Why Good Candidates Walk Away
And why companies pretend that’s fine.
Most companies say they want great candidates. Then they build a hiring process that seems carefully designed to make great candidates question their life choices.
The job description is vague. The salary range is missing, or so wide it might as well say “money exists”. The recruiter screen is fine, the hiring manager conversation is promising, then suddenly the candidate is being asked to complete a take-home assignment, meet six people, repeat the same story three times, and wait in silence while everyone “aligns”.
Good candidates walk away from this crap. Not because they are entitled. Not because “nobody wants to work anymore”. They walk away because they have options, self-respect, and a functioning bullshit detector.
Hiring is not just selection, it is signalling
Every hiring process tells candidates what the company values.
If the process is clear, respectful, and well-run, candidates learn something useful. They see a company that can make decisions, communicate expectations, and treat people like adults.
If the process is chaotic, secretive, and weirdly performative, they learn something else. They learn the company might not know what it wants. They learn feedback loops are probably slow. They learn decisions may need three pre-meetings, two gut checks, and a ritual sacrifice to the stakeholder gods.
That matters because candidates are not only trying to get picked. They are also deciding whether they want to pick you.
Companies forget this constantly. They act like interviewing is a one-way evaluation, as if candidates should just be grateful to be considered. That mindset might work when the market is brutal and people are scared, but it still leaves a smell. People remember how they were treated when they had less leverage.
The job description is where the nonsense starts
A bad hiring process often starts before anyone speaks to a human.
The job description says the company wants a “rockstar”, which is already a tiny red flag wearing sunglasses. It asks for ten years of experience in a tool that became popular five years ago. It lists every technology the team has ever touched, including the one legacy service nobody wants to admit still exists. It says the role is “fast-paced”, “dynamic”, and “high ownership”, which can mean anything from “you will have autonomy” to “we have no plan and you are the plan”.
Then there is the compensation problem. Some companies still refuse to post salary ranges because they want “flexibility”. Very mysterious. Very mature. Definitely not a tactic to preserve information asymmetry and waste everyone’s time.
If the range is not posted, candidates have to guess whether the role is worth applying for. If the range is posted but absurdly wide, they still have to guess. A range like $80,000 to $220,000 is not transparency. It is a shrug in a trench coat.
Candidates notice. Good ones especially.
The interview loop should test the job, not endurance
A good interview process should answer a simple question: can this person do the work, and can we work well together?
Somehow, companies turn that into a gauntlet.
There is a recruiter screen. Then a hiring manager screen. Then a technical interview. Then a systems design interview. Then a values interview. Then a cross-functional interview. Then a panel. Then a final conversation that is absolutely not a final conversation because someone important was on vacation and now wants “just thirty minutes”.
At some point, you are no longer evaluating the candidate. You are testing whether they can survive your organisation’s inability to make a decision.
That is not rigour. It is drag.
Rigour means each step has a purpose. It means each interviewer knows what they are evaluating. It means the team does not ask the same generic questions five times because nobody bothered to compare notes. It means the candidate can see how the process connects to the actual role.
Drag is what happens when everyone wants input, nobody owns the decision, and “culture fit” becomes the junk drawer for whatever feelings people cannot explain.
Take-home assignments need to calm down
There are reasonable take-home exercises. A short, scoped task can help candidates show how they think without forcing them to perform under a fake whiteboard timer.
Then there are the other ones.
The “small assignment” that takes eight hours. The unpaid product strategy document. The full redesign. The data analysis project with suspiciously real business context. The code challenge that asks for production-level polish while pretending it is just a quick exercise.
No. Absolutely not.
If the assignment takes more than a couple of hours, pay people or cut it down. If it resembles real company work, pay people. If you cannot evaluate the candidate without asking them to donate a weekend, your process is not thoughtful. It is lazy with a calendar invite.
And before someone says “but we need to see how they work”, yes, you do. You also need to respect that candidates have jobs, families, disabilities, caregiving responsibilities, burnout, and lives. Shocking stuff, I know.
Whiteboarding is not automatically bad, but it is often used badly
Whiteboarding gets defended as a way to see problem-solving in real time. In theory, fine. In practice, it often becomes an anxiety simulator with markers.
The issue is not always the format. The issue is the mismatch.
If the job requires designing systems, talk through systems. If the job requires debugging, debug something realistic. If the job requires collaboration, create a collaborative exercise. But if the role is building accessible product experiences and the interview is reversing a linked list under pressure while three people stare silently, what exactly are we doing here?
Some interviews measure confidence more than competence. Some measure familiarity with interview games more than ability to do the job. Some measure whether the candidate had enough spare time to grind algorithm problems after work, which is not the same as engineering judgement.
That does not mean lower the bar. It means point the bar at the actual work.
“Culture fit” is where bias puts on business casual
Few phrases deserve suspicion faster than “culture fit”.
Sometimes people mean reasonable things by it, like communication style, collaboration habits, or whether the candidate can work in the environment the company actually has. Fair.
But often, “culture fit” becomes a polite little container for bias. It can mean “they did not remind me of myself”. It can mean “they were too direct”. It can mean “they did not perform enthusiasm in the exact dialect we prefer”. It can mean “I have a feeling, and I would like that feeling to count as evidence”.
That is how companies reject strong candidates while telling themselves they are protecting the team.
Use better language. Say what you mean. Evaluate behaviours. Define what good looks like before the interview. If you cannot explain why someone failed a category without sounding like you are reading tea leaves, the category is probably garbage.
Silence is not neutral
One of the fastest ways to make good candidates walk away is to disappear.
Candidates are told the company wants to move quickly. Then they hear nothing for a week. Then two weeks. Then a cheerful note arrives saying the team is “still calibrating”. On what? The moon phase?
Silence tells candidates something. It says the company either cannot make decisions, does not value communication, or thinks candidate anxiety is free. It is not free. It costs trust.
Even bad news is better than ambiguity. A simple update goes a long way: “We are still interviewing this week and expect a decision by Friday.” That takes thirty seconds to write. If your hiring process cannot manage that, candidates are right to wonder what the internal communication looks like.
Feedback does not have to be a legal dissertation
Companies often avoid feedback because they are worried about risk. Fine, nobody is asking for a twelve-page personality diagnosis.
But there is a middle ground between “here is a detailed psychological profile” and “we went with another candidate”.
You can say:
“We needed stronger experience with X.”
“The role requires more depth in Y than we saw in the interview.”
“We were looking for examples closer to Z.”
“The exercise did not show enough detail in A and B.”
That kind of feedback is useful, respectful, and not especially hard. It also forces the company to know why it made the decision, which is apparently a bold innovation in some hiring processes.
Good candidates leave when the process stops making sense
Good candidates are not only assessing compensation, title, and tech stack. They are assessing risk.
They ask themselves:
Does this company know what it wants?
Do these people communicate clearly?
Are decisions made thoughtfully or endlessly deferred?
Is the work real, or is this role a dumping ground?
Do they respect my time before they even employ me?
If the answers start looking bad, good candidates leave.
They withdraw after the third rescheduled interview. They decline the take-home assignment. They accept another offer from a company that moved with clarity. They stop responding because the company already taught them what working there might feel like.
Then the company says, “We are struggling to find talent.”
No, you found talent. You just annoyed it into leaving.
What a better hiring process looks like
This is not complicated. Hard, maybe. Complicated, no.
Start with the actual job. Define what the person will do in the first six months. Write the job description around that, not around a fantasy candidate assembled from everyone’s wish list.
Post the salary range. A real one. Not a range so wide it needs its own weather system.
Limit the interview loop. Every step should have a purpose, an owner, and a decision attached to it. If an interview does not change the decision, delete it.
Use realistic assessments. Test the work people will actually do. Keep exercises short, scoped, and respectful. Pay candidates when you ask for substantial work.
Train interviewers. Do not just throw people into interviews because they had a free calendar slot. Make sure they know what they are evaluating, what good evidence looks like, and how to avoid turning personal preference into hiring criteria.
Communicate like adults. Give timelines. Send updates. Close the loop. Do not make candidates chase you for basic information.
Debrief with evidence. Not feelings. Not “I liked them”. Not “I did not get senior energy”. Evidence. Examples. Role-related signals.
The bar is not the problem
Companies love to say they have a high bar. Fine. Have one.
But a high bar is not the same as a bloated process. A high bar does not require confusion, silence, unpaid labour, or six people asking the same question with slightly different facial expressions.
A high bar should make the process sharper. It should make the evaluation clearer. It should help strong candidates understand why the company is worth their time.
If your process filters out good people because it is exhausting, vague, biased, or slow, that is not a high bar. That is a broken door.
The uncomfortable part
Hiring processes reveal companies more than companies want to admit.
They reveal whether leaders can make decisions. They reveal whether the organization respects time. They reveal whether teams can define success. They reveal whether the company understands the work well enough to assess it. They reveal whether “people first” means anything once there is no marketing page involved.
Candidates see all of that.
So when good candidates walk away, maybe the question should not be “what is wrong with candidates these days?”
Maybe the question should be: “What did our process tell them about us?”
Because they listened.