By the Connecting Odds Research Team

Key takeaways

Every listing in our October 16 sample of about 5,100 public jobs was imported from an outside board. None was posted directly by an employer on Connecting Odds. The public feed shows only the hiring employers chose to syndicate.
About half of the listings carry no salary field at all, and only about 1 in 5 carries a clean, usable pay range after checking. Even the visible market is only partly visible.
Contract roles are about 2 percent of the public feed. Interim and contract work, which often fills through referrals and staffing relationships, barely appears in public listings at all.
Senior and leadership titles together make up about 40 percent of public titles, while entry-level and junior titles are about 2 percent. Boards skew senior. Early-career hiring still runs heavily through campuses, internships, and referrals.

First, a correction to the story everyone repeats

Our first piece on [the hidden job market](/blog/hidden-job-market) started with the folk statistic that 70 percent of jobs are never advertised. We said then, and we say again here: that number has no solid recent study behind it. Treat it as an estimate people repeat, not as a verified fact.

What we can do instead is measure the public market properly and show where it stops. If the visible market is this thin in places, the case for a large off-market flow does not need a made-up percentage. The gaps in the public data make the argument on their own.

This piece uses our own listings. On October 16, 2026 we pulled the full Connecting Odds public jobs feed, about 5,100 listings from 560-plus companies. The feed re-orders while it is being read, so two pulls differ by a few hundred records. Where that matters we report shares rather than exact counts. The method is the same one behind our [Salary Guide 2026](/blog/salary-guide-2026-what-us-jobs-are-paying-right-now-koatq).

Finding 1: the public feed is 100 percent syndicated

Every single listing in the sample has the same source label: imported. The providers behind them are familiar names: Greenhouse, The Muse, Workday, Ashby, Lever, and Adzuna, among others. Zero listings in the sample were posted directly by an employer on this platform.

That matters for how you read any job board, not just ours. A public board is not a census of hiring. It is a syndication layer. An employer writes a role once, pushes it to an applicant tracking system, and that system decides where the role appears. Roles that never enter that system never appear anywhere public, no matter how many boards you check.

So the first limit of the visible market is simple: you only see the roles employers decided to put through the machine. A founder who hires a former colleague, a manager who promotes a contractor, and a team that creates a role for a person they already know produce no listing at all. There is nothing to scrape, nothing to alert you, and nothing to apply to.

Finding 2: even visible jobs hide their pay

About half of the listings in the sample carry no salary field at all. Of the half that do, many are partial, in the wrong currency for the location, or stated in a form that does not survive basic cleaning. After the same cleaning we applied in the Salary Guide, only about 1 in 5 listings in the wider sample carries a clean, usable pay range.

Think about what that means in practice. For most public listings, a job seeker cannot answer the most basic question, what does this pay, without applying, interviewing, and asking. The information asymmetry is not a side effect. It is the default state of the public market.

The hidden market works the opposite way. In a referral conversation, pay comes up early, because neither side wants to waste the other person's time. That is one practical reason experienced professionals prefer the referral path: the numbers are on the table from the start. Our free [Salary Explorer](/salaries/explorer) exists to close part of this gap for public roles, by showing triangulated ranges before you apply.

Finding 3: contract work is almost invisible in public listings

Only about 2 percent of the listings in the sample are labelled as contract roles. Add temporary roles and the share is still small. Full-time roles dominate the feed.

Anyone who has worked contract knows that does not match reality. Contract and interim hiring is a large, fast-moving part of the labor market, and it runs on staffing relationships, past performance, and who is available this week. By the time a contract need becomes a public listing, the agency or the manager has usually already called the people they trust.

For job seekers, the lesson is direct. If you want contract work, being findable and reachable matters more than refreshing boards. The roles move through people first and listings second, if they reach listings at all. For employers, it is the mirror image: posting a contract role publicly is still worth doing for reach, and posting on Connecting Odds is free, but the fastest fills will keep coming from known names.

Finding 4: public boards skew senior, so early-career hiring must happen elsewhere

About 20 percent of titles in the sample carry senior, staff, principal, or lead wording. Another 20 percent or so carry manager, director, vice president, head, or chief wording. Together, senior and leadership titles are roughly 40 percent of the public feed.

Entry-level, junior, graduate, and intern titles together are about 2 percent.

That split cannot mean companies hire forty senior people for every junior one. It means junior hiring uses different channels: campus recruiting, internship conversion, apprenticeships, and referrals from people who can vouch for someone with a short track record. Those channels leave almost no public trace.

If you are early in your career, this is the most useful number in this piece. The boards are not built for your stage. Your best routes are the ones that create evidence someone can point to: internships, project work, and a profile that states clearly what you can do. Our [2026 resume and job search playbook](/blog/resumes-and-job-search-the-2026-playbook-eb21q) covers how to build that evidence into a resume that screens well when you do use the public path.

What the data cannot tell us, stated plainly

Three limits, so nobody over-reads this piece:

We cannot count jobs that were never posted. By definition they are not in any feed. Anyone who gives you an exact hidden-market percentage is estimating, including us if we tried.
Our feed skews toward tech and white-collar roles, because those employers syndicate listings most consistently. Healthcare hiring, for example, is large in the wider economy and much smaller in this sample, as our [Q3 2026 hiring insights](/blog/q3-2026-hiring-insights-what-90-000-new-jobs-and-4-1-unemployment-mean-for-your--ey79s) showed from ADP data.
A listing is not a hire. Some posted roles are filled by referrals anyway, some are paused, and some stay open for months. In our sample only about 13 percent of listings were posted in the prior seven days. The public feed is a snapshot of open advertisements, not a flow of completed hiring.

How to work both markets at once

The practical conclusion is not to abandon boards. It is to stop treating them as the whole market.

Use boards for what they are good at. Salary research, spotting which skills employers name most, and roles at companies you do not know yet. In this sample the most-named skills were artificial intelligence, Python, and Go, which tells you where employer language is moving.
Make yourself findable for the other market. A clear public profile in the hiring manager's own vocabulary, an explicit note about what you are open to, and a way to message you without a paywall. That is the design we built into Connecting Odds: public profiles, free messaging between connections, and a chronological feed so you see when people you know move and hire.
Keep pay knowledge current. Check a range before every serious application, whether the role came from a board or from a person. The hidden market rewards people who can talk numbers early and calmly.
Invest in the twelve people who know your work. Part 1 made this point and the data supports it: the channels that fill senior, contract, and junior roles fastest are all relationship channels. A small number of warm, specific relationships beats a large, thin network.

The bottom line

The hidden job market is not a secret list somewhere. It is the hiring that happens before, beside, and instead of the syndication machine. Our own feed proves how partial the public picture is: fully imported, half without pay, almost no contract roles, and heavily skewed senior. Work the public market well, and build the profile and relationships that put you in the other one. That combination, not either half alone, is how most durable careers actually get built.

Frequently asked questions

Is it true that 70 percent of jobs are never advertised? Nobody knows the exact share, and we treat that figure as an unverified estimate. What we can verify from about 5,100 live listings is that the public feed is entirely syndicated, about half of it shows no pay, and contract and entry-level roles barely appear. A large off-market flow is the reasonable conclusion, not a precise percentage.

Why are there no employer-posted jobs in your sample? Every listing in the October 16 sample arrived through an outside provider such as Greenhouse, Lever, Ashby, Workday, The Muse, or Adzuna. Employers can post directly on Connecting Odds for free, but in this sample none had. That is itself a finding about how public boards get their inventory.

If boards are so incomplete, should I stop using them? No. Use them for research, discovery, and the roles that do get posted, and score your resume before you apply. Just do not mistake board coverage for market coverage, and put equal effort into being findable through people.

How can I make myself findable for roles that are never posted? Keep a public profile written in the language hiring managers use, state what you are open to next, make messaging easy, and stay in light contact with former colleagues who have seen your work. Those are the signals a referral conversation runs on.