Case File First: Why the Order You Build Evidence In Decides the Petition

Most people who start an EB-1A, O-1A, or EB-2 NIW petition begin the same way: they open a folder, and they start dropping in everything that might be relevant — CVs, award certificates, citation reports, reference letters, press mentions, whatever they can find. Only much later, often after a first analysis or a first RFE, do they ask which of the regulatory criteria each piece of evidence is actually supposed to support. By then the case has already taken a shape, and reshaping it costs real time.

The order matters more than most petitioners expect. A petition built criteria-first — decide which evidentiary categories you're pursuing, then gather evidence specifically to fill each one — produces a materially different, and usually stronger, record than a petition built evidence-first and organized afterward. This isn't a stylistic preference. It's the direct result of how USCIS evaluates these petitions, and it's the reason Merito's own architecture is built around a single structured Case File rather than a pile of uploaded files.

The criteria are fixed, and they're public

EB-1A extraordinary ability petitions are assessed against ten regulatory categories defined in USCIS's own Policy Manual — original contributions, published material about you, judging others' work, leading or critical roles, high remuneration, and the rest (USCIS Policy Manual, Volume 6, Part F, Chapter 2). O-1A extraordinary-ability nonimmigrant petitions draw on a closely related but distinct eight-criterion list under a separate part of the manual (Volume 2, Part M, Chapter 4). EB-2 NIW cases don't use a criteria checklist at all — they're assessed against the three-prong Dhanasar framework (substantial merit and national importance of the proposed endeavor, well-positioned to advance it, and balance-of-factors favoring a waiver) codified in Volume 6, Part F, Chapter 5.

The exact counts — 10 for EB-1A, 8 for O-1A, 3 prongs for NIW — aren't a rough approximation; they're the literal number of distinct evidentiary categories Merito's own scoring rubric tracks per visa type (apps/web/src/lib/marketing/stats.ts's getCriteriaCounts(), which reads the live rubric rather than a hardcoded guess). The reason this matters for sequencing: a petitioner who gathers evidence opportunistically — whatever they happen to have handy — almost never distributes it evenly across these categories. It clusters. Someone with a strong publication record tends to over-document "scholarly articles" and under-document "leading or critical role," not because the second category doesn't apply to them, but because they never framed their own career against it before deciding what to collect.

Deciding the target criteria first inverts that. Instead of asking "what do I have," you ask "what does each of these ten (or eight, or three) categories actually require, and which of my own achievements, reframed, could satisfy it" — a fundamentally different and more complete search.

Final merits is a second, separate question — and it changes what "enough" evidence means

For EB-1A specifically, meeting three of the ten criteria is not the end of the inquiry. USCIS runs a second-step "final merits determination" — a review of the record as a whole to determine whether it demonstrates sustained national or international acclaim, independent of which individual boxes got checked. This two-step framework traces to Kazarian v. USCIS and is now built directly into the same Policy Manual chapter (Volume 6, Part F, Chapter 2) as its own analytical step, not an informal gut check.

That second step is exactly why criteria-first planning beats evidence-first collecting. Three technically-met criteria supported by thin, single-source evidence reads very differently under final merits than three criteria each supported by two or three independent, well-documented items — even though both records "meet three criteria" in the narrowest sense. If you don't know in advance that final merits exists as a distinct review, you have no reason to over-invest in any one criterion beyond the bare minimum, and the totality of your record ends up looking exactly as thin as the least effort you put into your weakest included category. For a deeper walkthrough of how this second step actually gets applied, see the Final Merits Determination guide on Learn — it's worth reading before you finalize which criteria you're even pursuing, not after.

Where a free score fits before you've gathered anything

This is the specific gap Merito Score is built to close early. It's a short, self-reported questionnaire — no login and no uploaded documents required to see a preview — that scores your answers against the real EB-1A, O-1A, or EB-2 NIW criteria and renders the result as a dual-ring gauge: an outer ring for how many individual criteria your self-described evidence plausibly reaches, and a separate inner ring specifically for the final-merits question described above (/help/merito-score walks through exactly what each ring measures and why they're kept apart rather than blended into one number).

Because it needs no documents, it's genuinely useful before you've built anything — the criteria-readiness ring gives you an honest first map of which categories you're already close on and which you're not, which is precisely the ordering information a folder full of unsorted PDFs can't give you. It's not a substitute for real evidence gathering, and it says so plainly on its own results page — it's a self-reported orientation, not a document analysis. But orientation is exactly what's missing when someone starts a case by uploading everything first.

The Case File is the actual mechanism, not just a philosophy

The "plan first" argument above would just be generic advice if nothing in the product actually depended on it. It does. Every substantive Merito tool beyond Score — Case Canvas, Merito Audit, Officer Simulator, RFE Studio — reads from a single structured Case File, built once from your uploaded evidence and referenced by every tool afterward, rather than each tool separately re-reading your raw documents. This is a deliberate architectural choice, not an implementation detail: it's what lets Case Canvas render your entire case as one graph of criteria mapped to evidence, mark the items that back several criteria, and flag mismatches between a document's actual kind and the criterion it's tagged against — all without a fresh document scan on every view (/help/case-canvas describes the graph, radar, heatmap, and timeline views this single file structure enables).

The practical consequence: the shape of your Case File is set largely by the shape of the evidence you feed into it, in the order you feed it. If your first pass of uploads is unsorted and evidence-first, your first Case File — and your first honest look at where the gaps are — reflects that same disorganization. If your first pass is criteria-first, deliberately gathering toward specific gaps identified by Score or by your own reading of the ten (or eight, or three) categories, the resulting Case File already reflects a considered structure instead of an accidental one. Rebuilding a case's shape after the fact is possible — Canvas's Compare view exists specifically to show what changed between two versions — but it's strictly more expensive than getting closer to the right shape the first time.

Practical sequencing, not a rigid formula

None of this means real evidence gathering should wait for a perfect plan, and it's not a claim that any specific sequence guarantees a particular outcome — Merito is not a law firm and this is not legal advice about your specific case. What the mechanism above does support is a practical ordering: read the actual criteria for your visa category before you start collecting, take a free Merito Score pass to get an honest first-pass map of where your self-described record is already close and where it isn't, and let that map — not whatever happens to be sitting in your downloads folder — decide what you go looking for next. If you're weighing EB-1A against O-1A in the first place, the EB-1A vs. O-1A comparison guide is worth reading at this same early stage, since the two categories' evidentiary overlap is large but not total, and which one you're actually building toward changes what "criteria-first" means in practice.

This same ordering question comes up constantly in petitioner communities — Merito's community is a reasonable place to compare notes with people mid-case on how they sequenced their own evidence gathering, alongside eb1mentor.com's own writing on the subject; Why Strong EB-1A Evidence Still Needs a Final Merits Strategy covers the same final-merits gap from the strategy side, independent of any specific tool. If you're ready to see what a real Pro-tier case build looks like once the criteria map exists, the pricing page lays out what the trial includes.

FAQ

Do I need to pick one visa category before I do anything else?

You need a working hypothesis, not a locked-in decision. Merito Score supports EB-1A, O-1A, and EB-2 NIW separately and can be retaken for a different path any time — but even a first, tentative choice gives you a real criteria list to plan evidence gathering against, which an undecided "gather everything, decide later" approach doesn't.

Isn't more evidence always better, regardless of order?

Not in a way that matters for final merits. A record with excess, redundant evidence on two criteria and nothing on the rest reads as narrower under a totality review than a more modest record with real coverage of five or six categories. Volume without distribution doesn't move the final-merits needle the way people often assume.

What if I already have a folder full of unsorted evidence — is it too late to plan criteria-first?

No. The Case File pattern means reorganizing is a mapping exercise, not a re-collection exercise — Case Canvas is built specifically to take existing evidence and show you, per criterion, where it's thin. It's more efficient to plan before collecting, but "criteria-first" is a lens you can apply retroactively to evidence you already have.

Does a strong criteria-readiness score mean I'll be approved?

No — and Merito is deliberately built to never say so. Every result carries a standard disclaimer that it's a readiness review, not a probability estimate or legal advice. Criteria readiness and final-merits readiness describe how your record currently reads against published USCIS standards, not what any officer will ultimately decide.

This article is general information about how EB-1A, O-1A, and EB-2 NIW petitions are evaluated and how Merito's tools are built — it is not legal advice, and nothing in it should be read as an assessment of any specific case. Merito is not a law firm.