Home › Methodology
Methodology and sources
Every number on this site comes from a public government file. This page lists the files, explains what each measure means and how our few derived numbers are calculated. Data last updated 2026-11-14.
Sources
| Data | Publisher | Coverage on this site | Update |
|---|---|---|---|
| Number of Service-wide Forms by Quarter, Form Status, and Processing Time | USCIS | Oct–Dec 2021 – Apr–Jun 2026 | Quarterly |
| Report to Congress: Number of Service-Wide Forms by Month (FY22 appropriations requirement) | USCIS | November 2022 – August 2026 | Monthly |
| N-400, I-130 and I-485 by USCIS field office or service center | USCIS | FY2019 – Apr–Jun 2026 | Quarterly |
| Pending employment-based I-485 inventory by priority date | USCIS | March 2024 – October 2026 | Monthly |
| Visa Bulletin | U.S. Department of State | December 2001 – November 2026 (295 bulletins) | Monthly |
| OFLC processing times (PERM, prevailing wage) | U.S. Department of Labor | 2020-11-30 – 2026-10-31 (59 monthly snapshots; older ones from the Internet Archive) | Monthly |
Three different "processing times"
- Median (quarterly report). Half of the cases completed in the quarter took less than this many months from receipt to completion. Used for the headline number on every form page.
- Average (monthly report to Congress). The mean months for cases completed in the month. A few very old cases raise it, so it is usually higher than the median.
- 80th percentile (USCIS Check Case Processing Times tool). How long it took to complete 80% of cases at a given office over the past six months. USCIS uses this tool, and a separate 93rd-percentile threshold, for case inquiries. We link to it but do not copy it, because USCIS publishes it only as a live lookup without history.
The same form can therefore show three different numbers. None of them predicts an individual case; half of all cases take longer than the median.
Derived numbers
- Change vs a year ago: latest quarterly median minus the median four quarters earlier.
- Backlog estimate (field offices): pending cases at the end of the latest quarter ÷ average monthly completions (approvals + denials) over the last four quarters. It shows how long the current queue would take at the recent pace. It ignores new filings, transfers between offices and changes in staffing, and it is not a USCIS figure.
- Visa Bulletin movement: difference in days between the same cell in two consecutive bulletins. "Current" and "Unavailable" are kept as such and never converted to dates.
- Average advance: months of priority dates gained per calendar year between the first and last dated bulletin in the window (1, 5 or 10 years). Retrogressions count.
- Priority date projections: the gap between your priority date and the current final action date, divided by the average advance. This is a straight-line extrapolation of the past, not a forecast. Visa availability, demand and policy decide the real outcome.
- Pending applications ahead: sum of the USCIS employment-based I-485 inventory for your category and country with an earlier priority date. USCIS withholds counts below 10 ("D"); we count each as 5. Consular cases abroad are not in this inventory.
Known limitations
- USCIS changes report layouts and form labels; we map them to stable series (for example, the N-400 military row, which before FY2025 carried the overall N-400 median).
- Office-level files sometimes have zero or suppressed values, and offices open, close or merge.
- Visa Bulletin tables were parsed from the State Department's HTML archive and checked against the published values; contact us if you find a mismatch.
Corrections
Found an error? Write to hello@processingtimes.app with the page and the official source. We fix data errors as soon as we can verify them.