Applying for a job used to feel like a relatively understandable exchange.

A company had a need. A candidate believed they could meet it. The candidate submitted evidence of their experience. The company reviewed that evidence, spoke with a small group of people, and made a decision.

It was never perfect. Hiring has always involved uncertainty, bias, missed opportunities, and imperfect judgment. But the basic purpose was clear: Find the right person for the right role.

Today, that purpose has become buried beneath volume. Job seekers submit hundreds of applications and hear nothing. Companies receive hundreds or thousands of applications and struggle to identify meaningful differences between them. Candidates use artificial intelligence to tailor résumés at scale. Employers use artificial intelligence and applicant tracking systems to filter those résumés at scale. Applicants apply more broadly because the odds feel unknowable. Employers add more filters because application volume becomes unmanageable.

Candidates receive less information.

Employers receive more noise.

Both sides trust the process less.

Hiring is no longer functioning primarily as a matching system. It has become a congestion system. And the more each side tries to protect itself from the problems created by that congestion, the worse the system becomes.

The experience of looking for work has become strangely impersonal

Talk to almost anyone who has searched for a job recently and you hear variations of the same story. They carefully revise a résumé. They tailor a cover letter. They answer application questions already addressed in the résumé. They upload the résumé. Then they re-enter the same information into another form. They wait. Nothing happens.
So they apply again.

And again.

And again.

Eventually, applications stop feeling like thoughtful expressions of interest and begin to feel like lottery tickets.

The individual role matters less because the candidate has no reliable way to know:

  • whether the job is still open
  • whether the company intends to hire externally
  • whether a recruiter will review the application
  • whether an automated system will reject it first
  • whether the hiring timeline is real
  • whether the position has already been filled
  • whether anyone will respond

This uncertainty changes behavior.

When people believe a thoughtful application has only a tiny chance of being seen, submitting more applications becomes rational.

The problem is that when everyone behaves rationally in response to a broken system, the system becomes even more overwhelmed.

Employers receive more applications.

Recruiters face larger backlogs.

Hiring teams add more filters.

Candidates receive fewer meaningful signals.

The cycle repeats.

Ghosting destroys information

There is a tendency to discuss employer ghosting as a matter of etiquette. It is more serious than that. Ghosting removes information from a system that already has too little of it.

Greenhouse reported that a majority of job seekers had experienced ghosting during the hiring process, while earlier Indeed research found that 40% of applicants reported being ghosted after a second- or third-round interview.

That distinction matters. Silence after submitting an application is frustrating. Silence after several conversations is something different.

By that point, a candidate may have:

  • researched the organization
  • prepared work samples
  • rearranged their workday
  • taken time away from family
  • completed assessments
  • met with several members of the company
  • begun imagining what accepting the role would mean

The candidate has invested real labor. Then the company disappears.

No decision.

No explanation.

No closure.

The result is not merely disappointment. It teaches candidates that employer communication cannot be trusted. That changes how candidates behave the next time.

They invest less emotionally.

They apply more broadly.

They continue interviewing after accepting offers.

They assume silence means rejection.

They become less willing to believe timelines or promises.

Poor communication creates defensive behavior. Then companies interpret that defensive behavior as evidence that candidates are unreliable. Trust erodes from both directions.

Ghost jobs make the market harder to read

A healthy market depends on useful signals. Job postings are supposed to communicate demand. A listing tells candidates:

A company has a problem.

It has allocated resources to solve that problem.

It is prepared to hire someone.

But that signal has become unreliable. Greenhouse reported that approximately 18% to 22% of jobs posted on its platform in a given quarter were classified as ghost jobs.

A 2024 academic study using job-posting data estimated that as many as 21% of job advertisements may represent ghost jobs and suggested that these listings contribute to job-search fatigue while distorting broader labor-market signals.

Not every inactive listing is intentionally deceptive. Hiring plans change. Budgets get frozen. Leadership changes direction. Roles get filled internally. Priorities shift.

That is normal. The problem is leaving the signal active after the intent disappears.

A candidate cannot distinguish between:

  • a company actively hiring
  • a company casually building a talent pipeline
  • a company waiting for budget approval
  • a role likely to be filled internally
  • a position that no longer exists

All of those listings look the same. So candidates spend time responding to opportunities that may not be opportunities. That wasted effort has consequences beyond inconvenience. It creates false demand. It encourages mass applying. It makes job seekers question whether their lack of response reflects their qualifications or a market that was never accurately represented.

The system produces uncertainty, then treats the behavior caused by that uncertainty as a candidate problem.

More interviews do not automatically create better judgment

Hiring is difficult because hiring involves risk. A poor decision can be expensive. It can disrupt a team. It can reduce performance. It can damage morale.

It is understandable that employers want more information before making a commitment. But more steps do not always produce better information. Sometimes they only delay judgment.

Candidates increasingly describe processes involving repeated interviews, one-way video screenings, personality assessments, skills tests, take-home assignments, presentations, and unpaid project work before receiving an offer—or any decision at all. Reporting from TIME documented applicants moving through numerous interviews and assessments while hiring timelines continued to expand.

Every additional step should answer a specific unresolved question.

Can this person do the work?

Can they communicate effectively?

Do they understand the problem?

Can they collaborate with this team?

Do their expectations align with the role?

Instead, interview rounds often accumulate because no one feels confident making the decision.

Another stakeholder is added.

Another conversation is scheduled.

Another assignment is requested.

Another opinion is collected.

The process grows, but the signal may not improve.

This is analysis paralysis disguised as rigor. The organization is not necessarily learning more. It is distributing responsibility. No individual wants to make the wrong hire, so more people become involved.

More involvement creates more opinions.

More opinions create more uncertainty.

More uncertainty creates more process.

Eventually, the strongest candidate accepts another offer.

The organization concludes that hiring is difficult.

Unpaid work creates an uneven exchange

Work samples can be useful.

A portfolio reflects past work, but employers may reasonably want to understand how someone thinks about a specific type of problem.

The issue is not whether candidates should ever demonstrate their ability. The issue is proportionality.

A short exercise designed to understand a candidate’s reasoning is different from asking someone to produce a finished strategy, campaign, design direction, or operating plan.

The more closely an assignment resembles work the company could use, the more the exercise begins to feel less like evaluation and more like uncompensated labor.

Applicants are increasingly questioning extensive unpaid assignments, particularly when the scope is unclear and there is no assurance that the work will not influence an active project.

And a hiring process communicates how an organization values time. Not only employee time. Human time.

When a company asks ten finalists to complete eight hours of unpaid work, it may perceive the request as a reasonable investment from each candidate.

Collectively, however, it has transferred eighty hours of hiring cost onto people who do not work for the company.

That cost does not disappear. Someone absorbs it.

A better process asks:

What is the minimum amount of work necessary to evaluate the question we still need answered?

That is judgment under constraint.

Automation is solving the volume problem by creating a trust problem

Companies did not adopt applicant tracking systems and automated hiring tools for no reason. They face real operational pressure. When a role receives hundreds or thousands of applications, manual review becomes expensive and slow. Automation promises efficiency.

Applications can be sorted.

Qualifications can be identified.

Candidates can be ranked.

Interviews can be scheduled.

Screening can happen at any time.

Some of these improvements are useful. But efficiency is not the only measure of a successful hiring system.

Candidates increasingly suspect that applications are rejected before a person meaningfully evaluates them.

Near-instant rejection emails reinforce that belief. Whether that assumption is correct in every case is almost beside the point. The process provides so little explanation that candidates are forced to guess.

Research on rejected applicants has found that generic or absent explanations create frustration. Applicants want to understand why they were unsuccessful and what they could improve.

More recent research has also noted that applicants interacting with automated employment systems often lack transparency into how decisions are made and may distrust higher levels of automation.

The problem is not simply that software is involved. Software can improve a process.

The problem is opacity.

A candidate submits personal information into a system. The system produces a decision. The candidate receives no useful explanation.

That is not the way to start a professional relationship.

One-way interviews remove the thing interviews are supposed to provide

An interview should be a mutual exchange.

The employer evaluates the candidate. The candidate evaluates the employer. Both sides gather information. Both sides ask questions. Both sides decide whether continuing makes sense.

One-way video interviews weaken that exchange.

The candidate speaks to a screen. A question appears. A timer begins. The answer is recorded.

There is no conversation.

No opportunity to clarify.

No ability to read the room.

No chance to ask what the interviewer means.

AI-led interviews move this further. The candidate may interact with a synthetic voice or automated interviewer before speaking with anyone from the organization.

Supporters argue that these tools increase access, improve scheduling, and help employers process large applicant pools.

Those are real benefits.

But candidates increasingly describe automated interviews as impersonal and restrictive, and some are choosing to leave hiring processes that rely on them.

Recent reporting found that some applicants experience AI interviews as awkward or dehumanizing and question whether a person will ever review the interaction.

Again, the problem is not technology itself.

The problem is using technology to remove reciprocity from a process that depends on trust.

An interview should reduce uncertainty for both sides.

If it only collects information from one side, it is not functioning as an interview. It is functioning as a filter.

Employers are overwhelmed too

It would be easy to frame this entirely as employers mistreating candidates.

That would be emotionally satisfying. It would also be incomplete.

Companies are struggling with the same system.

Indeed reported that 56% of employers who describe hiring as difficult cite too many candidates as part of the problem. At the same time, 51% of job seekers also identify excessive candidate volume as a challenge.

That is revealing.

Both sides see the same congestion.

Employers receive:

  • applications from candidates who may not meet the basic requirements
  • AI-generated résumés optimized around job-description keywords
  • mass applications submitted with little intent
  • duplicated submissions
  • exaggerated qualifications
  • generic materials that make candidates harder to distinguish

The response is predictable.

More automation.

More screening.

More requirements.

More steps.

Less communication.

But each defensive layer creates a new candidate response. Candidates optimize résumés for automated filters.

They use AI to generate more applications.

They apply to jobs they may not fully want because the probability of receiving a response feels low.

They treat every opportunity as uncertain because the system gives them little reason to do otherwise.

Employers then receive even more volume and even weaker signals.

This is the loop.

Low trust creates defensive behavior. Defensive behavior creates more congestion. Congestion creates more automation. Automation creates less transparency. Less transparency creates lower trust.

No individual participant needs to be acting maliciously for the overall system to fail.

We are optimizing throughput instead of match quality

This may be the central design mistake. Modern hiring systems are often evaluated by operational metrics:

How many applicants entered the funnel?

How quickly were applications screened?

How many candidates advanced?

How long did the role remain open?

How efficiently did recruiters process volume?

Those metrics are understandable. But they can encourage organizations to optimize the movement of applicants rather than the quality of understanding.

A hiring process is not successful because it processes a thousand applications efficiently. It is successful because it helps the right company and the right person make a clear, informed decision.

Throughput and match quality are not the same thing.

A system can move quickly while producing poor matches. It can collect enormous amounts of information without improving judgment. It can eliminate candidates efficiently without understanding what it eliminated.

The purpose of hiring is not applicant processing. The purpose is confidence. The employer should become more confident that a candidate can succeed.

The candidate should become more confident that the organization is a place where they can succeed.

Every stage that does not improve one of those forms of confidence should be questioned.

A better hiring process would probably be smaller

The solution may not be another platform. It may not be more automation. It may not be a more sophisticated scoring system.

It may be a process with fewer promises, fewer steps, and clearer decisions.

A better system could begin with a few principles.

Post fewer jobs, but make every listing real

A job posting should represent current intent. If the role is exploratory, say so. If approval is pending, say so. If the company is building a future talent pipeline, say so.

Clarity may reduce application volume. Which is probably a benefit.

Tell candidates what the process is before it begins

How many interviews? Who will be involved? Will there be an assignment? How long should it take? When will the company make a decision? What criteria will matter?

Candidates should not discover a six-stage process one meeting at a time.

Use fewer interviews

Every round should answer a specific question.

If the team cannot explain what new information an additional interview will provide, the interview may not be necessary.

More opinions are not automatically better judgment.

Respect the cost of candidate work

Keep assessments focused. Use hypothetical problems when possible. Do not request work the company could reasonably use without compensation. Set clear time limits. Evaluate thinking, not endurance.

Communicate even when there is no news

Silence creates stories.

Candidates fill information gaps with assumptions. A brief update is better than uncertainty.

“We are still interviewing.”

“The timeline has changed.”

“The role is temporarily paused.”

Those messages are not difficult and they preserve trust.

Give meaningful feedback after meaningful investment

Not every applicant can receive personalized coaching. That expectation is unrealistic. But someone who completed several interviews or a substantial assessment deserves more than a generic rejection.

A few specific observations can create closure and preserve dignity.

Use automation to support human judgment, not avoid it

Automate scheduling. Automate reminders. Automate administrative work. Use software to reduce repetitive labor.

But do not confuse automated filtering with understanding. The more consequential the decision, the more important human accountability becomes.

Better hiring requires better judgment, not bigger systems

The hiring crisis is often described as a technology problem.

Or a labor-market problem. Or an applicant-quality problem. Or a recruiter-capacity problem.

It contains all of those things.

But underneath them is a problem of judgment. Employers are trying to avoid making bad decisions. Candidates are trying to avoid wasting effort. Both respond by increasing volume.

Employers create more process. Candidates submit more applications. The system grows larger while producing less confidence.

That is the paradox.

More activity. Less information.

More automation. Less trust.

More applicants. Worse signals.

The answer is not necessarily to process more. It is to decide better.

Good hiring requires the ability to:

  • define the real need
  • communicate it honestly
  • identify the evidence that matters
  • ignore information that does not
  • make reasonable tradeoffs
  • accept that certainty is impossible
  • treat people respectfully while making difficult decisions

Those are acts of judgment and they cannot be replaced by scale.

The healthiest hiring system may not be the one capable of processing the most applicants. It may be the one that creates the clearest expectations, asks the fewest necessary questions, communicates honestly, and helps both sides reach a confident decision.

Because hiring was never supposed to be an application-processing contest.

It was supposed to help people find each other.

And until we design the process around that purpose again, both employers and job seekers will continue optimizing against one another inside a system neither side trusts.