Lorem ipsum dolor sit amet, consectetur adipiscing elit. Suspendisse tincidunt sagittis eros. Quisque quis euismod lorem. Etiam sodales ac felis id interdum. Proin viverra nulla sem, vel molestie lacus volutpat nec. Integer ut bibendum erat.
Start for free
EB-2 NIW approval rates have fallen continuously since FY2022. Based on USCIS quarterly I-140 receipt and adjudication data, the trajectory:
| Period | Approval rate | Notes |
|---|---|---|
| FY2022 | ~95.7% | Pre-tightening baseline |
| FY2023 | ~80% | First sustained decline |
| FY2024 | ~71% | Some analysts report 43% with a different denominator |
| FY2025 (full year) | ~55.2% | Half of all FY2022 approvals |
| Q4 FY2025 (Jul-Sep 2025) | 35.7% | 5,356 denials vs 2,968 approvals (first quarter on record where denials outnumbered approvals) |
The inflection point. The sharpest decline accelerated after January 15, 2025, the effective date of USCIS Policy Alert PA-2025-03. That guidance clarified and tightened how USCIS officers apply the Dhanasar three-prong test. PA-2025-03 did not change the legal test. It changed how strictly the test is applied at the officer level.
These numbers are not cause for panic and they are not cause for complacency. Approvals are still happening at significant scale. But the evidentiary bar is materially higher than it was two years ago, and the kinds of cases that quietly approved on auto-pilot in FY2022 are now drawing detailed RFEs.
Three mechanisms compound in the data.
PA-2025-03 explicitly tightened the national-importance prong. Endeavor descriptions that read as field descriptions are now drawing RFEs by default. Substantial merit alone, without a sharp national-importance argument, no longer carries the prong. Officers are reading petitions more skeptically.
NIW became popular as a self-petition path during FY2021-2023 as wait times for other green card categories ballooned. The pool of filings grew, but the average evidence depth dropped. Petitions that would not have been filed under stricter market conditions ended up in the denominator.
Newer adjudicators trained under PA-2025-03 apply the test more uniformly across cases. The variance that existed in FY2022 (some officers approving thin cases, others denying them) compressed.
The base question shifted. In FY2022 the question was "Do I meet the test?" In 2026 the question is "Will the specific officer reading my file be convinced that my evidence meets the test under PA-2025-03?" Different question, higher bar.
Practical implications: invest in endeavor framing before you spend on prep, get an attorney consult before you commit to the category, and budget for the possibility of an RFE response even on a strong case.
RFE rates are up alongside denial rates. Most RFEs cluster around the national-importance prong and the well-positioned prong. RFE response is the attorney's job, but the underlying evidence work usually requires another round of recommender letters or independent expert validation. Plan for 4 to 8 weeks of response prep.
Anonymized case. A clean-water engineering researcher filed in March 2025 with an endeavor described as "developing improved wastewater treatment processes for urban environments." RFE arrived in August 2025. Response packet narrowed the endeavor to "deploying a specific membrane-bioreactor configuration at municipal scale in rural water districts where federal infrastructure grants require demonstrated technology readiness." Added two independent expert letters from people at unaffiliated water-engineering firms and one letter from a USDA rural utility program officer. Approved on RFE response.
The 35.7 percent Q4 FY2025 number gets a lot of attention. It is real, it comes from USCIS data, and it represents an actual shift in adjudication strictness. It is also not the same as "your individual case has a 35.7 percent chance of approval." Population-level rates mix strong and weak filings. Your individual rate depends on the strength of your specific evidence.
What the rates do tell you: cases that would have been borderline two years ago are now mostly being denied. Cases that would have been clearly strong two years ago are still mostly being approved, but they are drawing more RFEs along the way. The bar moved, the distribution did not move uniformly.
What the rates do not tell you: whether your specific endeavor framing will survive PA-2025-03. That is what the attorney consult and the eligibility check are for.
USCIS publishes quarterly I-140 receipt and adjudication data through its public reports portal. The percentages above are computed by applying consistent denominator methodology across quarters. Different denominator choices produce slightly different numbers, which is why FY2024 sometimes shows as ~71% and sometimes as ~43% depending on whether the denominator counts receipts, completions, or final actions. USCIS itself does not publish a single canonical NIW approval rate.
Unknown. Recovery requires either a policy reversal of PA-2025-03 (no signals of this) or a market correction where weaker filings drop out of the pool. The market-correction effect is already starting: NIW filings dropped about 18 percent quarter-over-quarter in Q4 FY2025 as practitioners advised clients to wait or pivot to EB-1A.
If you can plausibly clear 5 or more EB-1A criteria with deep evidence, yes. EB-1A is harder to qualify for in principle but easier to defend once the evidence is strong, because the test does not depend on the policy interpretation that PA-2025-03 tightened. EB-1A approval rates have held more stable.
Not at the I-140 stage. The same legal test applies regardless of country of birth. Country of birth affects priority date and wait time for I-485 (adjustment of status), particularly for India and China. China-born and India-born EB-2 NIW petitioners face years-long waits even after I-140 approval.
The eligibility check scores your endeavor framing against PA-2025-03 standards and shows whether your evidence supports the well-positioned prong before you spend money on prep.
Check my NIW caseApproval-rate data sourced from USCIS quarterly I-140 reports (uscis.gov/tools/reports-and-studies/immigration-and-citizenship-data). Methodology and denominator choices affect reported percentages. Population-level rates do not predict individual case outcomes. VisaNow.AI is not a law firm and does not provide legal advice. Each case is unique and must be evaluated by a licensed immigration attorney. Current as of May 2026. Nothing on this page constitutes legal counsel.