How to Build an EB-1A Evidence Portfolio That USCIS Officers Can't Dismiss

An EB-1A evidence portfolio is a structured two-layer package — the petition brief (your legal argument) and the exhibits (your proof) — that must demonstrate extraordinary ability under the Kazarian two-step framework. You need at least three of ten criteria satisfied with documentary evidence, plus a final merits argument showing you are in the small percentage at the top of your field. The most common mistakes are unquantified claims, non-independent expert letters, and missing the Step 2 totality argument.

The Officer's Desk

Picture the USCIS adjudicator who will decide your case. She works at the Nebraska Service Center in Lincoln. Her name, for this exercise, is Officer Chen. She handles roughly 15 to 20 EB-1A petitions per week. She is well-trained, thorough, and not a domain expert in your field. She has never attended a machine learning conference, performed surgery, or managed an engineering team.

Your petition arrives on her desk: a manila envelope, potentially 300 to 500 pages of documents printed, tabbed, and organized. Or submitted digitally. Either way, she has a limited amount of time to assess whether the evidence in that package establishes that you are among the small percentage at the very top of your field.

She is not deciding whether you are talented. She is deciding whether your evidence, evaluated against a regulatory framework she knows extremely well, meets the legal standard for extraordinary ability.

That framing — the adjudicator's perspective — should govern every decision you make about what evidence to include, how to present it, and how to contextualize it.

The Fundamental Problem With Most Evidence Packages

Most EB-1A evidence packages fail not because the applicant lacks achievement, but because the achievement is presented without context.

Consider Priya Krishnaswamy. Priya grew up in Chennai, India, completed her MD at Madras Medical College, and did her oncology fellowship at MD Anderson Cancer Center in Houston. She stayed, built a research program in triple-negative breast cancer, and by her eighth year in the US had published 14 peer-reviewed papers, presented at ASCO (the American Society of Clinical Oncology), and was serving on two NIH grant review panels.

Her first attempt at EB-1A failed — not because she lacked credentials, but because her petition said things like:

"Dr. Krishnaswamy has published extensively in the field of oncology. Her research on triple-negative breast cancer has been widely recognized. She has served on numerous grant review panels."

Every word of that is true. None of it is evidence in the EB-1A sense. Officer Chen has no idea whether 14 publications is a lot or a little for an oncologist of Priya's career stage. "Widely recognized" is an assertion, not a fact. "Numerous" is vague.

Priya's second petition — after she rebuilt her evidence package — said:

"Dr. Krishnaswamy's 14 peer-reviewed publications have accumulated 423 independent citations as measured by PubMed. Her h-index of 11 places her in the top 8% of oncology researchers at equivalent career stage per the Clarivate Essential Science Indicators norms for clinical medicine. Her 2021 paper in the Journal of Clinical Oncology, 'Predictive Biomarkers for Response to PARP Inhibitors in Triple-Negative Breast Cancer,' has been cited 94 times in 36 months and is referenced in the current NCCN Clinical Practice Guidelines."

Same career. Completely different evidentiary picture. Second petition was approved.

The Two-Layer Architecture of a Winning Evidence Package

Every piece of evidence in your package should operate simultaneously at two levels:

Layer 1 — Criterion level: This exhibit demonstrates that I meet Criterion X.

Layer 2 — Totality level: This exhibit also contributes to the overall picture of a person whose sustained national and international acclaim places them among the small percentage at the very top of their field.

If an exhibit only works at Layer 1 without contributing to the totality narrative, it weakens the package by adding volume without adding story. If a piece of evidence contributes to the totality story without clearly supporting a specific criterion, it is decoration.

The best evidence packages are edited, not accumulated. Every exhibit earns its place.

The Architecture of the Exhibit Package

The Exhibit Index

Start with a single-page index listing every exhibit by letter (Exhibit A, Exhibit B...), a one-line description, and which criterion it supports. This is not optional. It is the roadmap Officer Chen will use throughout her review.

A well-constructed exhibit index looks like:

Exhibit A — Award Certificate, NSF CAREER Award (2022) — Criterion 1

Exhibit B — Documentation of NSF CAREER Award Selection Process (2022 press release, 1 in 847 applications) — Criterion 1 (corroboration)

Exhibit C — Letters from Journal Editors Confirming Peer Review Activity, Nature Chemical Biology and ACS Nano (2019-2024) — Criterion 4

Exhibit D — Peer-reviewed publications list with citation counts from Web of Science — Criterion 6

Exhibit E — Citation analysis memo placing h-index at 88th percentile for field/career stage — Criterion 5 and 6

Exhibit F — Organizational chart showing role as Associate Director of Research — Criterion 8

...

Every exhibit has a cover sheet with: what it is, which criterion it supports, and a one-paragraph explanation of why it is relevant. Officer Chen should never have to guess why you included something.

Primary Documents and Corroborating Evidence

Each claimed criterion should be supported by both a primary document and corroborating context.

Criterion 1 (Awards):

• Primary: The award certificate itself
• Corroboration: Documentation of how many people applied or were eligible, who administered the award, what the selection criteria were, how the award is regarded in your field

Criterion 4 (Judging):

• Primary: Letter from journal editor or conference chair confirming your reviewing activity
• Corroboration: The journal's impact factor or acceptance rate, a description of the review process, confirmation of how many papers you reviewed

Criterion 5 (Original Contributions):

• Primary: The published work, patent, or documented contribution
• Corroboration: Citation analysis, expert declarations describing why the contribution is significant, documentation that others have adopted or built upon it

Criterion 6 (Scholarly Articles):

• Primary: Reprints or printouts of published papers
• Corroboration: Publication metrics (impact factor, acceptance rate), citation counts

Criterion 8 (Critical Role):

• Primary: Organizational chart, employment letter, or position description documenting your role
• Corroboration: Letter from CEO or senior executive describing the criticality of your role, documentation of the organization's distinguished reputation

Criterion 9 (High Salary):

• Primary: Employment contract, W-2, or pay stubs showing compensation
• Corroboration: BLS OES data for your occupation and metropolitan area showing your percentile position

Quantification: The Non-Negotiable Foundation

If there is one principle that runs through every successful EB-1A petition, it is this: replace adjectives with numbers, and replace numbers with percentiles.

Here is the progression:

Level 1 (adjective — fails): "Dr. Torres has an impressive citation record demonstrating significant influence in his field."

Level 2 (number — weak): "Dr. Torres has 312 citations."

Level 3 (contextualized number — works): "Dr. Torres has accumulated 312 independent citations as measured by Web of Science. His h-index of 14 places him in the top 11% of researchers in biochemistry at the associate professor career stage per Clarivate ESI norms."

Every piece of numerical evidence needs its percentile. Every award needs its selection ratio. Every salary needs its BLS comparison. Every journal needs its impact factor and acceptance rate.

Miguel Ángel Ríos, an architect and urban planner from Mexico City who had been working in Chicago on a TN visa for six years, built his Criterion 5 case around a transit-oriented development methodology he had developed that had been adopted in three major city planning projects in Chicago, Mexico City, and Vancouver. His initial petition just described the methodology and attached project renderings.

His revised petition included: a letter from the City of Vancouver's Chief Planner specifically crediting Miguel's methodology by name, a citation in an Urban Land Institute white paper, an expert declaration from a tenured professor of urban planning at MIT comparing Miguel's approach to standard industry practice and explaining why it represented an advance, and a list of five subsequent projects by independent firms that had published credits to Miguel's methodology in project documentation. Each piece answered the same question: "How many others in this field have done something comparable, and how does Miguel's contribution compare to theirs?"

The Expert Declaration: Your Most Powerful Tool

An expert declaration is a formal letter — sworn to be accurate — from a recognized expert in your field who vouches for the significance of your work and your standing among peers. It is not a reference letter. It is not a character reference. It is expert testimony delivered in writing.

Independent expert declarations are nearly universal in successful EB-1A petitions. They perform a function that no other exhibit can: they give Officer Chen the domain expertise she does not have, delivered by a credentialed source she can assess independently.

Types of Declarations

Personal knowledge declarations: The expert knows your work firsthand — as a collaborator, supervisor, or close observer of your career. These are easier to obtain but carry less weight because the expert has an obvious relationship with you.

Independent expert declarations: The expert knows your work through your publications, contributions, reputation, or public record — not through a personal relationship. These are harder to obtain but significantly more persuasive, because they demonstrate that your reputation extends beyond your immediate professional circle.

You need both types, but prioritize independent experts.

What an Expert Declaration Must Contain

A compelling expert declaration has six elements:

1. The expert's credentials: Who is this person? Their full title, institution, publication record, awards, standing in the field. USCIS needs to be able to verify that this is a credentialed expert whose assessment carries weight. Four to six sentences minimum on credentials. The expert's h-index if applicable.

2. How they know your work: Be specific. "I have read Dr. Torres's 2019 paper in Nature Chemistry and the subsequent work building from it. I was on the program committee for ACS National Meeting in 2022 when Dr. Torres presented his research on enzyme design." Not: "I have been following Dr. Torres's career."

3. The specific contribution they are addressing: Name the work. Describe it in plain terms. This section should make the contribution accessible to Officer Chen without oversimplifying it.

4. Why the contribution is significant: Not just "this is important work" but: "This is significant because it resolved a 15-year open problem in the field. Three of the leading research groups in enzyme engineering, including the groups led by [Name] at MIT and [Name] at ETH Zürich, have incorporated Dr. Torres's approach into their subsequent publications."

5. Comparative standing: This is the most important part. "In my 28 years in biochemistry, I have encountered fewer than a dozen researchers at the assistant professor career stage who have made contributions of this caliber. Dr. Torres is, without question, among the top 5% of researchers in enzyme engineering globally."

6. A clear conclusion: A direct statement that the applicant has demonstrated extraordinary ability and that their standing in the field is at the national or international level.

Getting Independent Experts to Write for You

The most common obstacle: "I don't know anyone eminent enough who would write for me."

Here is the approach that works. Go through your field's leading figures — Nobel laureates, members of national academies, named professors at top research universities, distinguished engineers, department chairs — and identify those who are likely familiar with your published work or major contributions even if they do not know you personally. Someone who has cited your work, someone whose work you have cited, someone who attended the same conference where you presented.

Reach out with a brief, respectful email. Explain that you are preparing an EB-1A petition and that you believe your work in [specific area] may be known to them, particularly [specific paper or contribution]. Ask if they would be willing to provide an expert declaration. Provide a one-page summary of the contributions you would like them to address. Offer to draft the letter for their review and revision.

Most eminent researchers are aware of this practice — it is standard in EB-1A petitions. A meaningful fraction will decline. A meaningful fraction will agree. Target ten to fifteen to end up with five to seven committed experts.

Common Evidence Failures and How to Fix Them

Failure 1: Award without selection documentation

Every award exhibit must be accompanied by evidence of: who administered the award, how many people were eligible or applied, how the selection was made (peer panel? editorial committee?), and how the award is regarded in your field. Without this, USCIS will issue an RFE on Criterion 1 regardless of the award's real prestige.

Failure 2: Citation count without context

Raw citation numbers without field/career-stage benchmarking are nearly meaningless to a non-expert officer. Get a citation analysis from Web of Science, Google Scholar, Scopus, or similar. Calculate your h-index and compare it against ESI norms or similar published benchmarks for your field. This is one place where Lumova's benchmark analysis tool can help — it generates citation percentile context for your field automatically.

Failure 3: Salary evidence against wrong peer group

The comparison must be: same occupation code (BLS Standard Occupational Classification), same metropolitan statistical area (MSA). A software engineer in San Francisco at $350K total comp is not demonstrating high remuneration compared to US national medians — she needs to be compared to software engineers in the San Jose-Sunnyvale-Santa Clara MSA specifically.

Failure 4: Published material about your company, not you

Review every article you plan to submit for Criterion 3. Does it prominently feature your specific work and contributions? Or does it cover your organization, with you as one of many mentioned? If you cannot clearly argue that the article is substantively about your work, find better articles or strengthen the existing ones with supplementary materials.

Failure 5: Expert letters without credentials in the opening paragraph

A beautiful letter that opens with "Dear Officer, I am happy to write this letter in support of my colleague..." tells USCIS nothing about why they should trust this person's assessment. The opening paragraph must establish the expert's credentials. Every time.

Building Your Evidence Timeline

Evidence collection for a strong EB-1A petition takes six to twelve months for most applicants. Rushing this process is one of the most common mistakes. Here is a realistic timeline:

Months 1–2: Audit and planning. Identify your three to four strongest criteria. Pull preliminary evidence — citation counts, award certificates, publication list. Identify gaps and start developing strategies to fill them.

Months 2–4: Expert outreach. Begin identifying and contacting independent expert declarants. This process has a significant lag — experts need time to review your work, agree to write, draft the letter, and revise it to your feedback. Start earlier than feels necessary.

Months 3–5: Documentation requests. Contact journal editors for peer review confirmation letters. Request salary benchmarking data. Obtain organizational documentation. Order certified translations of any foreign-language documents.

Months 4–6: Petition brief drafting. Write the petition brief in parallel with evidence collection. The brief and the exhibits must cross-reference precisely — every assertion in the brief should cite an exhibit by label.

Month 6+: Review and filing. Final assembly, review by an immigration attorney if using one, and filing.

Ask Lumova to audit your current evidence against all ten criteria and identify your specific gaps. Lumova will show you exactly where your package is strong, where it needs development, and what evidence you should prioritize collecting next.

Start your evidence audit →

(Lumova is educational only, not legal advice.)

Frequently Asked Questions

Do I need to translate all my foreign-language documents?

Yes. USCIS requires certified English translations of any document not in English. A certified translation includes a statement from the translator attesting to their competence and the accuracy of the translation. Machine translations are not acceptable.

How do I document contributions that are under NDA?

This is a real challenge in industry settings. Strategies that have worked: general descriptions of the contribution's nature and scale without proprietary technical specifics, organizational documentation showing adoption (internal memos referencing the contribution, provided by your employer), expert declarations from senior colleagues who can speak generally about the significance of the work without disclosing proprietary details, and press coverage or patent filings that reference the work in non-proprietary terms.

Can I include social media metrics as evidence?

Generally, no as primary evidence. LinkedIn recommendations, Twitter/X follower counts, and similar social metrics are not professional publications and do not satisfy any of the ten criteria directly. They can potentially appear as supplementary context in an expert declaration or published material argument if relevant, but should not be stand-alone exhibits.

How do I get a citation analysis?

Web of Science, Scopus, and Google Scholar all provide citation counts. Web of Science is often considered the most credible in USCIS proceedings because it excludes self-citations and uses a consistent methodology. ESI (Essential Science Indicators) provides field-normalized citation benchmarks. For a formal citation analysis memo comparing your h-index to career-stage norms in your field, consider commissioning one from a research analytics firm or having an expert declarant address it specifically in their letter.

Is there a maximum number of exhibits?

No hard limit, but there is a practical one. A 400-page petition of evenly weak evidence is worse than a 100-page petition of strong, well-contextualized evidence. Edit rigorously. Every exhibit must earn its place.