What PR Insight is, what it isn’t, and where the data comes from.

PR Insight is a data analytics tool for international students navigating the Australian skilled migration program. It takes the same statistics the Department of Home Affairs (DOHA) publishes — points, invitation rounds, state nomination allocations, peer cohort composition — and reads them against your profile so you can see where you actually stand.

It’s a data tool, not advice. We describe what the published numbers show; we never tell you what to do.

No.We are not migration agents, migration lawyers, or an “immigration assistance” service as that term is defined in section 276 of the Migration Act 1958 (Cth).

Under s.276, providing personal immigration assistance without MARA registration is a criminal offence. We deliberately stay on the descriptive side of that line. For advice on your specific application, consult a MARA-registered agent — the official register is at portal.mara.gov.au.

All of it is public. Specifically:

DOHA SkillSelect EOI dashboard— pool composition, invitation rounds, occupation cutoffs, status breakdowns. Ingested monthly into our database after DOHA publishes each snapshot.
State / territory nomination portals— round results, occupation lists, target sectors, annual allocations.
DOHA processing-times portal— published 50% / 90% processing benchmarks per subclass.
FOI releases— DOHA’s monthly Data Access Request PDFs which give authoritative primary-applicant counts (we use them to cross-check the SkillSelect dashboard).
Skills-assessing authorities— ANZSCO → authority mapping, assessment fees.

Every figure we show is traceable back to one of these public sources. We never make up numbers and we never claim figures are predictive.

DOHA’s dashboard publishes the raw numbers. We do three things on top:

1. The bulk-band filter (typical-range).DOHA suppresses any cell with fewer than 20 EOIs to a midpoint of 10. That means a single suppressed cell at, say, score 70 makes it look like 70-point invitations are still happening — even when 95% of recent invitations actually landed at 85-95. We compute the smallest score window that holds at least 75% of recent invitation volume and lead with that as the “usual range,” demoting the thin tails. The dashboard floor for VIC 190 Software Engineer drops from a misleading 70 to an honest 80–100.

2. Cross-subclass pollution correction.A single EOI selecting 189+190+491 shows as LODGED for all three subclasses on DOHA’s dashboard, even though invitation was issued for only one. We anchor pollution ratios against monthly FOI releases and correct the inflation — 190 LODGED counts drop by roughly 1.74×.

3. SUBMITTED-only competition framing.“In the queue” only means SUBMITTED — the EOIs actually competing for the next invitation. INVITED and LODGED are past the invitation gate; counting them as competition (which many tools do) overstates the pool by 13–15%.

For any (occupation × subclass × state) slice, we look at the recent credible-score points and ask: what’s the smallest contiguous score window that covers at least 75% of recent invitation volume?

That window is the “bulk band” or “usual range.” Anything outside it is a “thin tail” — still shown, still real, but not where the bulk of invitations actually land.

Example: for VIC 190 Software Engineer over the past 2 years, the full credible range was 70–105 but the bulk band is 80–100. Most of the score=70 evidence is suppression-midpoint outliers; surfacing 70 as the floor implies an achievable cutoff that isn’t real.

DOHA publishes a new SkillSelect snapshot at the start of each month for the previous month’s end-of-month state. Our ingest pipeline runs on a cron the morning after, so most figures on the dashboard reflect last month’s snapshot within 24 hours of DOHA’s publication.

Every data surface on the dashboard carries an “as of” date stamp so you can see exactly which snapshot is being read.

No.Per the MARA s.276 boundary, every dashboard surface, every reel, every email is descriptive only — we say what the published data shows, we never tell you what to do.

You will not see phrases like “you should pursue X visa,” “your best chance is Y,” or “wait until the round next month.” That kind of guidance is reserved for a MARA-registered agent who can review your full circumstances.

What you will see is “the bulk of recent invitations for occupation Z on subclass 190 land at 85–95 points.” That’s information. What you do with it is your call, ideally with a registered agent involved.

The peers cohort narrows the active EOI pool to people with your exact filter set: same occupation (ANZSCO), same English level, same nominating state preference, same partner-skills status. Then it shows how many people sit in that exact cohort and how many score above you.

No competitor publishes this. DOHA’s dashboard lets you slice on subclass or occupation but not both at once with the demographic filters layered on top. Our 8-dim master cube was built specifically to support this cut.

Most other tools count anyone whose EOI is still “active” on the dashboard — SUBMITTED + INVITED + LODGED + HOLD. We only count SUBMITTED for “competition” framing, because INVITED and LODGED EOIs are already past the invitation gate. Including them inflates the apparent pool by 13–15%.

The pollution correction on LODGED counts compounds this further: when you correct for single EOIs being marked LODGED across multiple subclasses, the visible state-allocation numbers drop materially from what raw SkillSelect shows.

Both choices are deliberate. Smaller numbers, more honest numbers.

Yes. Subscriptions run through Stripe Customer Portal — you can pause, cancel, or change tier at any time. No call required, no cancellation flow designed to wear you down.

If you’re on the Founding tier and cancel, you can’t re-claim the A$5 lock; new subscriptions are at the standard A$9/month.

The Founding tier closes at 100 active subscribers. New signups go on the standard launch list and convert to the A$9/month tier at public launch.

Founding members stay at A$5/month for life — that price is grandfathered and never increased while the subscription remains active.