13,597 New Business Filings Across 4 Cities We Track

US Tech Automations runs a daily clock against four cities' own open business-registration data. It is not a survey and it is not modeled — it is a running count of what each city's own dataset shows, captured every day and hashed so a later change is visible rather than silently overwritten. As of August 7, 2026, that clock has captured 13,597 new business registrations across the four metros we track.

These are new business registrations captured from each city's own open data, not a count of every business formed in America or in any city. A city that publishes only some licence types will show fewer registrations than actually happened there — these counts compare our coverage of each metro, not the whole economy of each metro.

Key Takeaways

• 13,597 new business registrations captured cumulatively across 4 tracked metros, from March 28, 2026 through August 6, 2026.

• 7,913 of those filings landed in the trailing 90-day window (May 9 – August 6, 2026) — a separate count from the cumulative total, not the same number viewed two ways.

• Los Angeles leads with 6,557 filings, a 48.2% share of everything the four tracked metros captured together.

• Construction is the largest single sector at 1,362 filings, a 16.1% share, in both the cumulative and the recent-window breakdown.

• New York City's latest filing on file dates to April 16, 2026 — its recent-window count reads 0, which reflects a stalled feed, not a stalled economy.

The Ranking, Cumulative and Recent Side by Side

The two columns below answer different questions and are never the same number: the cumulative count is everything captured since collection began, the recent count is only the trailing 90 days.

Los Angeles alone accounts for a 48.2% share of every filing the four tracked metros captured combined. That ranking covers only the four metro areas we collect daily — Chicago, IL; Los Angeles, CA; New York City, NY; and San Francisco, CA — not every metro in the United States, and it is not a claim about how many businesses started in the country as a whole.

New York City's row needs its own read: 303 cumulative filings against 0 in the trailing 90 days, with a latest filing dated April 16, 2026 — over three months before this edition's snapshot date. That gap is a feed that has gone quiet in our collection, not evidence that business formation stopped in New York City.

The four cities also do not publish the same kind of registration in the first place, so their raw counts are never a like-for-like comparison of formation activity on their own — they are a comparison of what each city's own open dataset shows us.

What Kind of Business Is Being Registered

San Francisco and Los Angeles publish a NAICS industry code on each filing; Chicago and New York City publish a free-text activity description instead. Records carrying a NAICS code are grouped here into 2-digit sectors, read directly from each record's own code — nothing here is our own categorization layered on top.

Construction holds a 16.1% share of every sector-coded filing, cumulative and in the trailing 90 days alike. Sector coverage sits at 62.0% of the cumulative total — the rest of the filings either lack a NAICS code entirely or come from Chicago and New York City's free-text activity fields, which are not folded into the sector percentages above.

The Recent-90-Day Sector Breakdown, on Its Own

The trailing 90-day window gets its own sector table rather than being folded into the cumulative one above, because the two periods are answering different questions and this report does not want a reader mentally averaging them into one number.

NAICS-code coverage in the trailing 90-day window runs slightly lower than the cumulative figure — 54.7% of recent filings carry a usable NAICS code, against 62.0% cumulatively. That gap is itself a coverage artifact worth naming: it does not mean fewer sector-coded businesses formed recently, only that a smaller share of the most recent batch of filings happened to carry a code when this edition was sealed.

The two sector tables in this report are never meant to be read as a before-and-after. Neither is a survey of every sector-coded filing that will ever exist for its window — each is a snapshot of what carried a NAICS code at the moment this edition was sealed, and a later capture of the same window could show a different coverage rate as more records get coded.

Why a Bigger Number Is Not the Same as More Business Formation

A city that publishes a broader slice of its licence and registration types in open data will show a larger count here than a city that publishes a narrower slice — even if the two cities have identical numbers of businesses actually forming. This report measures what four specific municipal datasets show us, not the underlying economy of any of the four cities. New York City's near-empty recent window is the clearest example on this page: it almost certainly does not mean business formation in New York City nearly stopped in the second quarter of 2026.

It means the specific dataset this clock reads from New York City has not been refreshed with new filings since mid-April.

A registration is also not the same thing as a business that opened its doors. It is a piece of paper: a new filing captured in a city's own open-data system. Nothing in this report says whether any given registration led to a business that hired staff, opened a location, or is still operating today.

Method in Plain English

The business-formation clock behind this report checks each of the four cities' own open municipal datasets daily and records every new filing it has not seen before — renewals are excluded, only new registrations count. Every count here is a verbatim count of distinct registrations the clock actually captured; nothing is estimated, modeled, or extrapolated. Each registration is counted once, at the moment it is first observed in a city's dataset.

This particular edition re-shapes one already-sealed capture of that clock rather than re-collecting anything itself — if the underlying capture changes, its own seal changes, and this report's seal refuses to silently follow along with an edit.

Our Search Console operating data is built from the same discipline: a daily capture, a seal, and a report that describes exactly what the capture shows rather than rounding it up to something bigger. The same rule holds even when a number is thin or unflattering — the same way our live blog-page experiment commits in advance to publishing whichever result its held-back control group actually produces, this report prints New York City's near-empty recent window instead of quietly narrowing the ranking table down to the three cities with fuller data.

Put the Formation Data to Work

A market-entry or location-strategy lead trying to gauge where new competitors and potential customers are showing up cannot get a same-format, same-window comparison like this from any single city's own portal. US Tech Automations builds the automation workflows that run this clock daily against four separate municipal systems and reshape the result into one consistent report every month, instead of someone manually pulling four different exports and reconciling four different formats by hand. Our permits-by-metro roundup applies the same daily-clock, side-by-side-cities approach to residential construction, if formation and permitting together are useful signals for the same market.

Frequently Asked Questions

Q: Which city is registering the most new businesses?
A: Of the four metros we track, Los Angeles leads with 6,557 cumulative filings and 3,608 in the trailing 90 days — a 48.2% share of everything the four tracked metros captured together. This is a ranking of our four tracked metros, not a claim about every US city.

Q: Does a higher registration count mean a city has more new businesses forming?
A: Not necessarily. Cities publish different licence and registration types in their open data, so a bigger number can mean a city publishes a broader slice of its data rather than that more businesses actually started there.

Q: Why does New York City show 0 filings in the recent window?
A: New York City's latest filing on file dates to April 16, 2026, well before this edition's 90-day recent window (May 9 – August 6, 2026) even begins. That reads as a dataset that has not been refreshed with new filings in our collection, not as business formation stopping in the city.

Q: Is a registration the same as a business that actually opened?
A: No. A registration is a new filing captured from a city's open data — a piece of paper. This report does not say whether any given registration led to a business that hired anyone, opened a location, or is still trading today.

Q: What counts as a "recent" filing versus a "cumulative" one?
A: Cumulative filings cover everything the clock has captured since collection began on March 28, 2026. Recent filings cover only the trailing 90-day window ending August 6, 2026. The two numbers answer different questions and are never combined into one figure in this report.

Method and Provenance

Every figure in this report is read directly from our own sealed business-formation clock, re-shaped from an already-sealed capture dated August 7, 2026. This is a census of four municipal open-data systems — Chicago, IL; Los Angeles, CA; New York City, NY; and San Francisco, CA — collected between March 28, 2026 and August 6, 2026, not a census of every business formed in the United States or in any single state. Every count is a verbatim count of distinct business registrations the clock actually captured; nothing is estimated, modeled, or extrapolated.

Source: US Tech Automations Research — Business Formation series, sealed 2026-08 edition, from our own daily business-formation clock.