Big Sites Post 2.96 Openings Per Hire. Apply Where It's 1.14.

If you've sent forty applications into big-company portals this fall and heard nothing back, the August 2026 JOLTS data explains a lot of it. Worksites with 5,000 or more employees posted 243,000 openings and made only 82,000 hires, which is 2.96 openings for every hire. Sites with 50 to 249 employees converted at 1.14. Here is what that gap means, why it's worse than it looks, and how to rewrite your resume for the smaller teams that are actually hiring.

What the August 2026 JOLTS size table shows

In August 2026, the largest worksites turned about one opening in three into a hire. Sites with 50 to 249 employees came close to one hire per opening, and that band made roughly 19 times as many hires as the 5,000+ band (1,549,000 vs 82,000).

JOLTS (the Job Openings and Labor Turnover Survey) is the Bureau of Labor Statistics' monthly count of open jobs, hires and separations. noon.ai published the size-class breakdown behind this article. For the whole economy there were 7,079,000 openings on the last business day of August and 5,192,000 hires during the month, so 1.36 openings per hire. Against that average, the size classes split hard:

Source: BLS Table 7 (private sector, seasonally adjusted, preliminary) via noon.ai. Ratios are derived. The 1,000 to 4,999 band is left out here. BLS Table 7 has it.

At 5,000+ sites the openings rate was 5.2% and the hires rate was 1.9%. Plenty of postings, very little hiring. The band is also a small slice of the market: 243,000 of roughly 5.8 million private-sector openings. The ratio looks dramatic because those sites carry a lot of open requisitions and have low turnover, not because they account for much of the hiring.

Company size or site size? The correction that changes the advice

JOLTS sorts employers by establishment size, not company size. An establishment is a single physical worksite, so a 200-person Amazon or JPMorgan office lands in the 50 to 249 band. The real advice is "apply to smaller teams and sites," not "avoid big brands."

It's easy to read this table as "5,000+ employers," and that reading is wrong. The BLS wording is "establishments with 5,000 or more employees": a single site with that many people on it, like a headquarters campus. It isn't every company with a big headcount.

That gives you more room than the headline suggests:

• Regional offices of large companies can hire like mid-size employers, because the worksite is mid-size even when the parent company isn't.
• Headquarters-campus postings are the ones most likely to sit in the 5,000+ band, and in the slow-converting pool.
• The location field on the posting is your best clue to which band a job falls in. A bank job in a 150-person regional office and a bank job at its flagship tower are different bets.

Are big companies actually hiring in 2026?

Some are, but large worksites carry far more open requisitions than they fill. HubSpot, Workday, Netflix and Amazon all announced cuts after the August reference period. Openings on the books aren't the same as hiring.

Why open requisitions and layoffs show up together

It comes down to how the survey counts. JOLTS records openings on the last business day of the month and hires across the whole month. A requisition that stays posted for six months adds to openings six times and never adds to hires. Large sites also have low turnover, which is why their 1.9% hires rate sits so far below their 5.2% openings rate. The BLS noted that 5,000+ establishments "showed little or no change in job openings, hires, and separations rates" in August. So the gap isn't a one-month blip. It's how those sites run.

Then came the layoffs, all announced after August:

• HubSpot: about 660 positions, roughly 7% of its global workforce.
• Workday: 525 people (2.5%) on September 29, its second round this year after about 400 in February.
• Netflix: about 850 jobs (5%) on October 9.
• Amazon: about 1,000 on October 7.

Any req that stays live while a team shrinks makes the openings column look busy and does nothing for hires.

Ghost jobs make the large-company gap worse

A ghost job is a posting the company isn't actively trying to fill. Greenhouse estimates that 18% to 22% of jobs on its platform fit that description, including about 19% in the second quarter of 2026. Clarify Capital looked at more than 175,000 listings and found about 1 in 7 were ghosts, rising to 1 in 5 for senior and executive roles.

Indeed Hiring Lab measures the same problem through what it calls vacancy yield: hires in a month per opening at the end of the previous month. By that measure there were about 0.7 hires in August for every opening in July. The JOLTS size table tells the same story more sharply: 34 hires per 100 openings at 5,000+ sites, against 88 at sites with 50 to 249 employees.

Where tech and finance candidates are piling up

Finance and insurance employers posted 3.21 openings for every hire in August, against 0.81 in construction. At the same time, Refolk's index shows 1.61 open-to-work tech profiles for every finance one. If you come from either sector, the big-portal pool you're applying into is a slow and crowded one.

In Refolk's index of professional profiles, 1,555 US profiles in software development, IT services and technology-internet put "open to work" in their headline. In financial services and banking, the count is 967.

The employers in those samples matter. On the finance side they include JPMorganChase (3 of the 25 sampled), Capital One, Prudential, KeyBank and Navy Federal. On the tech side: Autodesk, PayPal, Zscaler and LinkedIn. These are large companies. If you're coming out of one, the reflex is to apply back into the same tier, because those are the names you know and the titles that match. That's my read, not a measured pattern, but it puts you in the same portals as everyone else leaving the same employers.

One caveat: the index counts only headline text. Plenty of people looking for work never change their headline, so treat these numbers as a comparison between sectors, not a measure of market size.

Should I apply to small or mid-size companies?

Yes, but aim at sites with 10 to 249 employees, not the smallest shops. The 1 to 9 band converts at 1.50 openings per hire, worse than both 10 to 49 (1.18) and 50 to 249 (1.14).

That surprises people who assume smaller always means faster. The BLS noted that in August the openings rate fell only in the 1 to 9 size class. My read is that the smallest employers are the most likely to post a role and then pull it. The 10 to 249 range is big enough to have a budget and small enough that an opening usually means a seat somebody needs filled.

How to put that into practice:

1. Filter by headcount first. Build your target list from employers in the 10 to 249 range in your function, not from a job board's "top companies" carousel. A 150-person satellite office of a large company counts.
2. Discount reposted and evergreen reqs. A posting that keeps reappearing, especially at a 5,000+ campus, fits the ghost-job pattern.
3. Be more skeptical the more senior the role. Clarify's 1-in-5 ghost rate for senior and executive roles means a director application at a big campus carries the most dead-req risk.
4. Keep a few big-company applications, but make them targeted. I'd rather send two or three tailored applications to large sites, with a referral where you can get one, than thirty cold portal submissions.

The hard part of moving toward mid-size employers is triage. There are many more of them, and each posting needs a careful read before you know whether you fit. Refolk scores how well you fit each posting before you apply, so your hours go to the roles in the 1.14 band where you're a strong match.

How to write a resume for a mid-size company

Write for a flat team. Lead with scope and end-to-end outcomes, not titles and reporting lines. A company of 50 to 249 people hires someone who can own a problem without layers of managers around them.

HubSpot CEO Yamini Rangan explained her company's cuts as building "a flatter organization," with fewer "management layers" and decisions moved "closer to the people doing the work." That describes how a mid-size company already works. The big companies are restructuring to look like the employers in the 1.14 band. Your resume should look like it already belongs there.

What follows is my reasoning, not survey data. I haven't found a verified study of how mid-size hiring managers screen resumes. But the logic follows from how these teams are built. With no layer of specialists around you, the reader wants proof you can cover more of the work yourself.

Five rewrites for a flat team

• Swap the title ladder for a scope line. "Senior Manager II, Platform Partnerships" means little outside the company that issued it. "Owned partner integrations end to end: sourcing, contracts, technical launch" tells a 120-person company what you'll do.
• Name the whole loop. Mid-size teams want to see that you scoped it, built it, shipped it and measured it. If your bullets only cover the middle step, add the ends.
• Cut internal vocabulary. Internal tool names, codenames and org acronyms read as noise to anyone who never worked there. Replace them with the public product name or a plain description.
• Show range across functions. If you wrote the SQL, presented to sales and fixed the onboarding flow, say all three. At a giant that reads as unfocused. At a 60-person company it's the job.
• State team size. "On a five-person team" signals you're used to having no backup, which is exactly the situation a mid-size hire walks into.

Before and after

Fill the brackets with your real numbers. A rewrite with a vague outcome still beats the original, but a real figure is what gets the call.

Doing this for every posting is slow, and it's where most people drift back to one generic resume. Refolk does that rewrite for you: paste the posting, and you get your own resume back reworded for that team's scope and vocabulary, plus a cover letter drafted from the same history. The facts stay yours. Only the framing changes for each employer.

FAQ

Are ghost jobs more common at large companies?

The studies don't say directly, because they don't break the rate out by company size. Greenhouse puts ghost jobs at 18% to 22% of postings on its platform, about 19% in Q2 2026. Clarify Capital found about 1 in 7 across more than 175,000 listings, rising to 1 in 5 for senior and executive roles. The size data points the same way, though: 5,000+ sites made 34 hires per 100 openings in August 2026, so a lot of their requisitions are not turning into hires.

Does a big company's smaller office count as mid-size in this data?

Yes. JOLTS measures establishment size, meaning one worksite, so a 200-person office of a large company counts in the 50 to 249 band alongside independent mid-size firms. The location on the posting tells you which kind of site you're applying to, and a smaller office may convert far better than the headquarters campus.

Should I stop applying to large companies entirely?

No. Rebalance instead. Put most of your effort into sites with 10 to 249 employees, where August conversion ran at 1.14 to 1.18 openings per hire. Keep a short list of large-company roles where you have a referral or a near-exact match. And don't swing all the way to the smallest employers: the 1 to 9 band converted at 1.50, worse than mid-size.