Entific Research

Every Federal Court decision on temporary resident refusals. Taken apart.

845 study permit, work permit and visitor visa judicial reviews since 2022 — each one broken into the grounds the officer relied on, the errors the Court found, and the paragraph where it said so. Free to read, no account.

845
decisions, each taken apart
436
refusals set aside
2022→
unbroken since
60
judges
63
countries of origin

Officer refusal decisions only — not tribunal appeals. Refugee, PRRA and RPD matters are deliberately outside this series.

The query nobody supports

Search by the officer's error, not by keyword.

A refused study permit lands on your desk. You do not need cases about study permits — you need the cases where the Court held it unreasonable to discount a parent's affidavit of support, on facts like these.

No search engine indexes law that way, because the language varies with every judge. So the work gets done from memory, a precedent bank, and guessing at phrases. We read all 845 and wrote down what the officer got wrong.

Refusal groundOfficer error typeDispositionRemedyVisa postCountryJudgeYear

What the Court actually does

Officers lean hardest on the ground the Court is least interested in.

A generic "purpose of visit" finding carries 282 refusals. The Court engages it in 103 of them. And where it does engage a ground, the spread is wide — two thirds of study-comparison findings are set aside, against one in seven business-viability ones. None of this is visible in the reported decisions. It appears only once all 845 have been read the same way.

Refusal ground Officer relied Court engaged Set aside
Financial sufficiency 304 255 50%
Purpose of visit / generic 282 103 61%
Family ties 276 222 64%
Study plan & academic progression 184 147 65%
Employment & economic ties 138 77 58%
Eligibility / status compliance 124 99 30%
Career advancement & prospects 94 49 59%
Ability to perform the work 86 70 46%
Misrepresentation (s.40) 82 71 51%
Business viability / significant benefit 62 53 15%
Travel history 53 32 53%
Language ability for the job 41 35 34%
Similar studies at home 41 31 68%
Immigration status in country of residence 41 21 38%
Limited prospects at home 32 12 50%
Credibility / genuineness 30 11 45%

Officer relied — grounds the refusal rested on. Court engaged — grounds the Court addressed in its reasons. Set aside — of those engaged decisions, the share where the refusal was quashed; the disposition is of the decision as a whole, not of the single ground. Merits decisions 2022–2026; leave-stage dismissals are not counted. Grounds appearing in fewer than 15 decisions, and the residual "Other" bucket, are omitted.

Grounded in the record

Nothing here is invented, because it can only cite what is in the record.

A closed corpus of 845 decisions, not the open web. Every quote is checked against the text of the decision it came from. Every proposition carries a paragraph number.

The decision itself is always one click away, so the answer is never the last step — it is the shortest route to the page you were going to read anyway.

Where this sits

CanLII

Free, complete, authoritative — and it is text.

It has every one of these decisions. Ask it which turned on a financial-sufficiency finding, or which of those the Court quashed, and it has no answer. Not a failing — a library indexes documents, not findings.

Westlaw · Immigration Law Reporter

Human headnotes, written for every area of law at once.

Editorially excellent and genuinely expensive. Still organised around what the court held, not around what the officer did — so the search you need to run is not a search it supports.

General-purpose AI

Fluent, unbounded, and it will invent a citation.

Canadian lawyers have been sanctioned for filing authorities that did not exist. A model asked about immigration law reaches for plausible-sounding case names. This one can only reach for 845 decisions it has actually read.

Free, and not

Reading is free

The full text of every decision, the structured brief, the citable quotes, who cited whom, and the outcome tables above. No account, no trial, no card.

Your file is the paid part

Bring the client's record to the case law: this applicant's bank statements against this officer's exact finding, matched to the errors the Court has already held unreasonable — with the memo, the authorities and the pin cites written out.

Roughly two hundred of these are decided every year. We read all of them.

Search the library

Entific Research · Federal Court of Canada, 2022–present · Study permit, work permit, visitor visa