Why recommendation letters get rejected: 20 reasons from RFEs
Twenty USCIS officers' formulations that strip weight from letters under the contribution and leading-role criteria - from the author's address to 'written by AI' - and what to fix in each case.
Author: Alina Kanametova- updated 8 min read
Key answer
Support letters are the most common and the weakest document in a petition: almost every packet has them, and in almost every RFE the officer explains why they prove nothing. Below are twenty such explanations from real 2023-2025 notices under the two criteria where letters matter most - original contributions and a leading or critical role - and what to fix in each case.
Source: community members' review of RFEs, NOIDs and denials from 2023-2025 (o1eb1.com), reassembled by the editors: the officers' wording is verbatim, the examples and advice are ours. Not legal advice.
Formal requirements: what 8 CFR 204.5(g)(1) asks for
Evidence relating to qualifying experience or training shall be in the form of letter(s) from current or former employer(s) or trainer(s) and shall include the name, address, and title of the writer.
Formally the rule is about experience and training letters, not recommendations. But some officers apply the same standard to every support letter - and the result looks like this:
Since these letters were missing the required address information of the authors, they do not meet the regulation at 8 C.F.R § 204.5(g)(1) and were given no credible value under this criterion.
The minimum that settles the question: full name, title and organisation, a physical address (not just e-mail), contact details, date, signature, letterhead - and the author's CV or a short biography attached.
Ten reasons under "original contributions"
-
No author details. A letter without an address, title, date or signature gets "no credible value" - see the quote above. Fix: the full details block plus an attached CV.
-
The author's expertise is not shown. The officer cannot see why this person is qualified to judge "major significance". Fix: one or two paragraphs on the author's qualifications tied directly to your contribution.
-
The author is from your circle of colleagues, clients or employer.
The shared relationships with the reference letter authors calls into question their ability to remain independent and objective.
One would assume that contributions of 'major significance' would garner attention from individuals outside your own professional network.
- Solicited letters cannot be the foundation of the case.
Letters, though not without weight, cannot form the cornerstone of a successful extraordinary ability claim.
Letters from individuals who have had a prior advisory, collegial, or collaborative relationship with you carry less evidentiary weight than independent advisory opinions.
- Conclusions instead of facts. Hyperbole with no mechanics of "why this matters".
Letters that lack specifics and simply use hyperbolic language do not add value and are not considered to be probative evidence.
The fix is the "claim → evidence → reasoning" format: what was done, which document proves it, why it matters to the field.
- Your personal contribution is not separated out. A company's projects and the fact of publication are not your contribution.
The founding of a business and the publication of an article do little to establish how these contributions are considered to be of major significance in the greater field.
-
Significance shown only inside the employer or its clients. The standard from Visinscaia v. Beers: influence must reach beyond one employer and its clients - adoption by others, citations, standards, licences.
-
No independent corroboration of the author's statements.
The statements made by the witnesses should be corroborated by documentary evidence in the record.
Fix: every strong claim in the letter points to a specific exhibit.
- The letters look templated or AI-written.
The letters are highly inflated, and they appear to have been written with artificial intelligence (AI) chatbots due to how the letters are similarly worded and structured.
- Reliance on unverifiable sources - social media, wikis, marketing pages, screenshots without a URL or date.
Wikipedia, web portals, domains, blogs, social media: there are no assurances about the reliability of the content from these open, user-edited Internet sites.
Ten reasons under "leading or critical role"
-
No author details - the same "no credible value" formula as above.
-
The letter is not from an employer, or comes from a contractor.
Evidence of experience must consist of letters from employers.
- The author is outside the company and does not know the hierarchy.
Many of the letters come from individuals outside of the companies themselves and therefore do not offer any insight into the petitioner's roles with his employers.
- Praise only, no details.
The letters do not provide probative details regarding how your role was considered leading or critical for the organization as a whole.
A "leading" role is a place in the hierarchy with managerial duties; a "critical" one is significant influence on the organisation's or department's outcome. The letter must show one of the two concretely.
- No comparison with peers.
The letters do not indicate how the petitioner was leading or critical when compared to others in similar, let alone, more senior staff.
- No organisational chart.
Please submit organizational charts that depict your position in relation to the company as a whole.
A separate trap: a project-level role when the officer wants influence on the organisation, department or division.
- Routine duties described.
Simply fulfilling employment obligations is a measure of competency... not necessarily how you performed in a leading or critical role.
Fix: two or three episodes of "what would have happened without you and what changed after", with metrics.
- Inconsistencies in title, dates or employer. In one case the officer listed four different job titles from different documents and cited Matter of Ho.
It is incumbent upon the petitioner to resolve any inconsistencies in the record by independent objective evidence.
- The organisation's distinguished reputation is not shown.
USCIS does not consider evidence about an organization, from an organization, related to its own distinguished reputation to be independent and objective evidence.
- The organisation cannot be verified externally, or does not fit the criterion. A company search finds nothing, or it is a government body:
USCIS is unable to evaluate the distinguished reputation of a government agency since government agencies hold a monopoly on their area of jurisdiction.
The template that shows
Officers compare letters with each other: headers, fonts, unfilled fields, paragraph order, phrasing.
We note that all of the submitted letters appear to use the same template while attempting to appear different.
The use of templated letters suggests that the language in the letters is not the authors' own.
The Policy Manual says the same in its own words:
Letters of support, while not without weight, should not form the cornerstone of a successful claim for this classification.
Submitted letters should specifically describe the person's contribution and its significance to the field and should also set forth the basis of the writer's knowledge and expertise.
The practical conclusion: every letter is written in its author's own words, with its own structure and examples; a shared "skeleton" is acceptable only as a list of topics.
How to reduce the "solicited letter" effect
- Instead of "Dear USCIS Officer", "Dear Immigration Officer" or "Dear Sir or Madam" - "To Whom It May Concern" or a subject line "Professional Reference for …"; no case number in the letter.
- The author describes how they learned of your work independently - through publications, a product, conferences - rather than "we worked together".
- Experts who never worked with you weigh more than colleagues; documents that existed before the petition weigh more than any letter.
- No single letter covers two criteria: an employer for the role, an independent expert for the contribution.
What a strong letter is built from: nine groups of statements
The review identifies nine themes through which an author unpacks a contribution. They are not phrases to copy but topics, each of which must be closed with a fact:
- Originality - what exactly was proposed first, which approach or method did not exist before you.
- Significance and impact - how understanding or practice in the field changed.
- Dissemination - who else uses it: organisations, countries, curricula, standards, tools.
- Influence on peers and the industry - follow-on work, licences, consultations, regulatory change.
- Measurable effects - percentages, sums, fewer errors, higher productivity.
- Recognition by the field - which authorities assessed the work, where it is called landmark, where you were invited to speak.
- Longevity - what still works years later, which sub-field grew from the contribution.
- Implementation - products, services, commercialisation, long-standing problems solved.
- The author's own authority - years in the field, positions, publications, roles in professional bodies.
What to avoid: "pleasant", "talented", "with great potential"; promises of future achievement instead of accomplished ones; "I have read his CV"; intra-university awards as the centrepiece; emotion and personal qualities outside the profession.
Checking a letter before it goes out
Details. Name, title, organisation, full address, contact details, date, signature, letterhead, the author's CV attached.
The author's competence. One or two paragraphs on why they can judge this contribution or this role.
Independence. How the author learned of the work; for the role criterion - confirmation that this is an employer with personal knowledge of the position.
Specifics. One or two contributions with the mechanics of novelty and scale; for the role - the place in the hierarchy, decision rights, two or three episodes with metrics, a comparison with peers.
Anchors. Every key statement points to an exhibit: publications, contracts, an org chart, KPIs, performance reviews, external sources on the organisation's standing.
Language. No regulatory wording repeated back, no phrases shared with other letters, no hyperbole; terms explained as to a non-specialist.
FAQ
Related
Recommendation letters: how to assemble them
Whom to ask, how many, how to spread them across criteria
Letter structure block by block
The six blocks of a strong letter and a skeleton
Tool: letter draft
Build a letter skeleton from your facts
USCIS wording on contributions
How the criterion reads in real notices
Official sources
- 8 CFR 204.5(g)(1) - requirements for experience letters
- USCIS Policy Manual, Volume 6, Part F, Chapter 2 - the section on letters of support
- Matter of Caron International, 19 I&N Dec. 791 (Comm. 1988) - solicited letters create no presumption of eligibility
- Matter of Ho, 19 I&N Dec. 582 (BIA 1988) - inconsistencies in the record
Reference material, not legal advice: officers reach different decisions on identical evidence.
Related materials - Gather your evidence
- Recommendation letters for O-1, EB-1A, NIW: a strong set
- Strong recommendation letter: structure by section
- Document Translations for O-1, EB-1, and EB-2 NIW Petitions
- Translator Certificate for O-1/EB-1/EB-2 Petitions
- How to Prove a Journal or Publication Is Authoritative
- Journal authority: impact factor, quartile, indexing
- Major Media or Scholarly Journal: Required Metrics
- Business plan structure for a visa petition: sections
- EB-2 NIW business plan: what it includes and how it works
- A US work plan for EB-1A: what 8 CFR 204.5(h)(5) requires
- Which boxes you are closing
- Is the case ready to file
Criteria breakdown
From the catalog
Documents and templates from the library
NIW Exhibit: Principal Investigator on Two ProjectsTemplate1 page anonymized access on request
USCIS Proposed Rulemaking - Employment-Based Immigrant Petition ReformsGuide1 page anonymized
Exhibit 13.1: Funded Research Project Proposal - Grant Evidence (NIW)Template1 page anonymized access on request
Exhibit 13.2: Final Report of a Funded Research Project (EB-2 NIW)Template1 page anonymized access on request
From community discussions
«And if there are no old articles, how suspicious will it be that I started publishing. Even if I keep a 2-month interval? Can the release of articles be tied to publishing, for example, of a unique method or a "contribution" to the community? For example, I published some unique method and it started getting cited in the media? For an O1 visa is that also critical or does it relate more to EB1?»
community member · from public community chats
«in our case the contribution to the industry was courses (two of them). We showed the course ratings on educational platforms, how many engineers bought the courses, and recommendation letters from companies that implemented changes based on these courses.»
Tanya · from public community chats
«If we're talking about contribution that can be confirmed by a significant number of citations or a developed method or a patent. What specifically counts as a "contribution" to the endeavor isn't in the requirements. But right now, because there are a huge number of applications and the quota is limited, if a petitioner doesn't have this kind of clear contribution but is just a great specialist, then the chance the approval goes to someone who has all this is much higher. Keep in mind that right now every applicant has to compete with a large number of top-class professionals from all over the world, and just being a very good specialist in your field isn't enough.»
Yernar · from public community chats
Personal opinions of community members from public discussions, not legal advice.
Did this answer help?
Open database - you can improve it
Found something off, or know what to add? Select text and hit "Suggest an edit", or propose an article/topic below.