Four Interviews. One Hire. Did Anyone Actually Learn Anything New?
American employers say candidates now go through an average of four interviews before receiving an offer.
That is actually down from five in 2023.
Yet in the same 2026 survey, 43% of hiring managers said hiring takes longer than it did two years ago, 44% said their companies currently have positions they cannot fill, and hiring managers reported that an average 21% of open positions close without anyone being hired at all.
Read those findings together and something interesting appears.
Companies have apparently removed an interview.
Hiring has not magically become easy.
Perhaps the industry's obsession with how many interviews a candidate completes has been aimed at the wrong variable.
The more useful question is:
What does each interview allow the organization to know that it did not know before?
Because four interviews can represent four distinct measurements of job-relevant capability.
Or four people can spend four hours asking variations of:
"Tell me about yourself."
Those processes have the same interview count.
They do not have the same information value.
And Talent Acquisition leaders who measure interview processes primarily by speed, candidate satisfaction or number of stages may be overlooking something more fundamental:
An interview is supposed to produce evidence for a decision.
If it does not, it may simply be expensive conversation.
The hiring market has created a strange contradiction
The current labor market should theoretically make some hiring easier.
LinkedIn reported earlier this year that U.S. applicants per open role had doubled since spring 2022.
At the same time, 66% of recruiters surveyed globally said finding qualified talent had become harder over the preceding year.
Those findings are not necessarily contradictory.
More applicants do not automatically produce more qualified applicants.
More candidates do not automatically produce clearer decisions.
And a larger applicant pool can actually increase the burden on a weak selection system.
Fresh August research from Express Employment Professionals and The Harris Poll illustrates the tension.
The survey of 1,006 U.S. hiring decision-makers, conducted May 13 through June 1, found that:
44% had open positions they could not fill.
43% said hiring takes longer than it did two years ago.
26% said their process takes at least four weeks when résumé review, interviews and final decisions are considered.
Hiring managers said an average 21% of open positions close without a hire.
Candidates undergo an average of four interviews before an offer, down from five in spring 2023.
These are self-reported survey findings, not audited operational data, and they do not establish why any particular vacancy remained open.
But they create a useful question.
If companies are interviewing fewer times and hiring is still difficult, perhaps interview quantity was never the primary problem.
Hiring has a measurement problem disguised as a meeting problem
Imagine a candidate interviewing for a product leadership role.
Their process looks like this:
Interview 1 — Recruiter
Background. Motivation. Communication. Career goals.
Interview 2 — Hiring Manager
Background. Motivation. Leadership style. Career goals.
Interview 3 — Peer Panel
Background. Leadership style. Stakeholder management. Career goals.
Interview 4 — Executive
Background. Motivation. Leadership style. "Culture fit."
Everyone leaves with an opinion.
The hiring manager likes the candidate.
One peer is unsure.
The executive "didn't feel enough executive presence."
Another interviewer thought the candidate was excellent.
Then everyone meets for a debrief.
What exactly happened?
Four conversations occurred.
But the organization may not have conducted four independent assessments.
It may have conducted one loosely defined assessment four times.
That distinction is the heart of interview design.
Selection science has known this for decades
Structured interviewing is not a new HR trend.
It is one of the more established findings in personnel selection research.
A landmark 1988 meta-analysis by Willi Wiesner and Steven Cronshaw reviewed 150 usable validity coefficients from published and unpublished interview research.
The authors found that interview structure substantially moderated predictive validity; structured interviews produced mean validity coefficients approximately twice those of unstructured interviews in their analysis.
Research since then has made the picture more nuanced.
A 2004 meta-analysis by Frank Schmidt and Ryan Zimmerman found mixed evidence regarding why structured interviews outperform unstructured ones. Interestingly, their analysis suggested that averaging three to four independent unstructured interviews could achieve predictive validity comparable to one structured interview administered by a single interviewer.
That is an important complication.
Unstructured interviews are not automatically useless.
Nor does adding structure magically make an interview excellent.
But a substantial research literature has continued to support interview structure as an important design characteristic.
A major review by Julia Levashina, Frederick Morgeson and Michael Campion synthesized decades of research on structured employment interviews and the mechanisms through which interview structure affects reliability, validity and other outcomes.
And a newly published July 2026 chapter in the Oxford Handbook of Personnel Assessment and Selection again summarizes the evidence around standardized questions, behaviorally anchored ratings and job-analysis-based design, concluding that structured interviews improve reliability, validity and fairness compared with less structured formats.
Nearly four decades after the 1988 meta-analysis, the core lesson remains surprisingly relevant:
How an interview is designed matters at least as much as the fact that an interview occurred.
What does "structured" actually mean?
This term gets watered down in corporate hiring.
A recruiter putting six questions into a Google Doc does not necessarily create a structured interview.
The U.S. Office of Personnel Management defines a structured interview as an assessment designed to measure job-related competencies by systematically asking candidates about past behavior or hypothetical situations.
At high levels of structure, candidates receive the same predetermined questions, and their responses are evaluated against common standards or proficiency benchmarks.
The important ingredients are not bureaucracy for bureaucracy's sake.
They are comparability.
If Candidate A is asked about conflict management while Candidate B spends 20 minutes discussing their favorite leadership book, comparing the two afterward becomes difficult.
If one interviewer considers a good answer "strategic" and another considers a good answer "scrappy," the organization may not actually share an evaluation standard.
If interviewers decide which competencies matter after meeting candidates, criteria can quietly move.
Structure attempts to constrain those problems.
Not by eliminating human judgment.
By giving human judgment something consistent to operate on.
Structure does not mean turning interviewers into robots
This is where practitioners sometimes push back.
They hear "structured interview" and imagine:
Candidate enters.
Interviewer reads Question 1.
Candidate answers.
Interviewer reads Question 2.
Humanity dies somewhere around Question 4.
That is not the only way to structure an interview.
OPM explicitly recognizes both behavioral and situational interview approaches, and structured interviewing can include predefined probing questions.
Recent research is also examining how structured probing can improve technology-mediated interviews rather than forcing candidates into rigid one-way experiences.
A well-designed interview can still contain:
conversation,
curiosity,
clarification,
rapport,
candidate questions,
follow-up,
and genuine human interaction.
The structure belongs primarily around what the organization needs to learn and how evidence will be interpreted.
Think of it less like a script.
Think of it like experimental design.
You can still have a human conversation.
You just shouldn't change the experiment every time a new participant walks into the room.
Four interviews should not mean four votes
There is another common failure hidden inside interview panels.
Organizations sometimes treat interviewers as jurors.
Each person meets the candidate.
Each develops a general impression.
Then everyone votes.
Yes.
No.
Strong yes.
Weak no.
Maybe.
The problem is that hiring is not supposed to be a popularity contest.
Different interviewers should ideally be collecting evidence against defined requirements.
That means an interviewer evaluating technical problem-solving may have extremely valuable evidence about technical problem-solving and very little evidence about executive stakeholder management.
Another interviewer may have the reverse.
If both simply cast one equal global vote, information gets flattened.
Even worse, interviewers may unknowingly evaluate the same thing repeatedly.
Five people assessing "communication" does not necessarily produce five times more information about communication.
It may simply produce five correlated opinions influenced by five different definitions of the word.
The design question becomes:
What evidence is each interviewer responsible for producing?
That is much harder than assigning people to calendar slots.
It is also much more useful.
The interview process should have diminishing returns
Suppose Interview 1 establishes that a candidate possesses Competencies A and B.
Interview 2 measures C and D.
Interview 3 validates E through a job-relevant work sample.
Then someone proposes Interview 4.
The correct question is not:
"Is four interviews too many?"
It is:
"What unresolved decision uncertainty will Interview 4 reduce?"
If nobody can answer, the interview may not belong in the process.
This creates a concept hiring organizations rarely discuss explicitly:
marginal interview value.
Every additional interview imposes costs.
Candidate time.
Interviewer time.
Recruiter coordination.
Calendar delay.
Potential rescheduling.
Additional data requiring reconciliation.
Another opportunity for inconsistent evaluation.
Possibly another period of candidate silence.
Those costs can be worthwhile when the interview produces meaningful new evidence.
But the burden exists whether the interview produces new information or not.
This is why "three interviews good, seven interviews bad" is too simplistic.
A six-stage executive process could be defensible if each stage evaluates materially different high-consequence requirements.
A three-stage process can still be badly designed if all three stages duplicate one another.
Count is a proxy. Purpose is the variable that matters.
Even the federal government is telling hiring teams to reconsider interview rounds
This year, the U.S. Office of Personnel Management issued updated assessment guidance as part of federal merit-hiring changes.
For federal attorney hiring, OPM encouraged agencies to use structured, job-related interviews alongside other high-quality assessment methods rather than relying on repeated interviews.
Its March 2026 guidance is unusually explicit:
"Numerous rounds of interviews are unnecessary and inefficient."
OPM recommends that agencies consider no more than three interview rounds, including a final leadership or culture/mission-alignment interview, while pairing interviews with other job-relevant assessment methods where appropriate.
That three-round recommendation is federal hiring guidance for a particular context.
It is not a universal scientific threshold for every employer or role.
A startup hiring its first CFO and a hospital hiring a nurse should not automatically use identical selection architectures.
But the underlying principle travels well:
Don't add interviews merely because you need more evidence. Ask whether another assessment method could produce that evidence better.
Sometimes the answer is not another interview
Consider what an organization actually wants to know.
Can the candidate write?
Give them a proportionate writing task.
Can they code?
Use a relevant technical assessment or work sample.
Can they analyze an unfamiliar business problem?
Consider a job-relevant case exercise.
Can they perform a physical or technical task essential to the role?
Observe the task where lawful and appropriate.
Can they navigate interpersonal ambiguity?
A structured behavioral interview may be useful.
OPM's current assessment guidance explicitly recognizes work samples, technical skills tests, structured interviews, simulations, situational judgment tests and other methods as distinct tools for measuring job-relevant capabilities.
Its assessment strategy materials also report estimated validity coefficients of .54 for work samples and .51 for structured interviews, compared with .38 for unstructured interviews, based on the research synthesis underlying those federal resources.
Those numbers should not be treated as immutable rankings for every job or modern selection context.
Validity depends on what is being measured, how the assessment is designed, the criterion used and the population involved.
Indeed, a more recent meta-analysis specifically examining constructs assessed through interviews emphasizes that interview validity varies according to what the interview is actually trying to measure.
That reinforces the broader point.
There is no universally "best interview."
There is only a better or worse match between:
what you need to know
and
how you attempt to know it.
Candidate experience and decision quality are not opposing goals
Hiring teams sometimes behave as though candidate-friendly design requires sacrificing rigor.
The research does not support such a simple trade-off.
OPM notes that structured interviews can provide candidates equal opportunities to present information and that job-related structure contributes to perceptions of fairness.
A major review of 145 applicant-reaction studies found that candidate perceptions of fairness are meaningfully related to organizational attractiveness, job-pursuit intentions, intentions to accept offers and intentions to recommend employers to others.
A three-year longitudinal study also found relationships between procedural-fairness perceptions during selection and subsequent offer acceptance, although the authors appropriately caution against treating all applicant reactions as simple predictors of long-term performance.
And more recent research examining rejection timing found that fairness perceptions shift as candidates move deeper into a hiring process, with later-stage candidates becoming increasingly invested in the outcome.
The practical implication is important:
A better-designed interview can simultaneously improve the evidence available to the employer and make the process more intelligible to the candidate.
Candidates can know what to prepare for.
Interviewers can know what they are evaluating.
Hiring managers can know why a particular interview exists.
Recruiters can know what evidence remains missing.
Debriefs can focus on evidence rather than reconstructing everyone's feelings.
Structure is not merely an efficiency intervention.
It is an information-quality intervention.
The hidden cost is not interview time. It is organizational time.
Let's talk money carefully.
There is no credible universal dollar amount for a redundant interview.
Compensation varies.
Panel size varies.
Recruiter involvement varies.
Candidate volume varies.
The business consequence of delayed hiring varies enormously by role.
So we should not pretend that "every unnecessary interview costs $X."
But we can demonstrate the mechanics.
Consider a modeled scenario, not a benchmark.
An organization hires 500 people per year.
For every hire, five candidates reach a panel stage.
The organization requires one additional 60-minute interview that does not materially change the decision.
Three employees participate.
That creates:
500 hires
× 5 panel-stage candidates
× 3 interviewers
× 1 hour
= 7,500 employee hours per year.
At a hypothetical loaded labor cost of $75 per hour, that equals:
7,500 × $75 = $562,500 in employee time.
Again:
That is not a claim that companies lose $562,500 because of extra interviews.
It is an illustrative calculation showing how seemingly small design choices scale.
And it still excludes:
recruiter scheduling time,
candidate preparation,
rescheduling,
debrief time,
administrative coordination,
and any vacancy delay created by calendar availability.
Change any assumption and the number changes.
The responsible financial question is therefore not:
"What does an interview cost?"
It is:
"How much organizational capacity are we consuming to obtain this evidence, and does the evidence justify the cost?"
That is measurable.
Hiring teams rarely measure information gained per interview
Most recruiting dashboards can tell you:
time to fill,
time in stage,
offer acceptance,
source of hire,
candidate volume,
interview-to-offer ratio.
Far fewer can answer:
Which interview stages actually change decisions?
That is a fascinating operational question.
Suppose 96% of candidates who pass Stage 2 also pass Stage 3.
That does not automatically prove Stage 3 is redundant.
Perhaps Stage 3 assesses a rare but critical risk.
Perhaps the candidates reaching it have already been heavily selected.
Perhaps the interview influences leveling rather than pass/fail.
But it is worth investigating.
Similarly:
Which competencies are assessed in multiple stages?
Which interviewers frequently disagree?
Which stages generate "no new evidence" during debriefs?
Which interview questions actually differentiate candidates?
Which interviews are added informally for senior stakeholders?
Which stages produce candidate withdrawals?
Which interviewers routinely submit evidence late?
Which assessments are ignored once the hiring manager forms an opinion?
Those questions move TA Operations beyond funnel administration.
They turn interview design into an operating discipline.
The debrief may be where good interview design goes to die
Even a carefully structured interview can be undermined after it ends.
Imagine five interviewers independently assess a candidate.
Then they enter a room.
The hiring manager speaks first:
"I loved her."
Now everyone else interprets their evidence through that statement.
Or:
"I don't think he's senior enough."
Someone who rated the candidate highly begins reconsidering.
A person with weak evidence becomes more confident because another executive agrees.
The purpose of independent assessment is partly to preserve information before social influence begins.
That does not mean hiring teams should never discuss candidates.
They should.
Ambiguous evidence needs interpretation.
Conflicting observations need resolution.
Context matters.
But there is a major difference between:
independent evidence → controlled deliberation
and
group conversation → collective vibe.
PathPair's public hiring architecture reflects this distinction: interview execution should produce independent evidence before deliberation, and the subsequent decision stage should operate through defined decision authority rather than assumed ownership.
That is not about removing humans from hiring.
It is about protecting the value of what humans observed.
Hiring managers are not the problem
This is important.
When interview processes break down, recruiting teams sometimes default to:
"Hiring managers need more training."
Maybe.
But training cannot repair architecture that does not exist.
If a hiring manager receives:
no defined competencies,
no interview purpose,
no calibrated scoring standard,
no candidate context,
no preparation expectations,
and no clear decision authority,
then tells the recruiter afterward:
"I just didn't love them,"
the organization created the conditions for that outcome.
The person may need coaching.
The system definitely needs design.
PathPair's own public operating principles treat interview quality as an organizational capability: interview stages should have defined purposes, criteria, evidence expectations and accountable ownership, while additional interviewers should not automatically mean additional interviews.
That distinction matters.
Because blaming hiring managers produces workshops.
Designing the system produces repeatability.
The interviewer's résumé is not an interview qualification
Organizations often give interview authority according to hierarchy.
You're a director?
You can interview.
VP?
Definitely.
You've worked here six years?
Put you on the panel.
None of those facts demonstrates that someone knows how to conduct a reliable selection interview.
Interviewing is work.
It requires understanding:
what the role requires,
what the interviewer is responsible for assessing,
how questions should be asked,
what evidence matters,
what information should not influence the decision,
how candidate responses should be evaluated,
and what should be documented.
Yet organizations routinely give people significant influence over employment decisions with less preparation than they require to approve a purchase order.
That is an odd risk posture.
The "culture fit" interview deserves special suspicion
Few interview phrases conceal more ambiguity.
"I want to assess culture fit."
What does that mean?
Collaboration?
Integrity?
Risk tolerance?
Communication?
Customer orientation?
Adaptability?
Work style?
Values?
Or:
"Would I enjoy working with this person?"
Those are not interchangeable.
A vague culture-fit interview can easily become a permission structure for similarity preference.
If a characteristic genuinely matters to successful performance or organizational requirements, define it.
Then determine what evidence would demonstrate it.
If it cannot be defined well enough for interviewers to evaluate consistently, ask whether it belongs in the selection decision at all.
"Culture fit" should never become a professionally acceptable synonym for vibe.
There is a simple test for every interview on your calendar
Take one open role.
Write every interview stage across the top of a page.
Under each stage, answer:
What specifically are we trying to learn?
Why don't we already know it?
What evidence will demonstrate it?
Who is qualified to evaluate that evidence?
How will the evidence affect the decision?
Then look across the page.
You may discover:
three people assessing communication,
two people assessing leadership,
nobody assessing an essential technical requirement,
one executive interview with no defined purpose,
and a work sample that everyone requested but nobody actually uses during the decision.
That is not an interviewer problem.
That is architecture.
And architecture can be redesigned.
What TA leaders should measure next
Don't immediately launch a company-wide interview overhaul.
Start with evidence.
Take a representative sample of recent roles.
Map the interview stages.
Map the competencies or requirements each stage was supposed to evaluate.
Compare that with what interviewers actually documented.
Then examine:
Redundancy — Where is the same requirement repeatedly assessed without a clear reason?
Coverage — Which important requirements receive little or no evidence?
Decision influence — Which stages materially contribute to decisions?
Candidate burden — How much candidate time does the process require?
Organizational burden — How much interviewer and recruiter capacity does it consume?
Evidence quality — Are interviewers documenting observations against defined criteria or merely recording conclusions?
Decision latency — How long after sufficient evidence exists does the organization actually decide?
Do not begin with a predetermined target such as:
"Every role must have three interviews."
Begin with:
"Every material selection step should have a defensible purpose."
The right number can follow from there.
More interviews can actually be a symptom of low confidence
This is perhaps the most interesting possibility.
Organizations often add interviews because they want greater confidence.
But additional interviews may sometimes reveal that the organization does not trust its own selection system.
The hiring manager wants another opinion.
The executive wants to meet everyone.
A peer is unsure.
Someone asks for "one more conversation."
Nobody wants responsibility for the wrong hire.
So another meeting appears.
Then another.
This can create the illusion of rigor.
But consensus is not the same thing as evidence.
Six people agreeing on a weakly defined criterion does not make the criterion stronger.
And if an organization routinely needs another interview before it feels comfortable deciding, the problem may not be insufficient candidate exposure.
It may be insufficient decision architecture.
The future of interviewing should not be fewer interviews
Nor should it be more interviews.
It should be intentional interviews.
That sounds obvious.
The evidence suggests organizations still struggle with it.
We now operate in a hiring market where employers report difficulty filling roles while recruiters face more applicants, candidates experience multiple interviews, technology generates unprecedented amounts of hiring data, and artificial intelligence can summarize nearly every conversation.
Information is not the scarce resource anymore.
Decision-relevant information is.
That changes the job of Talent Acquisition.
The goal is no longer simply to move candidates through interviews efficiently.
It is to ensure the hiring system asks the right questions, of the right people, at the right time, for a defined reason — and then converts the resulting evidence into a responsible decision.
A candidate should not meet another employee because:
"That's how we've always done it."
An executive should not enter the process because:
"They like to meet finalists."
A panel should not exist because:
"We want multiple perspectives."
Those may eventually be defensible reasons.
But they are not yet designs.
Every additional interaction asks something from a candidate.
Their time.
Their preparation.
Their attention.
Their hope.
Organizations should be able to explain what they are receiving in return.
Because an interview is not valuable merely because it happened.
Its value is the uncertainty it responsibly removes from the decision.
And if the organization cannot identify that value, perhaps the interview does not belong there.
If this made you look differently at your interview process
You do not need to start by rebuilding every scorecard.
Start smaller.
Choose one role your organization hires frequently.
Map every interview.
Ask what each stage is supposed to establish.
Then compare that intended design with what candidates and interviewers actually experience.
If those two versions of the process look different, you have learned something important.
PathPair's complimentary Bridge Assessment exists to help organizations determine whether hiring-experience conditions like these warrant deeper investigation — without assuming in advance that a full Diagnostic is necessary.
Because the objective is not to create more process.
It is to make sure the process you already have is doing useful work.
SOURCES & FURTHER READING
U.S. Office of Personnel Management — Structured Interviews.
Primary federal guidance describing structured interviewing, job-related competency design, standardized questions, scoring and research-supported benefits.
OPM Structured Interviews
U.S. Office of Personnel Management — Guidance on Use of Hiring Assessments — February/March 2026.
Current federal assessment guidance addressing structured interviews, work samples, technical assessments and the appropriate use of multiple interview rounds.
OPM Hiring Assessment Guidance
Bourdage, J.S. & Powell, D.M. — “Employment Interviews: Past, Present, and Future,” Oxford Handbook of Personnel Assessment and Selection — July 22, 2026.
A new scholarly overview of interview structure, validity, reliability, fairness, applicant behavior and technology-mediated interviewing.
Oxford Handbook chapter
Wiesner, W.H. & Cronshaw, S.F. — “A Meta-Analytic Investigation of the Impact of Interview Format and Degree of Structure on the Validity of the Employment Interview” — 1988.
Foundational meta-analysis covering 150 usable validity coefficients.
Wiesner & Cronshaw meta-analysis
Schmidt, F.L. & Zimmerman, R.D. — “A Counterintuitive Hypothesis About Employment Interview Validity and Some Supporting Evidence” — Journal of Applied Psychology, 2004.
Useful counterweight showing why the structured/unstructured distinction should not be oversimplified.
PubMed record
Levashina, J., Morgeson, F.P., Campion, M.A. & colleagues — “The Structured Employment Interview: Narrative and Quantitative Review of the Research Literature” — Personnel Psychology, 2014.
Structured Interview Review
McCarthy, J.M., Bauer, T.N., Truxillo, D.M. et al. — “Applicant Perspectives During Selection” — Journal of Management, 2017.
Comprehensive review of 145 studies examining applicant reactions, fairness, organizational attractiveness, pursuit intentions and related outcomes.
Applicant Perspectives review
Express Employment Professionals / The Harris Poll — “44% of Employers Report Unfilled Roles as Hiring Process Takes Longer” — August 12, 2026.
Recent U.S. survey of 1,006 hiring decision-makers and separate survey of 1,000 job seekers.
Read the survey findings and methodology
LinkedIn — 2026 Talent Research — January 2026.
Current data on applicant volume, recruiter difficulty finding qualified talent and recruiting technology adoption.
LinkedIn 2026 Talent Research
