Columbus Has 396 Job Postings. 363 of Them Are One Bank.

Every metro with at least 300 metro-tagged postings, at the 2026-08-13 16:30 UTC snapshot. The bar is that metro's top employer count divided by that metro's own posting count, so each percentage has its own denominator. Higher means your search in that city depends more on one company's hiring decisions.

Metros with at least 300 metro-tagged active postings at the 2026-08-13 16:30 UTC snapshot; 31 metros clear that bar and 22 are shown here. 'Postings' counts only rows that carry both a metro and an employer name, which is 55,867 of 247,242 active postings (22.6%), so these are not counts of every job in the city. 'Distinct employers' is the number of different employer names appearing in that metro. A posting is not an opening, and employers that publish one listing per location inflate their own share more than employers who post once, which is a bias that pushes concentration up rather than down.

Source: Seeker job corpus, 247,242 active job postings, snapshot 2026-08-13. seekerscore.com/insights/one-employer-metro-job-markets-2026

The finding: a metro's posting count hides how many employers are behind it

Rank metros by posting count and you get a familiar list. Rank them by how much of that count belongs to one company and the list rearranges completely. Columbus has 396 metro-tagged postings and 363 of them (91.7%) are JPMorgan Chase, spread across just 11 distinct employers. Philadelphia is 71.3% the same bank. Seattle, with ten times Columbus's volume, is 62.5% Amazon. San Diego is 47.6% Apple. Then the other end: Boston has 1,062 postings across 131 employers with its largest at 10.5%, Atlanta 13.8%, Denver 16.1%. The spread between the most and least concentrated metro we measured is 8.7 times. Two cities can show the same number of openings and offer a completely different number of actual chances.

This corrects a claim we published ourselves

Our companion metro page, published 2026-08-12 from a 30,000-row sample, led with 'Columbus outranks Seattle and Austin'. That page named the exact test that would falsify it: decompose the metro counts by employer and see whether removing JPMorgan Chase drops Columbus below Seattle. We have now run it on the full corpus and the finding does not survive. Columbus's entire metro-tagged inventory is 396 postings, of which 363 are one bank. Strip that employer and Columbus is 33 postings, behind every other metro in the table. Running the same comparison on the full corpus also inverted that page's other headline: it reported New York ahead of Washington DC and the San Francisco Bay Area combined, where the full corpus puts the Bay Area alone at 11,005 against New York's 6,912. Both claims have been corrected on that page, /insights/job-market-metro-concentration-2026, as of 2026-08-13, with the original figures kept beside the correction rather than deleted, because a retraction that removes the evidence teaches nobody anything.

Which 22.6% of the corpus this is, because it is not a random 22.6%

Only 55,867 of 247,242 active postings carry a metro, and it is worth knowing exactly which ones, because a reader could reasonably assume the missing three quarters are a geocoding failure spread evenly across the market. They are not. Metro resolution by source runs from 73.8% on apple, 69.1% on google, 65.2% on amazon, 58.3% on ashby, 56.1% on oracle_cloud and 50.5% on greenhouse, down to 0.4% on reed and exactly 0.0% on himalayas, jobicy and remotive. The three zeroes are not a bug: those are remote-only boards, and a remote job has no metro to resolve to. himalayas alone is 44.1% of the corpus, which accounts for most of the gap. So this page describes the located, mostly large-employer segment of the market and is silent about remote work by construction. Two consequences follow honestly. First, absolute posting counts per metro are a floor, not a census, and reed's 0.4% means United Kingdom inventory in particular is badly undercounted, so London's 3,747 is far below its real size. Second, because the enterprise applicant-tracking feeds resolve best, this view over-represents exactly the kind of large employer that produces high concentration. Columbus at 91.7% is a true statement about postings that can be placed in Columbus; it is not a claim that no other work exists there.

Why concentration matters more than volume for your search

Volume tells you how many postings exist. Concentration tells you how many independent decisions stand between you and an offer. In a metro at 91.7%, you are effectively applying to one hiring process: one set of recruiters, one interview loop, one compensation band, one internal reorganisation away from the market disappearing. In a metro at 10.5%, a rejection removes about a tenth as much of your available market. This also changes what research is worth doing. In Columbus, learning how JPMorgan Chase hires covers 92% of what is listed and is unambiguously the highest-value hour you can spend. In Boston, the same hour spent on AstraZeneca covers 10.5%, and breadth beats depth.

The concentrated metros are concentrated for a structural reason

The top employer in the most concentrated metros is almost never a local company. It is a large financial or technology employer running an operations hub: JPMorgan Chase in Columbus and Philadelphia, Amazon in Seattle, Apple in San Diego and Austin, Citi in Pune and Chennai. These are places where one firm put a campus, not places where an industry grew. That has a practical implication most 'best cities for jobs' lists miss: the concentration is a decision made in another city and can be reversed there. It also means the roles available are shaped by what that hub does, so a metro that looks like a technology market may in practice be offering operations, risk and support functions.

How we measured this, and what it cannot tell you

One snapshot of the jobs table at 2026-08-13 16:30 UTC, 247,242 active rows. 55,867 of those (22.6%) carry both a metro label and an employer name, and only those rows are used here; 62 distinct metros appear, and we report the 31 with at least 300 postings. Every percentage on this page divides a metro's own numbers by that metro's own total, never by the corpus. Employer names are normalised for case but not for corporate structure, so a parent and a subsidiary filing under different names count separately, which biases the employer counts up and therefore understates concentration. Finally, a posting is not an opening. Employers that publish one listing per location will overstate their own share, and that bias runs in the direction of the headline, which is why we report the number of distinct employers next to it as an independent check: Columbus at 11 employers is concentrated on a measure that does not depend on posting counts at all.

Frequently asked questions

Related Data

Guides

• Hidden Job Markets →
• Company Clusters →
• How Many Jobs Should You Apply To? →
• Job Market Trends →