Six interviews are scheduled for Thursday. Two people arrive.
The explanation offered inside most businesses is about the candidates: nobody wants to work, people are unreliable, the workforce has changed. The explanation is emotionally satisfying, it is repeated widely enough to feel like evidence, and it predicts nothing. It does not explain why the same employer's no-show rate differs by role, or falls sharply when one thing about the process changes.
A no-show is not a character trait. It is the visible outcome of a decision the candidate made days earlier, usually for a reason the employer could have observed.
The candidate is running a portfolio
The single most useful correction is to stop modelling the candidate as choosing whether to attend your interview.
Someone who needs work within a fortnight does not apply to one job. They apply to eight or twelve, because the base rate of any single application producing work is low and they cannot afford to serialise. They are managing a portfolio under time pressure, and every application is a live option until something better resolves.
An interview slot four days out is one option among several. If a different employer makes an offer on day two, your Thursday slot is not declined, because declining requires an action and the candidate has no relationship with you that makes the action feel owed. It simply stops being relevant.
This reframing explains what the character theory cannot. It predicts that no-shows rise with the gap between application and interview, that they fall when an employer replies first, that they cluster in roles where alternatives are plentiful, and that they concentrate among the strongest candidates, who have the most options and resolve them soonest. All four are observable from an employer's own records.
Time is the dominant variable
Of everything an employer controls, elapsed time is the one that moves the number most.
- Hour 0Application submittedInterest is at its maximum. The role is specific in the candidate's mind, and they have just chosen to spend effort on it.
- Hours 1 to 24The window where a reply is decisiveA response here is frequently the first the candidate has received, and it converts an application into an actual relationship. Attendance is highest for interviews scheduled in this window.
- Days 2 to 4Other applications begin resolvingCompeting employers respond. Your role becomes one of several rather than the most recent thing considered.
- Days 5 to 7The application is effectively coldThe candidate may not recall the specific role. A message from an unfamiliar company name now reads as unsolicited rather than expected.
- Day 8 onwardScheduling reaches a strangerAny interview arranged here competes with commitments made since, and it is attended out of politeness rather than intent.
Two intervals matter and they are usually confused. The first is time to respond, being how long before the candidate hears anything at all. The second is time to interview, being how long until they are actually seen.
Time to respond is the one that determines attendance. A fast reply followed by an interview eight days out performs substantially better than silence followed by an interview in three, because the reply establishes that a real person is expecting them. The interview is attended for the relationship, not for the slot.
That is a useful thing to know, because it means the fix is cheap. Compressing scheduling is an operational constraint that may not be movable. Replying within a day is a decision about who is responsible for replying.
The channel is not a detail
The second largest controllable factor is where the message arrives.
A large share of the frontline workforce does not use email as a primary channel. An address exists because a form demanded one. It is checked rarely, filtered aggressively, and an interview invitation sent there is competing with promotional mail for attention it will not get.
A message that is not read is indistinguishable, in outcome, from a message never sent. Employers routinely record the resulting silence as candidate behavior, which makes the channel invisible in the diagnosis.
Process that produces no-shows
- First reply after four days
- Sent to an email address
- From a name the candidate does not recognize
- Written in English by default
- No confirmation before the day
- No way to reschedule except calling
Process that produces attendance
- First reply inside a day
- Sent where the candidate already replies
- Naming the role they applied to
- In the language they wrote in
- Confirmed the day before
- Rescheduling is one reply
The rescheduling point deserves more weight than it gets. A candidate whose circumstances change has two options: tell you, or disappear. If telling you requires a phone call during business hours to a company they have not yet met, disappearing is easier. Much of what is recorded as ghosting is a cancellation that had no cheap way to happen. Making rescheduling a single reply converts some portion of no-shows into a later interview, which is a strictly better outcome than an empty slot.
The causes, in order of size
- 1Days of silence after applyingThe dominant factor. Interest decays fast and the first employer to respond frequently converts the candidate before anyone else has read the application.
- 2Another offer resolved firstLargely unavoidable and partly addressable by speed. Its size is a direct function of how long the candidate had to wait.
- 3Contact sent to a channel the candidate does not readEmail in a workforce that answers messages. Produces silence that gets recorded as candidate behavior rather than as delivery failure.
- 4Facts that made the job unviable, discovered latePay, precise location, travel time, actual shift pattern. If these emerge after scheduling, the interview is already pointless and the candidate knows it before you do.
- 5No cheap way to cancel or rescheduleA candidate who cannot decline easily disappears instead. This portion is recoverable, and it is recorded as ghosting.
- 6Genuine unreliabilityReal, and much smaller than the folk explanation assumes. It is what remains after the five above are addressed, and it is the only one that is actually about the candidate.
The fourth item is worth separating out because it is entirely self-inflicted. A job post that omits pay, gives a city rather than an address, or describes hours as flexible generates applications from people who have not yet decided whether they can take the job. Some of them work it out before Thursday. Vagueness in a job post converts a screening problem into a no-show problem, and moves it to the most expensive point in the process, which is a scheduled slot with a manager in the room.
What to change first
In order of effect per unit of effort.
Reply within a day, even when the reply is holding rather than deciding. The message that a real person has seen the application does most of the work.
Send it where the candidate will read it. For this workforce that means messaging. It is the reason Kariaa runs candidate contact through WhatsApp and Telegram rather than email, detects the language of each incoming message on its own rather than treating language as a profile setting, and replies in the language the candidate wrote in.
Confirm the day before, and make rescheduling a single reply rather than a phone call.
Put the deciding facts in the post: the pay range, the precise location, the real shift pattern. Every candidate who screens themselves out on Monday is a slot that stays available for someone who will attend.
Then measure it. Record the date of application and the date of first response for every candidate, and compare attendance across that interval. The relationship is usually visible in a few dozen rows, and it will be a stronger argument for changing the process than any external benchmark, because it is about your own candidates. That is part of the four-date approach in Cost per Hire and Time to Fill Explained.
Key takeaways
- Candidates who need work apply to many roles at once. Your interview is one live option in a portfolio, and it stops being relevant when another resolves first.
- Time to respond drives attendance more than time to interview. A fast reply followed by a distant interview beats silence followed by a near one.
- A message sent to a channel the candidate does not read produces silence that gets recorded as candidate behavior.
- Much of what is called ghosting is a cancellation with no cheap route. Making rescheduling one reply recovers part of it.
- Vague job posts move a screening problem to the interview slot, which is the most expensive place it can surface.
- Genuine unreliability is real and much smaller than assumed. It is what remains after the controllable causes are addressed.
Common questions
Why do candidates not show up for scheduled interviews?
- Most often because time passed and another application resolved first. A candidate who needs work applies to many roles simultaneously, so any single interview is a live option rather than a commitment. Long silence after applying, contact sent to a channel they do not read, and job facts that made the role unviable discovered late account for most no-shows. Genuine unreliability is real but much smaller than commonly assumed.
How quickly should I respond to a job applicant?
- Within a day, even if the reply only acknowledges the application rather than deciding on it. Time to respond affects interview attendance more than time to interview does: a fast reply followed by an interview eight days out performs better than silence followed by an interview in three, because the reply establishes that a real person is expecting them.
Should I contact frontline candidates by email or text message?
- Messaging. A large share of this workforce does not use email as a primary channel; an address exists because a form required one. An unread invitation produces the same outcome as one never sent, and the resulting silence is usually recorded as candidate behavior rather than as a delivery failure.
How do I reduce interview no-show rates?
- Reply within a day, send it to a channel the candidate actually reads, confirm the day before, and make rescheduling a single reply rather than a phone call during business hours. Put pay, precise location and the real shift pattern in the job post so candidates who cannot take the job screen themselves out before a slot is booked.
Is candidate ghosting getting worse?
- The framing is worth questioning before the trend is. Much of what gets recorded as ghosting is a cancellation that had no cheap way to happen, or a message that was never read because it went to an unused channel. Both are process outcomes an employer controls, and both are measurable from an employer's own records by comparing attendance against the interval between application and first response.
Written by the team at Kariaa. Learn more at www.kariaa.com.