881,000 applicants are still waiting

+robots reading your ads & AI’s job-market wash

Author

Keith Weightman
November 02, 2026

In partnership with

Signal Summary: AI agents are now vetting employers before candidates ever see the job. Meanwhile, 881,000 hourly applications vanished into ATS purgatory, and 9,000 execs swear AI won’t actually shrink headcount. Plus, a candidate who brought her own medical consult to a customer service interview.

First time reading? Sign up to get weekly signals.

All images hand-drawn with an iPad Pro, Apple Pencil, and Procreate.

HERE’S WHAT HAPPENED LAST WEEK ↓

SIGNAL 1
Your next candidate might be a bot with opinions

The headline: Gartner says machines now decide whether candidates even hear about your jobs. Your talent brand has a new audience, and it reads Reddit.

What’s really happening:

  • By 2029, roughly 1 in 4 candidates may have an AI agent job hunting for them

  • Application volume has more than doubled in two years (some firms say it’s nearly tripled, thanks to bots)

  • AI tools size up employers using news, Reddit, and employee reviews you don’t control

Translation: Your Glassdoor reviews just got promoted to lead recruiter.

Signal → Strategy: Audit what ChatGPT says about your firm today. Then publish structured, text-heavy content (transcribed recruiter videos, real pay data, placement stories) so the bots stop citing that 2019 Reddit rant.

SIGNAL 2
881,000 applications walked into an ATS and never came out

The scoop: Fountain analyzed 12.5 million hourly applications. Candidates show up ready to work. Employers show up eventually.

The damage:

  • Only about 1 in 12 applicants gets hired

  • 881,000 applications timed out in 2025 with no offer and no rejection

  • 25% of successful hires waited six weeks or more for an offer

  • Over half of hourly workers can’t go a month without income

The irony: Food service managers log twice the interview hours of any other sector and still have the lowest hire rate (4.1%).

Signal → Strategy: Call hourly-heavy clients and offer to clear out their timeout bucket. Pitch same-week dispositions, self-scheduling, and offers in days, not six weeks. The candidates they ghosted are your bench.

SIGNAL 3
AI jobs math: it’s a wash (allegedly)

Breaking: An NBER survey of about 9,000 executives says AI hasn’t moved headcount. Fine print: these are executive predictions, not job counts. Execs also predicted the metaverse.

The numbers:

  • 87% say AI had zero net effect on employment over the past three years

  • Execs expect a modest 0.4% headcount dip at their firms over the next three

  • 78% of firms use AI, but 2% of firms account for over half the spending

The twist: Same total jobs, completely different jobs. Researchers predict “major labor market churn,” and Harvard data shows postings falling for automatable roles while demand rises for AI-augmented ones.

Signal → Strategy: Focus on churn, not headcount. Map which client roles are getting automated versus augmented, then redeploy displaced workers into the augmented ones before a competitor does. Churn is billable.

RECRUITING CONFESSIONAL
“Great With People” Meets “Please Diagnose My Leg”

Recruiting Confessional A weekly series featuring anonymous stories from recruiting and staffing professionals. Submit yours here.

The Setup: I was interviewing for a customer service role. Her cover letter said she was “great with people.” More than once. Bold claim. We were about to test it.

The Twist: Before she even sat down, she stuck her leg out toward me and pointed to a scab. Clearly infected. Clearly picked at. Then she asked for my professional opinion: was it infected, and should she see a doctor?

The Aftermath: The interview technically continued. My ability to focus on her answers did not.

Have a story? Submit it here

Your company data could be worth $100k–$3 million

There's a new channel of revenue that has just been unlocked for companies with over 20 people that have been around longer than a year. 

Few know it, but AI labs now let you monetize your company’s operational data, and they're spending $15 billion this year on it to train the next frontier of LLMs. Your workflows, documentation, processes, and internal convos like Slack are extremely valuable and they're paying anywhere from $100k to $3 million for it per company.

Before you ask, all the data is anonymized and cleaned before it's sent to a Lab. Bezel (Y combinator backed) helps find a data partner that is the best fit for you.

SHARE THE SIGNALS
3 people you should share these signals with:

  1. Your marketing lead who still thinks the careers page is “fine”

  2. The client ops manager sitting on 400 unanswered applications

  3. That exec who says “AI will replace everyone” at every happy hour