How Many Jobs Are Actually Remote in 2026? What Seven Years of Posting Data Shows

Every few months a headline announces that remote work is finished, and every few months another one announces that it is winning. Both get shared widely, usually citing a single company's press release. I wanted to know what the actual posting data says, so instead of reading the coverage I went and pulled the underlying numbers: Indeed Hiring Lab publishes a daily tracker of the share of job postings mentioning remote or hybrid work, open to anyone, going back to January 2019. That is 2,769 daily observations per country. I pulled the US series and the sector breakdown and did my own arithmetic on it. What came out was more interesting than either headline.

What this number actually measures, and what it does not

Before any conclusions, the caveat that most articles skip. Indeed's tracker measures the share of job postings whose text contains remote or hybrid work terms. It is a measure of what employers advertise, not of how people actually work, and not of how many applications those postings receive. It also lumps fully remote and hybrid together, so a role requiring three days in the office counts the same as one that never asks you to travel. That makes the tracker excellent for measuring change over time, since the methodology is consistent across all seven years, and poor for answering 'can I work from my couch.' I am using it for the first purpose only.

The headline: remote postings plateaued, they did not collapse

Here is the annual average share of US postings mentioning remote or hybrid work, computed from the daily series. The 2026 figure covers January through July.

The peak was February 2022 at 10.4%. The trough came in August 2024 at 7.8%. Since then the line has drifted gently upward, and July 2026 sits at 8.7%, which is slightly above where 2023 ended. Four years after the peak, remote and hybrid postings are down about 17% from the high and up about 235% from 2019. That is a plateau at a permanently higher level, not a return to the old normal and not a collapse. If you have been told that return-to-office mandates wiped out remote hiring, the posting data does not support it. Roughly one in twelve US openings still advertises flexibility, compared with one in thirty-eight before 2020.

The flat national line is hiding a large reshuffle between industries

This is the part I did not expect. When I broke the US data out by occupational category and compared July 2026 against the February 2022 peak, the aggregate stability turned out to be a coincidence of offsetting moves. Some fields gave up a great deal of remote hiring. Others quietly gained more than they ever had at the height of the pandemic.

Legal is now the most remote-friendly category in the entire US dataset, ahead of software development. I had to re-run that calculation twice because it contradicted everything I assumed going in. Software gave up about a quarter of its remote posting share while legal, a field with a reputation for conservatism about where work happens, added roughly a third to its own. Civil and electrical engineering, which involve physical infrastructure, both gained more remote share than software retained.

The likeliest explanation is not that lawyers suddenly discovered flexibility. It is labor market power. Software hiring cooled sharply after 2022, and when employers hold the leverage they stop advertising concessions they no longer need to make. Legal, accounting, and the engineering disciplines have had persistent shortages of licensed and credentialed people, and remote flexibility is one of the cheapest recruiting incentives available to an employer who cannot simply raise salaries. Remote work in a posting is not primarily a statement about whether the job can be done from home. It is a signal about who currently has the upper hand in that specific labor market.

Candidate interest in remote fell faster than employer supply

Indeed publishes a second series alongside the postings one: the share of searches on the site that use remote or hybrid terms. It measures what job seekers are asking for rather than what employers are offering. The two lines are worth reading together.

2023 is the only year in the entire seven-year series in which the search share exceeded the posting share, and it is the year the shortage narrative hardened into conventional wisdom. Since then, searches using remote terms have fallen further and faster than remote postings have. In 2025 the gap was the widest it has been since 2019, and it was employer supply running ahead of expressed candidate demand, not the other way around.

I want to be careful about how much weight this carries. A search is not an application, and it is entirely possible that people stopped typing 'remote' into the box because they concluded it was futile, or because remote filters became a standard checkbox rather than a search term. Discouragement and satisfaction produce the same number here, and this dataset cannot distinguish them. What I can say is narrower and still useful: the confident claim that a huge and growing pool of candidates is chasing a shrinking pool of remote jobs is not visible in Indeed's own demand-side data. Both sides have cooled.

Why 35% of people work from home but only 8.7% of postings offer it

This is the gap that matters most for anyone actually job hunting, and it explains why the posting data feels so wrong to people who look around their own office. The Bureau of Labor Statistics American Time Use Survey found that in 2025, 35% of employed people did some or all of their work at home on days they worked. Among people with a bachelor's degree or higher it was 51%, against 19% for those with a high school diploma and no college.

So more than half of degree-holding American workers do some work from home, while fewer than one in ten job postings advertises it. Those two facts are not in conflict. They describe two different transactions. Remote work is overwhelmingly something granted to people who already hold the job, negotiated informally, extended after a probation period, or simply inherited from a team's existing habits. It is far less often something an employer commits to in writing, in public, to a stranger, before knowing whether that stranger can be trusted to deliver without supervision.

What this changes about how you run the search

1. 1Check your own field's trend before you accept anyone's general advice. If you are in legal, accounting, or an engineering discipline, remote share is higher now than at the 2022 peak and you should be negotiating accordingly. If you are in software, marketing, or customer support, you are competing for a shrinking pool and should plan for hybrid.
2. 2Stop treating 'remote' as a hard filter and start treating it as a negotiable term. Apply to strong hybrid roles at companies within reach of you, then negotiate arrangement at offer stage, which is the point of maximum leverage and the same moment you should be negotiating salary.
3. 3Expect a wider applicant field on anything advertised remote. A hybrid role in Columbus competes against candidates who can drive to Columbus. A remote role competes against everyone in the country with a laptop, which is why remote listings close faster and go quiet more often.
4. 4Read the posting language closely. 'Remote' with a metro area attached usually means hybrid with occasional office days, and 'remote-first' at a company with one office usually means the office people get promoted. The distinction rarely appears in the job title.
5. 5Give yourself a realistic timeline. Filtering to a single-digit percentage of postings lengthens a search that already takes months, and the median US job search runs longer than most people budget for.

What this changes about your resume

When a role is advertised as remote, the employer is not only assessing whether you can do the work. They are assessing whether you can be trusted to do it unwatched, communicate without being prompted, and stay legible to a manager who will never walk past your desk. Almost nobody writes a resume that addresses this directly, which is exactly why doing so works.

• ▸Make prior remote or distributed experience explicit and specific. 'Remote (distributed team across 4 time zones)' after a job title does more work than a 'Remote Work' line in your skills section, which reads as a preference rather than a track record.
• ▸Show evidence of unsupervised output. Projects you scoped yourself, processes you documented, handoffs you designed, and anything you shipped while your manager was in a different country all speak directly to the risk the employer is actually weighing.
• ▸Quantify asynchronous impact, not just impact. 'Cut onboarding questions 40% by writing the team's first runbook' proves you make yourself less of a supervision burden, which is the specific fear behind remote hiring hesitancy. The general method is in our guide on quantifying achievements.
• ▸Handle your location line honestly. List your city and state, and add a short line such as 'Open to remote and hybrid roles nationwide' if that is true. Omitting your location entirely reads as evasive to recruiters, and many applicant tracking systems screen on location before a human sees the file.
• ▸Tailor harder than usual on remote postings, because the applicant pool is national rather than local. Matching the posting's exact language matters more when you are one of several hundred rather than one of forty.

This is the problem Resume Leap is built to solve: it reads the specific posting, including whether it is remote, hybrid, or onsite, and surfaces the experience that answers that particular employer's concerns into the top third of the page, instead of leaving you to guess which version of your history matters this time.