The real supply of remote jobs

In one industry, 12 employers held 363 open roles between them. About 15 were product jobs, and only 2 of those ever scored as a genuine match against a real resume, both at the one employer, out of twelve, that is fully remote.

What the sweep actually found

Twelve sports industry employers advertised those 363 roles across their own career pages. Most were never going to be a fit for a product person, since a company that size runs sales, support, operations, engineering, and a long tail of everything else alongside product. Narrowing to product roles, the kind a product manager, product designer, or product marketer would actually apply to, cut the pool to roughly 15 openings. Narrowing again, to roles that matched on seniority, function, and remote status all at once, cut it to 2.

Both of those 2 were at the same employer, and that was not a coincidence. It was the only one of the twelve that is fully remote. The other three named employers in the sweep were each tied to a specific place. FanDuel splits between New York and Atlanta. Peloton is New York. Hudl is Lincoln, Nebraska. Strava is San Francisco, and the role there was hybrid even so. Remote jobs that are quietly location locked goes through each of those four in more detail, including the specific reason each one holds its people to a place rather than letting them work from anywhere.

The remote flag is not reliable on its own

Part of why the real number is smaller than it looks is that postings misreport themselves. A listing can carry a remote flag and a workplace type field that says something else, on the same posting, at the same time. At one employer in this sweep, 21 of its 24 postings were flagged remote while the workplace type field disagreed. At another, it was 20 of 67. In both cases the workplace type was the honest answer, and the honest answer was not remote.

The likely cause is not deception. A remote flag often gets set once, early, when a posting first goes live, and is then left untouched even as the underlying workplace type field gets updated later by a different team or a different process. The two fields live on the same posting but are not always edited together, and nobody notices the gap until someone reads both side by side instead of trusting the one at the top of the page.

That is not a rare glitch on one company's careers page. It happens often enough that a search built only on the remote checkbox will overcount, sometimes by a wide margin, before you have read a single job description. Remote, hybrid and onsite: what a posting means walks through what each of those three words tends to hide, and which single field in a posting tells the truth even when the flags do not agree with each other.

Why the number is this small, and why that is not a broken search

Two real matches out of 363 postings looks, at first, like proof that something failed: the wrong keywords, the wrong boards, the wrong filters. In this case nothing about the search was broken. The industry itself is built around a handful of physical hubs, and nearly every employer in it hires according to that geography, remote flag or not. Your industry decides where you can live makes the wider case: it is usually the industry, not the job title, that decides whether remote work is realistically on the table at all.

That distinction changes what you do next. A broken search gets fixed with better keywords. A small industry gets addressed by widening the industry, widening the location you will accept, or accepting a smaller, slower pool of openings and giving each one real attention instead of speed.

What counted as a real match, and what did not

A posting scoring as a match here meant more than the word remote appearing somewhere in the listing. It meant the seniority matched what the resume actually showed, the function matched, and the location or remote status matched too, all at once, not one out of three. Plenty of the other 361 postings were fine jobs at fine companies. They simply were not a fit for this particular resume and this particular requirement, and no amount of rereading the listing changes a workplace type field back into a remote one.

That distinction matters for how a result like this should feel. Three hundred and sixty three postings sound like a market. Fifteen product roles sound like a shortlist. Two real matches sound like disappointment, until you see that the other thirteen were never going to work regardless of how the application was written, because the location was fixed before the req was ever posted.

Reading the real supply of remote jobs in your own search

A few things follow directly from a sweep like this one.

A search that combines remote, a senior title, and one specific industry can legitimately return almost nothing. That is a fact about the market you are searching in, not a failure of the person doing the searching or the tool they used to do it.

The remote filter on any job board is a starting point, never an answer. The workplace type field, read on the actual posting rather than trusted from a search result page, is the one worth believing.

Reading the source matters more than it sounds like it should. Aggregators copy postings and let the copies drift out of date long after the original changes or closes. Company job boards beat aggregators covers a different symptom of the same root cause, stale listings, but the fix is the same one that made a sweep like this possible at all: read the employer's own board, across Greenhouse, Lever, and Ashby, instead of a copy of it three steps removed.

None of this means remote work does not exist in a given field. It means the honest supply is usually smaller than a first search suggests, and knowing that going in changes how you plan the search rather than how discouraged you feel by its results.

None of it is permanent either. New employers occasionally decide to build a role, or a whole function, as fully remote, and a field that looks locked today can loosen within a year or two as more of its established players make that same choice. A sweep like this one is a snapshot of a single moment, not a verdict on an entire industry forever, which is one more reason it is worth rereading the workplace type field on a role every few months rather than trusting a memory of how the industry used to look.

iapplyai.app was built around exactly this gap. It sweeps a curated graph of verified employer boards and scores every posting against your actual resume, so a small real number shows up to you as a small, clear list, instead of 363 postings you have to sort through by hand to find the 2 that were ever going to work.