# The H-1B Sponsor Read, One Employer Scored to Target, Probe, or Drop

*Given one employer, your SOC code, metro, and offer, you will score six dimensions from public filings and reach a Target, Probe, or Drop call a second reader would match.*

- Canonical URL: https://www.refolk.ai/candidates/guides/h1b-sponsor-read-per-employer
- Pillar: Reading the market
- Format: Framework
- Published: 2026-11-06
- Last reviewed: 2026-11-06
- Reading time: 21 min
- Keywords: how to check if a company sponsors H-1B for my role, find H-1B sponsors by job title and city, H-1B wage level lottery odds for my offer, cap-exempt H-1B employers how to verify, LCA disclosure data how to read, does this company sponsor new grads on OPT

## Key takeaways

- Under wage-weighted H-1B selection, a Level IV registration gets four lottery entries and a Level I registration gets one, and DHS modelled Level I selection at 15.29% against 30.58% for Level II.
- When one person has several H-1B registrations, every one of them is weighted at the lowest wage level among them, so a single Level I registration from any employer drops you to one entry.
- The USCIS H-1B Employer Data Hub counts change-of-employer transfers as continuing approvals, so an employer that only hires people already on H-1B looks like it never sponsors when you read the initial column alone.
- A certified LCA shows intent, not a hire: score only Certified rows, and read the NEW_EMPLOYMENT or CHANGE_EMPLOYER count that matches your own petition path.
- In Refolk's index, 5 of the top 10 named employers in a sample of 25 US software engineers with H1B in their headline are staffing or IT-services firms, so high filing volume often reflects intermediaries, not end-client roles in your metro.
- A nonprofit is not automatically cap-exempt: an affiliated nonprofit qualifies only on a recognised basis such as a formal written affiliation agreement with a university, so ask to confirm it exists.

This is the read I run on one employer before spending an hour tailoring a resume for it. The question is whether this company will actually sponsor an H-1B for someone in my occupation, at my wage level, in my city. It is for anyone who needs US sponsorship: a first cap petition from F-1 OPT, a transfer from a current H-1B employer, or a move to a cap-exempt employer. You come out with six scored dimensions, a verdict of Target, Probe, or Drop, and the exact recruiter question to send when the record is ambiguous.

## What does a sponsorship read have to answer?

A sponsorship read has to answer one question: will this employer file the specific petition you need, for your occupation, at your worksite, at a wage level that gives you a real chance? "This company filed LCAs" doesn't answer that on its own.

Some terms, defined once:

- **LCA (Labor Condition Application):** the attestation an employer files with the Department of Labor before an H-1B petition. It names the occupation, the worksite, and the wage.
- **SOC code:** the Standard Occupational Classification code for your duties. It decides which wage table applies to you.
- **OEWS levels:** four wage thresholds, Level I to Level IV, set for each SOC code and area from the Bureau of Labor Statistics' Occupational Employment and Wage Statistics survey.
- **Cap:** the annual limit on new H-1B petitions, filled by a registration lottery. **Cap-exempt** employers file outside it.
- **Transfer:** a change-of-employer petition for someone already counted against the cap. It needs no lottery.

The read scores six dimensions. Each one proves something specific, and each has a typical way of misleading you.

| Dimension | What it proves | How it lies |
|---|---|---|
| 1. Role-and-metro LCAs | Intent to hire your SOC at your site | Withdrawn or unused LCAs, staffing volume |
| 2. Offer wage level | Your lottery weight on a cap petition | Ranges and multi-site roles fall to the lowest level |
| 3. Cap-exempt status | Whether the lottery applies at all | A nonprofit mistaken for exempt |
| 4. Initial vs continuing approvals | USCIS approved petitions on your path | Transfers hidden in "continuing", wrong tax ID |
| 5. Posting language | Stated policy for this requisition | Silence read as yes, stale boilerplate |
| 6. Your status path | The lottery and fee exposure you carry | Fee status moves with court orders |

You score each dimension Pass, Unclear, or Fail against an observable test. The verdict follows from rules you apply in order, so a second reader working from the same records reaches the same call. The rubric is in the scoring section below.

## How much does your wage level change your H-1B lottery odds?

Under wage-weighted selection, each registration gets one lottery entry per OEWS level the offer meets. Level I gets one entry and Level IV gets four. DHS modelled a 15.29% chance of selection at Level I and 30.58% at Level II.

| Level | Entries | DHS modelled selection | Multiple of Level I |
|---|---|---|---|
| I | 1 | 15.29% | 1.0x |
| II | 2 | 30.58% | 2.0x |
| III | 3 | ~46% | ~3.0x |
| IV | 4 | ~61% | ~4.0x |

The Level I and II figures come from the DHS final rule. The Level III and IV figures come from a secondary summary of DHS projections, and the multiples are derived. Treat the whole table as relative weights. Complete selection rates by wage level for the first weighted lottery had not been published, so nobody can give you a confirmed percentage for your level.

### How your offer is placed

USCIS places a registration at the highest OEWS level the offered wage meets or exceeds, for the relevant SOC code and the area of intended employment. An offer below the Level I figure is still treated as Level I. This weighting level is not the same thing as the prevailing wage determination the employer makes for the LCA, so the level printed on an employer's past LCAs is a clue, not your level.

Location changes the answer for the same salary. A $150,000 offer that sits at Level III in Austin can land at Level I in San Jose. That is why the read is per role, per metro, and never per company.

Some summaries describe the four levels as rough percentiles of the local wage distribution. I have not confirmed that against a DOL document. Use the actual numbers from the FLAG Wage Search instead of a percentile rule of thumb.

**17.7% - Share of selected registrations at OEWS Level I in the first weighted lottery**

Eligible registrations fell about 38.5%, from 343,981 to 211,600, yet the smaller pool did not rescue entry-level offers.

That number is the clearest sign of how much the weighting matters. A smaller field should have helped everyone. Instead the weighting more than offset it for Level I offers.

> **Watch out:** Your weakest registration sets your weight
>
> When one person has several registrations, every one of them is weighted at the lowest wage level among them, and the same rule applies to wage ranges and multi-site roles. One Level I registration from any employer pulls your Level III offer down to one entry.

> Applying widely can now lower your lottery odds, because your weakest registration sets the weight for all of them.

## Which public record answers which question?

Three free government sources cover the whole read. The DOL LCA disclosure files show intent by role, site, and wage. The USCIS H-1B Employer Data Hub shows petitions USCIS approved and denied. The OFLC Wage Search on the FLAG portal gives you the four wage levels.

### LCA disclosure files

DOL publishes these on its OFLC performance data page. Each quarterly release is cumulative for the fiscal year, so the fourth-quarter file is the full year. Every row carries a CASE_STATUS, the status of the last significant event or decision. The values include Certified, Certified-Withdrawn, Denied, and Withdrawn.

Each row also carries six count fields. Each field counts worker positions, and one LCA can mix several of them:

- **NEW_EMPLOYMENT:** the worker will begin employment with a new employer.
- **CONTINUED_EMPLOYMENT:** the worker continues with the same employer.
- **CHANGE_PREVIOUS_EMPLOYMENT:** a change to existing employment with the same employer.
- **NEW_CONCURRENT_EMPLOYMENT:** a second, concurrent employer.
- **CHANGE_EMPLOYER:** the worker is moving from another employer, which is a transfer.
- **AMENDED_PETITION:** an amendment to an existing petition.

Treating NEW_EMPLOYMENT as "a new hire who needed a cap slot" is an inference. The field also counts cap-exempt new hires. None of the record layouts I checked include a cap-exempt flag, because exemption is claimed on the petition (Form I-129), not the LCA. That gap is why dimension 3 is a separate check.

### USCIS H-1B Employer Data Hub

The Hub groups initial approvals, initial denials, continuing approvals, and continuing denials by fiscal year, two-digit industry (NAICS) code, tax ID, state, city, and ZIP code. Each count reflects USCIS's first decision on the petition. The detail that matters most: "continuing" covers every petition not filed as new employment or new concurrent employment, so **change-of-employer transfers are counted as continuing**.

Coverage starts at fiscal year 2009. The data runs about one quarter behind: first-quarter data arrives in April. Check the live page for the latest quarter it actually covers before you trust a recent count.

### OFLC Wage Search on the FLAG portal

DOL retired its old wage data site, and OEWS levels now live in the OFLC Wage Search on the FLAG portal. Each wage year runs from July 1 to June 30 and is built from the previous May's BLS estimates. Use "All Industries" for most employers and the ACWIA table for higher education.

#### Four layers of sponsorship evidence

1. **Posting language** - Stated policy for one requisition, cheap to change and often boilerplate
2. **Certified LCA** - Attested intent for a SOC, worksite, and wage, but not proof of a hire
3. **Data Hub approval** - A USCIS first decision on a real petition at a matched tax ID
4. **Approval in your path's column** - Initial for new hires, continuing for transfers, read for your role

*Each layer down is closer to an approved petition for someone on your exact path, and harder for an employer to fake.*

## How do cap-exempt employers and fee exposure change the read?

Cap-exempt employers skip the lottery entirely, so they go on their own track. On that track your wage level (dimension 2) no longer affects your odds, and the employer's exemption (dimension 3) becomes the thing to verify. Your status path then decides whether any fee applies.

### The four cap-exempt categories

- Institutions of higher education.
- Nonprofits affiliated with or related to an institution of higher education.
- Nonprofit research organizations.
- Governmental research organizations.

An affiliated nonprofit only has to qualify on one recognised basis. One basis is a formal written affiliation agreement with a university that sets up an active working relationship for research or education. Under that basis, a fundamental activity of the nonprofit must also directly support the university's mission. Research organizations need research to be a fundamental activity, not a side line.

The practical test is simple: a hospital with no written affiliation agreement is cap-subject, however nonprofit it is. Ask for the basis. If it is an affiliation, ask whether the agreement exists in writing.

### Fee exposure by petition path

Two fees have been in play, and their status has changed more than once. A $100,000 payment imposed by proclamation was extended by a second proclamation, but courts vacated the agency policies that collect it, and an appeals court refused to block that ruling. It never covered extensions, transfers, amended petitions, or change of status filed inside the US.

A separate $103,265 fee for certain cap-subject petitions has been proposed. Its comment period closed with no final rule published. DHS arrived at the figure by dividing $8,777,488,035 of costs it identified by an assumed 85,000 petitions a year. DHS's own analysis says the fee would hit 11,051 small entities, 76% of the small cap-subject filers it identified. A small employer that sponsors today may stop sponsoring new cap petitions if the fee is finalised.

| Path | $100k payment (vacated) | $103,265 fee (proposed) |
|---|---|---|
| F-1 change of status, cap | Not covered | Covered |
| Cap petition, consular | Covered | Covered |
| Transfer / change of employer | Not covered | Not covered |
| Extension | Not covered | Not covered |
| Cap-exempt employer | Unverified | Not covered |

If the proposed fee is finalised, cap-subject petitions would carry it while cap-exempt petitions and transfers would not. For a new grad, that pushes the best Target list toward universities, affiliated nonprofits, and research organizations. Treat this as a mechanism to re-check, not a settled fact: the rule may change before or after it is finalised.

## The scoring procedure, step by step

The full procedure takes about two hours for the first employer. Steps 1, 2, and 7 depend only on you, so later employers reuse them and take under an hour each. Run it before you tailor anything.

#### Per-employer sponsorship read

1. **Fix your SOC, metro, and offer floor** - Map your duties to a 2018 SOC code in O*NET OnLine, set the worksite metro area, and write down the low end of the expected offer. Done when you have one SOC code, one area, and one number, in about 20 minutes.
2. **Place the offer on the OEWS wage levels** - In the FLAG Wage Search, choose the current July-to-June wage year, enter your SOC and area, and use All Industries, or ACWIA for higher education. Done when Levels I to IV are recorded and your offer sits at the highest level it meets.
3. **Pull certified LCAs for your role and metro** - Filter the latest cumulative quarterly LCA file plus the prior full-year file by employer legal name, SOC, and worksite city, and keep Certified rows only. Done when you have NEW_EMPLOYMENT and CHANGE_EMPLOYER counts and the spread of wage levels, in 30 to 60 minutes.
4. **Read the Data Hub by petition path** - Search the employer and city, match the last four digits of the tax ID, and table initial and continuing approvals and denials for the last three fiscal years. About 15 minutes.
5. **Screen cap-exempt status** - Check whether the employer is a university, an affiliated nonprofit (ask for the written agreement), or a nonprofit or governmental research organization. Done when it is routed to the cap-exempt track or the cap track.
6. **Read the posting language** - Search the posting for "unable to sponsor" or "must be authorized without sponsorship now or in the future". Record it as offered, silent, or refused, in about 5 minutes.
7. **Map your own status path** - Decide whether you need a new cap petition, a transfer, or a cap-exempt petition, and check fee exposure for that path against a dated source. About 5 minutes.
8. **Score the six dimensions and call it** - Score each dimension Pass, Unclear, or Fail, apply the verdict rules in order, and record Target, Probe, or Drop. For a Probe, send the recruiter question before tailoring.

Some practitioner guides for immigration attorneys also choose the SOC before the wage lookup. Sponsor-list sites start from employer lists and never place your wage level at all. Start from yourself: the SOC and metro decide which rows in every later file are relevant.

> **Rule:** Score the floor, count only Certified rows
>
> If the offer is a range, score the low end, because ranges fall to the lowest level. In the LCA file, every count you use must come from a row with CASE_STATUS of Certified, never Certified-Withdrawn.

## The scoring rubric and verdict rules

Score each dimension Pass, Unclear, or Fail against the observable tests below, then apply the three verdict rules in order. A second reader with the same records should reach the same verdict. If they don't, one of you misread a column.

**Per-employer sponsorship scorecard**

```
EMPLOYER (legal name):
TAX ID, last four digits:
MY SOC:                     METRO / AREA:
OFFER FLOOR:                OEWS LEVEL MET:
LOWEST LEVEL ACROSS ALL MY REGISTRATIONS:
MY PATH: new cap petition / transfer / cap-exempt
DATE OF READ:     LCA FILE THROUGH:     HUB DATA THROUGH:

1. Role-and-metro LCAs (latest cumulative file + prior full year)
   Pass: Certified rows at my SOC in my metro, nonzero count in my path's column
         (NEW_EMPLOYMENT for new hires, CHANGE_EMPLOYER for transfers)
   Unclear: my SOC in another metro, my metro under a neighbouring SOC,
            or only staffing-firm rows
   Fail: no Certified rows at my SOC in the window
   Score:

2. Offer wage level (cap path only; transfer or cap-exempt = Pass)
   Pass: Level III or IV    Unclear: Level II    Fail: Level I or below
   Score:

3. Cap status
   Pass: exemption verified, or clearly cap-subject and I am not relying on exemption
   Unclear: exemption claimed with no named basis or agreement
   Fail: I need the exempt track and the employer fits none of the four categories
   Score:

4. Data Hub, last three fiscal years, tax ID matched
   (initial column for new hires including cap-exempt; continuing for transfers)
   Pass: approvals in my path's column
   Unclear: approvals only in the other column, or tax ID not matched
   Fail: no approvals in either column
   Score:

5. Posting language
   Pass: sponsorship offered    Unclear: silent    Fail: refused now or in future
   Score:

6. My status path
   Pass: no lottery and no fee exposure (transfer, extension, verified cap-exempt)
   Unclear: lottery, or a proposed or contested fee on my path
   Fail: a fee in force on my path that the employer says it will not pay
   Score:

VERDICT: Target / Probe / Drop
PROBE QUESTION SENT (dimension, date):
```

*Use one copy per employer. Fill in the inputs first, then score each line against its test.*

### Verdict rules, applied in order

1. **Drop** if dimension 5 is Fail, if dimensions 1 and 4 are both Fail, or if any two dimensions are Fail. A refusal in the posting drops this requisition only. The employer can stay on your list for other roles.
2. **Target** if no dimension is Fail, dimension 1 is Pass, and no more than one dimension is Unclear.
3. **Probe** in every other case. Send the recruiter question for each Unclear dimension and for the single Fail, if there is one.

A Level I offer at an employer with a strong record therefore comes out as Probe, not Drop. That is the right call: the question to ask is whether the role can be pinned to one site and one pay figure, or scoped at a higher level.

#### Record strength against offer level, cap path

Horizontal axis runs from Offer at Level I to Offer at Level III or IV. Vertical axis runs from Thin role-and-metro record to Strong role-and-metro record.

| Quadrant | What it means |
| --- | --- |
| Probe the level | Ask whether the role is a single site and single pay figure, or can be scoped higher |
| Target | Tailor and apply now |
| Drop | Unless you are on a transfer or cap-exempt path, spend the hours elsewhere |
| Probe the record | Send the recruiter question before tailoring anything |

*On a new cap petition, a strong role-and-metro record only becomes a Target when the offer also carries enough lottery weight.*

On a transfer or cap-exempt path, ignore the horizontal axis: dimension 2 passes by default and the record alone decides. Once an employer scores Target, that's where tailoring hours belong. [Refolk](/candidates) writes the resume from your own history, tailors it and the cover letter to the posting, and scores how well you fit, so running this read first costs less than an application that was never going to be sponsored.

**Recruiter question for a Probe verdict**

```
Hi [Name], I'm interested in [Role title] ([req ID]) at the [City] site. Before I apply, I want to check one thing so I don't take up your time on a mismatch.

I will need H-1B sponsorship: [a new cap petition from F-1 OPT / a transfer of my current H-1B / a cap-exempt petition].

[Dim 1] Has [Company] sponsored this kind of role at the [City] site, or only at other locations?
[Dim 2] Is the role budgeted at one pay figure for one worksite, or as a range across several locations?
[Dim 3] Does [Company] file H-1B petitions as cap-exempt, and on what basis: university, nonprofit with a written affiliation agreement with a university, or research organization?
[Dim 4] Does [Company] file new H-1B petitions for external hires, or only transfers and extensions?
[Dim 5] The posting doesn't mention sponsorship. Is it available for this requisition?
[Dim 6] If a new H-1B fee applies to my petition type, would that be a blocker for this role?

Thanks,
[Your name]
```

*Keep the opener and closing, and include only the lines for dimensions that scored Unclear or Fail. Send it before you tailor anything.*

## How this read goes wrong

Most wrong verdicts come from eight repeatable errors. Most of them make an employer look better than it is for you. One, misreading the Data Hub columns, does the opposite and hides a real sponsor.

1. **Counting LCAs as hires.** Certified-Withdrawn rows, and certified LCAs that are never used, inflate the count. *Check:* use NEW_EMPLOYMENT or CHANGE_EMPLOYER counts on Certified rows only, and confirm against Data Hub approvals.
2. **Treating Hub "initial" as lottery hires.** The initial column also includes cap-exempt and concurrent new employment. Meanwhile transfers sit under continuing, so an employer that hires only transfers looks like it never sponsors new people. *Check:* read both columns against your own path. For a transfer candidate, a high continuing count can be a real Target.
3. **Matching the wrong employer.** Hub rows are split by tax ID and ZIP code, and subsidiaries file under different legal names. *Check:* match on the last four digits of the tax ID, and pull the LCA rows under the same legal name.
4. **Overstating your wage level.** Ranges and multi-site roles fall to the lowest level, and so does a second registration from another employer. A $150,000 offer at Level III in Austin can be Level I in San Jose. *Check:* score the floor in the actual worksite area, and ask whether other registrations are planned on your behalf.
5. **Taking a nonprofit for cap-exempt.** Nonprofit status alone doesn't make an employer exempt. *Check:* name the category, and for an affiliate, confirm a written affiliation agreement exists.
6. **Staffing-firm false positives.** High LCA volume from consultancies doesn't mean there is a role at an end client in your metro. *Check:* look at the worksite city on each row, and whether the employer places workers at client sites.
7. **Stale data.** LCAs filed before July 1 use the previous wage year's OEWS levels, and the Hub runs about a quarter behind. *Check:* note the "data through" dates on your scorecard, and compare levels only within one wage year.
8. **Fee status drift.** The $100,000 payment briefly became collectable again while a court paused its own order. *Check:* look at the date of any fee source and prefer one dated after the latest court or rule event.

**5 of 10 - Top named employers that are staffing or IT-services firms, in Refolk's sample of US software engineers with H1B in their headline**

Sample of 25 profiles from Refolk's index; the staffing classification is Refolk's own.

Error 6 is the one that most often turns a sponsor-list search into wasted applications. A list sorted by filing count puts intermediaries on top. Your role-and-metro filter is what removes them.

## What does Refolk's index show about who signals sponsorship?

In Refolk's index of professional profiles, sponsorship signals in headlines are uneven across roles and cluster at intermediaries. Use this as a prompt to filter harder, not as a sponsor list.

| Title | Headline keyword | US profiles | Ratio |
|---|---|---|---|
| Software Engineer | OPT | 77 | base |
| Software Engineer | H1B | 93 | 1.21x SWE-OPT |
| Data Scientist | OPT | 22 | 0.29x SWE-OPT |

What it suggests:

- **H1B software engineers cluster at staffing firms.** In the H1B sample, half of the top ten named employers are staffing or IT-services firms, while the OPT sample is scattered. My inference is that volume sponsorship concentrates at intermediaries. That makes dimension 1's role-and-metro test more important, not less.
- **Visible OPT peers are thinner in some roles.** Data scientists with OPT in their headline number about a third as many as software engineers. Fewer visible peers means less informal evidence about which employers took someone on your path, so the filing record has to carry more of the read.

Limits: headline keyword matching is noisy, "OPT" can match unrelated words, and these are small counts. Read the ratios as direction, not size. To check your own segment, search profiles with your title and a sponsorship keyword in your metro, and note which employers repeat.

## Keeping the read current

Re-run the read whenever one of its inputs changes. The triggers are a new quarterly LCA release, a new Data Hub quarter, the July 1 wage-year rollover, a change in your own registrations, or a court or rulemaking event on fees. Your SOC, metro, and wage levels usually hold for a full wage year. The employer side moves every quarter.

Keep the scorecard with its "data through" dates. When a Probe answer comes back yes, change the dimension score, re-apply the rules, and record the date. A recruiter's written answer counts as evidence for that requisition only, not for the employer in general. When an employer moves to Target, Refolk tailors the resume and cover letter to that posting from your own history, so a recruiter's yes doesn't sit for days while you rewrite.

#### Before you record a verdict

- [ ] My SOC code, metro area, and the low end of the offer are written on the scorecard.
- [ ] The offer is placed against Levels I to IV from the current wage year in the FLAG Wage Search.
- [ ] If I have or expect other registrations, dimension 2 is scored at the lowest level among them.
- [ ] Every LCA count I used comes from a Certified row, in the column that matches my path.
- [ ] Data Hub rows are matched to the employer by the last four digits of the tax ID.
- [ ] I read both the initial and continuing columns for the last three fiscal years.
- [ ] Any cap-exempt claim names one of the four categories, and an affiliate claim rests on a written agreement.
- [ ] Staffing-firm volume is separated from direct roles at a worksite in my metro.
- [ ] Fee status for my path is checked against a source dated after the latest court or rule event.
- [ ] The verdict, all six scores, and the "data through" dates are saved with today's date.

## Frequently asked questions

### How do I check if a company sponsors H-1B for my role and city?

Filter the DOL LCA disclosure file by the employer's legal name, your SOC code, and your worksite city, keep only Certified rows, and read the NEW_EMPLOYMENT and CHANGE_EMPLOYER counts. Then check the USCIS H-1B Employer Data Hub to confirm petitions were actually approved, matching the last four digits of the tax ID. A certified LCA shows intent. A Hub approval shows a USCIS decision. You need both before you call an employer a Target.

### Does a company that sponsors H-1B also sponsor new grads on OPT?

Not necessarily. An employer can sponsor heavily through transfers of people already on H-1B and rarely file new cap petitions. In the Data Hub, transfers count as continuing approvals, so look at the initial column, and in the LCA file look at NEW_EMPLOYMENT counts at your SOC. Then check your wage level: a new grad's offer at Level I gets one lottery entry, while a Level IV offer gets four.

### What are my H-1B lottery odds at my wage level?

DHS modelled about a 15.29% chance at Level I and 30.58% at Level II. Secondary summaries of DHS projections put Level III near 46% and Level IV near 61%. Your level is the highest OEWS level your offered wage meets for your SOC code and area of intended employment. These figures are models, and complete selection rates by level have not been published, so use them as relative weights rather than promises.

### How do I verify an employer is cap-exempt for H-1B?

Confirm that it fits one of four categories: an institution of higher education, a nonprofit affiliated with or related to one, a nonprofit research organization, or a governmental research organization. For an affiliate, ask whether a formal written affiliation agreement with a university exists and whether a fundamental activity directly supports the university's mission. The LCA files do not appear to carry a cap-exempt flag, because exemption is claimed on the I-129, so ask the employer directly.

### Why does the Data Hub show few initial approvals for a company I know sponsors?

The Hub counts change-of-employer transfers as continuing approvals, not initial ones. An employer that mostly hires people already on H-1B will look low in the initial column while approving many transfers, so read both columns. Also check that you have the right legal entity. Hub rows are split by tax ID and ZIP code, and subsidiaries file under different names, so match on the last four digits of the tax ID.

### Does the $100,000 H-1B fee apply to my petition?

Check the date of your source first, because the fee's status has changed with court orders. In the sources behind this guide, courts had vacated the agency policies that collect the $100,000 payment. It never covered extensions, transfers, amended petitions, or change of status filed inside the US. A separate $103,265 fee was proposed but not finalised. It would not apply to cap-exempt petitions or to transfers of people already counted against the cap.

---

*From the Refolk guide library. I revise these guides rather than replacing them, so the current version is always at https://www.refolk.ai/candidates/guides/h1b-sponsor-read-per-employer*