Analysis · October 19, 2026 · data as of October 19, 2026

The 20 biggest employers hold 9.9% of Hired Creative's listings

In the 2026-10-19 export the top 20 employers hold 9.9% of 20,671 listings, 4.5% within LinkedIn search rows and 20.2% within rows from other boards and applicant-tracking pages.

In the Hired Creative listings export of 2026-10-19 (20,671 listings, 9,127 employers by normalised company name), the 20 employers with the most listings hold 2,045 listings, or 9.9%. The largest single employer holds 0.96%. The export covers only the boards and company pages Hired Creative reads, so these figures describe our listings. They do not describe the creative job market, and they are not a measure of local concentration of the kind used in labour economics. How concentrated the listings look depends on where they were read, and that dependence is the main result.

Key points

• In Hired Creative's 2026-10-19 export, the 20 employers with the most listings hold 9.9% of 20,671 listings, and the largest holds 0.96%.
• Seven in ten of the 9,127 employers (70.7%, employers matched by normalised name) have exactly one listing in the export.
• The top-20 share is 4.5% among 9,945 LinkedIn search rows and 20.2% among 8,745 rows from other boards and applicant-tracking pages, so it depends on where listings were read.
• Rankings of regions or disciplines by concentration are not reported, because they cannot be told apart from which boards Hired Creative reads. The figures describe our listings only.

How the listings are spread across employers

The 20,671 listings spread over 9,127 employers. It takes 921 employers to account for half of all listings. If the listings were spread evenly across employers, the top 20 would hold 0.22%. They hold 9.9%, so the spread is uneven, but no employer holds more than 1%.

The 9.9% does not come from repeated rows or from single-employer feeds. On a copy without exact repeat rows (same employer, title and location text; n=19,996) the top-20 share is 9.7%. On a copy without rows from single-employer career-site feeds (n=18,690) it is 9.6%. These checks do not test a larger effect: which companies an applicant-tracking page covers is chosen in advance.

The share depends on where listings were read

Every listing reached Hired Creative through a feed. The top-20 share was computed again inside each group of rows, using the 20 employers with the most rows in that group.

Among 9,945 rows from LinkedIn search feeds, the top 20 employers hold 4.5%. Among 8,745 rows from other boards and company applicant-tracking pages, they hold 20.2%. A third group of 1,981 rows comes from single-employer career-site feeds. It is concentrated by definition, because a feed belongs to it only when one employer supplies at least 90% of its rows. Its share is therefore not reported as a finding and is left out of the chart.

Collection design pushes the two figures in different directions. A LinkedIn feed returns only what its keyword list asks for, 2 pages per keyword per location, which holds its top-20 share down. Applicant-tracking pages cover companies chosen in advance, which can raise theirs. The data do not show how much of the gap between 4.5% and 20.2% comes from collection and how much from the employers themselves. The contrast does not show that the market is more or less concentrated on one kind of board than on another.

Why regions and disciplines are not ranked

A region-by-region ranking changes with the feed type, so it cannot be told apart from which feeds covered each region. On all listings, Europe's top-20 share within the region is 20.2% (n=4,289) and Asia-Pacific's is 12.0% (n=3,529). Within rows from other boards and applicant-tracking pages the order reverses: Asia-Pacific 52.3% (n=576), Europe 29.3% (n=2,099). Within LinkedIn rows Europe falls to 7.8% (n=1,526). These are examples of the order changing, not regional findings.

Feeds for different regions also use different keyword lists and numbers of locations, so even rows from the same feed type are not like for like. Hired Creative treats gaps between regions as a source-mix effect. The analysis did not show otherwise. The same applies to the export's discipline field: disciplines differ in their feed mix, so no discipline ranking is reported.

Method

Data: the Hired Creative listings export of 2026-10-19, 20,671 listings, one row per listing. Every listing is in every overall figure.

Employer: the company string, lower-cased, stripped of accents and punctuation, with these suffix words removed: inc, llc, ltd, limited, plc, gmbh, corp, corporation, co, group, the, holdings, sa, ag, bv. This turned 9,170 distinct raw strings into 9,127 employers.

Top-20 share: the number of listings held by the 20 employers with the most listings, divided by all listings in the group being measured. Cumulative shares use the same rule for the top 1, 5, 10, 50, 100 and 500 employers. The share of employers with exactly one listing is out of all 9,127 employers.

Feed classes: a row is "LinkedIn search" if its source name starts with "linkedin". A source with at least 2 rows is a "single-employer career-site feed" if at least 90% of its rows are one employer. All other rows are "other boards and applicant-tracking pages".

Repeat rows: exact repeats share employer, title and location text.

Regions are the export's coarse region field, with worldwide and other listings pooled into one group. No country or city analysis was made. Countries with fewer than 30 listings, and any whose separate reporting could identify a sensitive location, are pooled.

Minimum sample: a share is reported for a region or discipline only when the group has at least 300 listings. Smaller groups stay in all totals.

Tests: two resampling checks with a fixed seed and 1,000 random draws each. One compares each region and discipline with random same-size draws from all listings (a permutation null). The other draws equal-size samples within a feed class so that groups of different size can be compared.

Limits

• This is a snapshot of one export, not of the creative job market. Which employers appear depends on which boards and company pages Hired Creative reads. About one row in ten (1,981 of 20,671) comes from a single-employer career-site feed, and many more come from applicant-tracking pages for a chosen set of companies. Those employers are present by construction.
• A critic can fairly say that a spread across a whole-world export says nothing about local concentration. A widely cited study of US vacancies on CareerBuilder.com from 2010 to 2013 (Azar, Marinescu and Steinbaum) defines a market as an occupation within a commuting zone and measures concentration there. That is a different question from a share of a global export, and this piece makes no comparison with it.
• Online postings can distort concentration even when coverage is close to complete. An NBER working paper (Dalton, Kahn and Mueller, September 2026) that matches US Burning Glass online postings to the Job Openings and Labor Turnover Survey reports that the postings overstate labour market concentration nearly threefold, because small and mid-sized establishments post less online. That is US data from a different source. The direction of bias in Hired Creative's feeds is not known and may differ by feed.
• On all listings, most regions and most disciplines show a higher top-20 share than random same-size draws from the whole export. That is expected when the feed mix differs by region and discipline. It shows a difference in our data, not a difference in the market. Excluding single-employer career-site feeds removes the gap for at least one region and one discipline.
• Employers are matched by normalised name only. Parent groups, subsidiaries and regional brands listed under different names stay separate, and unrelated employers that share a short or placeholder name could merge. Some employers in the top 20 are agencies or networks that hire for others, and the counts do not separate them.
• Repeats are caught only when employer, title and location text match exactly. The same job posted on two boards with different text is counted twice. Removing exact repeats moved the top-20 share by 0.2 points.
• The category field was not checked for precision, and the figures say nothing about wages, hiring behaviour or market power. Employers appear only as counts and shares, and none is judged. This is a single snapshot and shows no trend.

Sources

• Azar, Marinescu and Steinbaum, NBER Working Paper 24147, December 2017, revised February 2019
• Azar, Marinescu and Steinbaum, Journal of Human Resources 57(S), 2022, S167 to S199
• Dalton, Kahn and Mueller, "Do Online Job Postings Capture Job Vacancies?", NBER Working Paper 35697, September 2026
• Hired Creative listings export, 2026-10-19 (n=20,671 listings).

Published by Hired Creative from its own listings and the sources above. Spotted an error? Tell us.