Data report · Halloween 2026

The Ghosting Index, Halloween edition 2026: 80% of job applications are ghosted

By Bryan Vine · Published October 29, 2026 · Data through October 28, 2026

We followed 3,263 job applications for at least 30 days each. 79.9% never got a human response: no rejection, no reply, no interview invitation. Another 17.7% were rejected, and just 1.6% led to an interview.

• 79.9%of applications ghosted
• 30.7%got an automated receipt, then nothing
• 17.7%were rejected
• 1.6%led to an interview
• 5.9median days to a rejection

The body count

• Interview
• Other human reply
• Rejected
• Ghosted after an automated receipt
• Ghosted, total silence

Of 3,263 applications: 1.6% reached an interview, 17.7% were rejected, 0.8% got some other human reply, 30.7% got only an automated receipt and 49.2% got nothing at all. In total, 20.1% heard back from a person.

The most haunted hiring systems

Applications through Workable were ghosted most often (96.6% of 145). Greenhouse applications drew the most answers (66.1% ghosted), and most of those answers were rejections (31.3%). The tracking system does not decide whether anyone reads an application, but it does decide how easy it is for an employer to send a closing notice, and some systems make it close to a single click.

• Interview
• Other human reply
• Rejected
• Ghosted after an automated receipt
• Ghosted, total silence

Fast frights, slow resurrections

When a rejection came, the median wait was 5.9 days; a quarter took longer than 16.0 days (577 rejections). The first human reply (an interview invitation, an assessment or a recruiter writing back) arrived after a median of 5.7 days, and a quarter took longer than 18.3 days (87 replies).

• Fastest to reject: Ashby, median 5.5 days to a rejection.
• Slowest to reply: Ashby, median 7.5 days to a first human reply.

Dot: median days to a rejection. Bar end: 75th percentile. Only systems with at least 10 rejections from 5 or more people are shown.

Nobody is too senior to be ghosted

Seniority changes the odds less than you might hope: the ghost rate moves by 8.1 points across the levels we can show. Level is read from the job title (for example Senior, Staff, Director); titles with no level word are grouped on their own.

• Interview
• Other human reply
• Rejected
• Ghosted after an automated receipt
• Ghosted, total silence

Remote, hybrid, onsite: nowhere to hide

Remote applications were ghosted more often than onsite ones (80.6% against 74.8%). Remote postings draw applicants from a whole country or more, so each one competes in a bigger pile.

• Interview
• Other human reply
• Rejected
• Ghosted after an automated receipt
• Ghosted, total silence

Ghost stories by occupation

Occupation is read from the job title. Titles that match no specific role are grouped by their general word (for example Engineer).

• Interview
• Other human reply
• Rejected
• Ghosted after an automated receipt
• Ghosted, total silence

Is the haunting getting worse?

Each bar is a cohort of applications by the month they were sent, judged at the same cutoff. Every month shown had at least a 30 day head start.

• Interview
• Other human reply
• Rejected
• Ghosted after an automated receipt
• Ghosted, total silence

How to keep your applications out of the graveyard

• Give it three weeks, then move on. Three quarters of the rejections in this data arrived within 16.0 days. An application with no word after about three weeks is far more likely ghosted than pending, so keep applying instead of waiting on it.
• Apply early. Postings close fast; half of the ones we track are gone within weeks (see how long job postings stay open). An application read while the shortlist is empty has a better chance of an answer.
• Plan for volume. At a 1.6% interview rate, it took about 62 applications per interview invitation. Tailoring each application still matters; it is the volume that makes the math work.
• A receipt is not a reply. 38.5% of ghosted applications got an automated confirmation first. It only means the form went through.

How we counted the ghosts

The Ghosting Index is built from real applications sent through Reach, an auto-apply service that submits applications on behalf of job seekers and receives employer mail at a dedicated address per person. That mailbox is what lets us see what came back.

• Sample. 4,977 applications sent by 96 job seekers between June 11, 2026 and October 28, 2026. Rates on this page use the 3,263 applications from 57 people that were sent at least 30 days before the cutoff (the last on September 28, 2026), so every one had a fair chance to get an answer. Reach staff and test accounts are excluded.
• Ghosted means no human response within the window: no rejection, no reply, no interview invitation. An automated "we received your application" email does not count as a response; those applications are counted as ghosted and shown separately.
• Outcome order. Each application counts once, at the furthest point it reached: interview, then rejection, then any other human reply, then receipt only, then silence. An application rejected after an interview counts as an interview.
• Reading the mail. Employer mail is classified automatically. Mail the classifier could not place was re-read with stricter rules: newsletters, job alerts and marketing were dropped, rejections and receipts it missed were reassigned, and any message that arrived within an hour of sending was treated as an automated receipt. A reply counts only if it reads as being about the application.
• Applicant tracking system comes from the host of the application link (for example greenhouse.io or lever.co); email applications are their own group.
• Privacy. Only aggregates are published. No employer, job title or person is named. A group is shown only when it has at least 30 applications from at least 5 different people, and no single person (other than the consenting founder) makes up more than half of it. Smaller groups are folded into "Other" or left out.
• Limits. Reach users skew toward technology roles in the United States, and an auto-apply service sends more applications per person than most people send by hand. 97 rejection emails could not be matched to a specific application; if every one of them belonged to a ghosted application, the ghost rate would be 77.0% instead of 79.9%.

One account (the founder's own job search, included with his consent) made up 55.1% of the applications in the measured group. As a check, we computed each person's own ghost rate: across the 31 people with at least 10 measured applications, the median person was ghosted on 81.2% of them, close to the pooled 79.9%.

Each edition is a frozen snapshot: the numbers here will not change after publication. The next edition publishes in the first days of next month.

Cite this report