we will send a quote, the contents list and the order form that carries the licence terms. Nothing is charged before you have all three, and there is no card field on this site. Or write to services@klarite.com.au.
Three lifecycle stages, measured on the same windows.
Short measurement briefs on the elapsed time from submission to decision for one lifecycle stage each, with the population and n printed beside every figure. No cause, no forecast. Each is free for an email address.
Or ask a question the reports do not answer.
The register records more than any one report can put in front of you. If the question you need settled is about a particular activity type, a region, a cohort of plans or your own filing record, describe it and we will scope it. If the public record cannot answer it, we will say so rather than estimate — which is the same rule every published figure here is held to.
All prices AUD ex-GST. The Environment Plan Census 2026 and the Market Comparison Report are each invoiced per deliverable and need no subscription. A custom report is quoted before anything is committed, and nothing on this site takes a card.
“Less than 10 per cent of plans are accepted by NOPSEMA on first submission.” Their words, not ours.
That is NOPSEMA's published position, last updated August 2021, and they publish no sample size behind it. What we can measure ourselves is the clock. Acceptance was never the risk — fewer than 3% of plans have ever been refused. The clock is the risk, and the clock broke. The median assessment for plans decided since 2023 is 304d (n=106) — roughly tripled since 2023. NOPSEMA “will typically offer two opportunities to modify and resubmit,” and of the returns we can measure from public notices, the median span from first return to final acceptance is a further +119 days (n=48; a floor, not a prediction). Getting it right the first time is worth the better part of a year. And the patterns that send a plan back are sitting in 720+ public plans nobody has time to read: the sections accepted plans consistently demonstrate, the reasons delegates actually give, the themes the regulator actually probes. That record is public. Until now it wasn't usable.
An RFI is a schedule event
An EP that NOPSEMA returns for further information takes a median of at least +119 days from first return to final acceptance (n=48 measurable returns; a floor, not a prediction). For an operator holding a rig or vessel window, that is not paperwork — it is a seven-figure delay.
The patterns are public
The sections accepted plans consistently demonstrate, the reasons delegates actually give, the themes RFIs keep targeting — all of it sits in 720+ public plans nobody has time to read.
Section quality is tribal
How a Spill Response or OPEP section should read is institutional knowledge — until people walk out the door with it. The accepted record is the benchmark that stays.
Three rules every figure in these reports has to survive.
Nothing here is a survey, an estimate or a model of what we think the register would say. It is the register, counted, with the counting rules printed.
The whole register, not a sample
Every Environment Plan NOPSEMA has published is in scope — 720+ of them — so there is no sampling error to argue about and no selection to defend. Where records are excluded, they are named and counted.
Every figure carries its population
No rate without its denominator, no median without its n. No median or rate computed on fewer than 20 records is presented as a finding; cells below that floor are printed with their n and marked in place rather than quietly dropped.
Re-derivable, and checked against the regulator
Every number is computed from a frozen, identifier-pinned snapshot, so any figure can be re-derived on request. Two independent reconciliations against NOPSEMA's own published counts sit in the report, on two publications and two time axes.
Built on the published NOPSEMA EP corpus: 720+ Environment Plans, indexed and searchable, with commitments cross-referenced against the Acts and Regulations that govern Australian offshore energy.
The questions enterprise buyers actually ask.
What is in the free edition, and what does the full one add?
The free edition covers the register at register level: filing volume by year, the assessment clock by decision year with its percentile tail, refusals against cancellations, and a plain statement of what the public record cannot answer. The full edition goes to the depth the free one does not: duration broken out by region, by activity type and by lifecycle stage across two eras, all fifteen activity types, every titleholder with ten or more filings, what returned plans actually failed, two independent reconciliations against the regulator's own published counts, the forward view over the plans currently under assessment, and a data appendix reproducing every table. Both are a census rather than a sample, so neither carries sampling error.
Why do you ask for an email address before the free edition?
So we know who is reading it, and so you have a copy of the link in your own inbox rather than in a browser download folder you will not find in a month. That is the whole use. There is no mailing list, no drip sequence and no third-party marketing tool behind it: the address arrives as an email to a person here, and if you would rather not give one, write to services@klarite.com.au and we will send the file back by hand.
What does A$10,000 actually buy?
A single-organisation licence to the full edition of The Environment Plan Census 2026 — the whole document, for everyone in your organisation, not a per-seat or per-report fee. Quotation is permitted with attribution and with populations named as printed, so figures can go into a board paper or a client report. It is invoiced on release and needs no subscription. Buying it starts an order: we send a tax invoice, the contents list and the order form that carries the licence terms.
Can you build a report on something else?
Yes, and that is what the custom request is for. The register can be interrogated on any axis it actually records — an activity type, a region, a lifecycle stage, a cohort of plans, your own filing history — and if what you want cannot be answered from the public record we will tell you that rather than estimate it. Describe the question in the scoping box and we come back with what is answerable, what it would cost and how long it takes.
Who is this for?
EP consultancies, titleholder regulatory affairs teams, and business development leads in Australian offshore energy. Anyone who needs to understand the NOPSEMA landscape faster than reading 720+ plans.
Why don't you score an EP or predict acceptance?
Because you'd be right not to trust it if we did. Only fewer than 3% of plans have ever been refused, so an "acceptance probability" is close to meaningless — and a black-box score is nothing you could defend to a client or a regulator anyway. The real question was never whether a plan gets accepted. It is how many times it comes back first, and what that costs. What we give you, with citations, is the record itself: what accepted plans demonstrated, what delegates actually wrote, and how long each assessment took. The professional judgment stays where the professional accountability sits: with you.
Where does the data come from?
The public NOPSEMA Environment Plan register. We index every published plan and track register status changes automatically. Deep section-level extraction currently covers 650+ plans — around 91% of the register — and is expanding. No confidential data, no scraped titleholder systems. Only what NOPSEMA already publishes.
How accurate are the benchmarks?
Benchmarks are computed from the actual register, not estimates. Processing times, outcomes, and activity splits come from the regulator's own published status fields. Where a number is modelled rather than counted — like the assessment-clock percentile ranges — we say so, and we show a range, never a false-precision point estimate.
Can we quote your figures?
Yes, with attribution, and with the population named as it is printed. That condition is not decoration: most of these figures are only true of a stated population, and a rate reproduced without its denominator becomes a claim we did not make. The licence terms travel with the order form.
Is there still a live dashboard?
There is, and it is not what this site sells today. app.proveintel.com runs the live register application and existing users can still sign in there. We have taken it off this site while the reports are the offer, so that what you can buy here is exactly what is described here. If a live seat is what you actually want, write to services@klarite.com.au and ask.
Start with the free edition. It costs an email address.
The whole register at register level, and it is the same census the licensed edition is drawn from. If what you need is not in it, tell us what the question is and we will scope it. Neither needs a card, a call or a procurement cycle.