There is one mistake I have made more than once during my years in recruitment.
I liked the candidate.
They were pleasant, stable, professional and funny. The conversation was easy, and they seemed like someone it would genuinely be enjoyable to work with.
And that was exactly where the problem started.
When I like someone, I have noticed that I can start asking fewer questions. Not because I consciously decide to ignore a potential problem, but because some things begin to feel obvious before I have actually tested them.
I experienced this with a candidate who was very close to what the client needed culturally. I genuinely liked the person and could easily imagine them fitting into the team.
The problem was not cultural fit.
Their current maximum simply was not senior enough for what the client needed from the role.
My positive impression of the person was not wrong. What would have been wrong was allowing that impression to influence an assessment of something entirely different.
For me, this became one of the most important lessons in candidate assessment: subjectivity does not always make us judge someone more harshly. Sometimes it simply makes us stop asking.
Experiences like this changed the way I conduct interviews.
Subjectivity Can Change the Questions Before It Changes the Conclusion
The biggest problem for me was not that I noticed contradictions and deliberately chose to ignore them.
Often, I never reached the contradiction because I did not ask the question that might have revealed it.
When a conversation flows easily and the candidate appears professional, competent and pleasant, it becomes surprisingly easy to assume that the areas we have not explored yet will probably be fine too.
Agency recruitment adds another layer to this.
Roles are often urgent. The client is waiting. The market is limited. When a promising candidate appears, it is a desirable outcome for everyone involved if that person turns out to be the right one.
That is exactly when the interviewer needs to become more disciplined.
The desire for someone to be a good candidate must not determine how rigorously we assess whether they actually are.
For me, professional assessment starts with the awareness that I am also part of the assessment process, and that my own reaction to a candidate can influence what I ask, what I hear and what I conclude.
I Start by Testing Why the Candidate May Not Be Right for the Role
To reduce my own subjectivity, I start every assessment with an internal hypothesis: this candidate may not be good enough for this particular role.
That sounds harsher than it is.
I do not enter an interview trying to find a flaw in the candidate, and I am not trying to prove that they are inadequate.
The hypothesis exists for me, not for them.
If I begin with the assumption that someone is an excellent candidate, it becomes very easy to start collecting evidence that confirms what I already want to believe.
If I start from the possibility that the candidate may not be right for this particular role, I have to test the areas that matter regardless of how much I personally like them.
And if, after a thorough interview, I cannot defend that initial hypothesis, that is a very strong signal that I may genuinely have an excellent candidate in front of me.
Being right for the role is also broader than technical ability.
A candidate can be very strong but want their next hierarchical step, while the company is hiring for a purely engineering position and already has established Team Lead, Project Management or other leadership roles above it.
A candidate may have salary expectations that the company simply cannot meet. Even if their knowledge and experience justify those expectations, there is little value in taking them through several interviews when we already know the approved budget will not change.
This can also distort the rest of the hiring process. Once a company interviews an exceptional candidate at a compensation level it cannot afford, it can become very easy to expect the same level of knowledge and experience from candidates whose expectations are substantially lower.
Sometimes that candidate exists.
Often they do not.
The same applies to working models. If a candidate clearly wants a remote position and the company is committed to hybrid work for at least the next year, good recruitment should not depend on the hope that one side will eventually change its mind.
A sustainable match requires responsibility on both sides.
Does the company genuinely need, value and want what this candidate offers? Does the candidate actually want what this role and organisation require? And can both sides realistically meet the expectations that matter most to the other?
A good match is not a situation in which one side manages to impress the other.
It is a situation in which what both sides offer and what both sides need are sufficiently aligned for the relationship to work.
A Good Interview Answer Still Needs to Be Tested
One of the things that frustrates me most about standard interviews is seeing a fixed set of questions being used without enough thought about what those questions are actually supposed to tell us.
A question has very little value if we do not understand what we are trying to learn from the answer.
And even less if we accept the candidate’s first sentence as the conclusion.
Take a very common statement:
“I work well under pressure.”
It sounds positive, but at that point I still know almost nothing.
What does pressure mean to that person?
For one candidate, pressure may be a tight deadline. For another, it may mean a project that is already a month late, an unhappy client and a team trying to deliver while the situation continues to deteriorate.
I would therefore ask about the greatest pressure they have experienced in their career, what made the situation difficult, how they handled it and what they would do if the same situation happened again.
I may also want to understand how long they could sustainably operate under those conditions and what kind of pressure would eventually become a deal-breaker for them.
Only then does “I work well under pressure” start to have meaning.
The same applies when someone says:
“I am very responsible.”
I am more interested in the most significant responsibility they have actually held.
What was the project or role? What exactly did they own? Why do they consider it particularly responsible? What was happening around them at the time, and what did they do when the situation became difficult?
If they worked on a complex product or project, I also want to understand how they acquired the domain knowledge they were missing.
Did they wait for someone to provide all the necessary information, or did they actively seek out the right people, documentation, additional conversations and context in order to understand the domain well enough to take responsibility for the outcome?
I can then explore the same underlying process through their relationship with a manager, the way they communicate emerging problems, how they involve other people, how they approach a difficult client or what they do when they do not immediately have the information they need.
The questions can be completely different in content while testing the same process.
That is much more useful to me than asking once whether someone is responsible and turning their answer into an assessment.
Gestalt-Informed Candidate Assessment Follows Process, Not Just Answers
The strongest influence of Gestalt in the way I interview candidates is not a specific list of questions.
It is the way I follow process.
By process, I do not simply mean the chronological sequence of events in someone’s career. I am interested in recurring ways a person approaches situations, relationships, responsibility, missing information and professional development.
For example, the fact that someone progressed from Developer to Project Manager is not particularly informative on its own.
What interests me is how that progression happened.
Perhaps the person repeatedly recognised what they did not yet know, actively sought new knowledge, accepted broader responsibility, earned relevant certifications and at some point realised that understanding people had become important to the role, so they deliberately pursued further education in that area.
The career path itself is not the conclusion.
What becomes interesting is the recurring way that person responds when they recognise a gap between what they currently know and what the situation requires.
I cannot assume that they will behave identically in every future situation.
But it gives me a meaningful line of inquiry.
Do they show the same initiative when they lack information about a project? When they do not understand enough about a product or domain? When they need to involve another person? When a problem appears that nobody has already solved for them?
The same logic applies to responsibility for difficult outcomes.
One dismissal is not a pattern. It may happen for many reasons that have very little to do with an individual candidate’s performance.
Several dismissals across consecutive roles, however, require much more exploration.
One unsuccessful project does not make someone irresponsible either.
But if, across multiple situations, every difficult outcome is explained entirely through what someone else failed to do, a manager who did not provide enough information, a team that did not deliver or another person who was responsible, I want to understand where the candidate sees their own initiative and responsibility inside those situations.
I do not conclude from the first answer.
I explore further and, where relevant, return to the same underlying process later through a different question.
The same applies to non-verbal changes during an interview.
A change in tone, swallowing, coughing, a sudden shift in facial expression or a noticeable change in energy is never a diagnosis and it is not proof that a candidate is hiding something.
It is simply a place where I may want to stay a little longer.
If a candidate says that a previous employment relationship ended mutually, but the entire way they communicate changes when we reach that subject, I do not automatically conclude that the termination was not mutual.
I ask another question.
And if the answer still feels incomplete, I may approach the same subject from another angle later in the interview.
A non-verbal signal is not a conclusion. It is a reason to explore further.
Following process also means paying attention to what happens in me as the interviewer.
Do I like this person unusually quickly? Do I feel irritated? Am I starting to lose energy? Do I feel the need to defend the candidate or find explanations for them?
In psychotherapy, part of this would be discussed through countertransference. In recruitment, I do not treat my reaction as a fact about the candidate.
I treat it as information that may tell me where I need to become more curious.
If I notice, for example, that I am carrying most of the conversation and constantly creating the next step, I may explore how much initiative the person naturally takes, how they function without strong direction and what type of management structure allows them to perform best.
That does not make the candidate good or bad.
A person who performs very well inside a clearly structured team with an engaged Team Lead may be an excellent hire for one company and a poor match for a role that requires them to define direction independently from day one.
The objective is not to identify the “right personality”.
It is to understand the relationship between the person and the system they would be entering.
A Sustainable Professional Match Requires Alignment on Both Sides
The strongest candidates I meet do not necessarily have a perfect answer to every question.
More often, they have a realistic understanding of themselves.
They know what they can already deliver. They know where their knowledge currently ends. They understand what they still need to learn, and they are able to explain what they need from a company in order to perform well.
This becomes particularly important with senior candidates.
If the product is genuinely complex and the candidate’s reaction is essentially “this will be easy”, I do not automatically interpret that as confidence.
Sometimes it is a signal that the person has not yet understood the complexity.
Someone who has worked with comparable complexity can usually explain much more precisely what they have already handled, where the similarities are, what would be new for them and what they would need to understand before taking full responsibility.
Professional expectations matter in the same way.
A candidate may urgently need a job, but if their clear professional direction is towards modern technologies, continuous development and increasing complexity, a company whose foreseeable work is centred around a legacy system may not provide what they need long term.
They may still accept the offer.
If that mismatch was already visible during assessment, the problem has not disappeared simply because the candidate signed the contract.
It may have only been moved a few months into the future.
That is why the purpose of this kind of assessment is not to claim that we can predict people.
We cannot.
And a recruitment interview is not psychotherapy.
The purpose is much more practical: to recognise as early as possible where the candidate, the role and the organisation are genuinely aligned, and where important differences are likely to reappear later through withdrawal from the process, a rejected offer or early attrition.
There is also a clear boundary around what is shared with the client.
If a candidate tells me about a difficult personal or family situation that affected a previous role, the private details of that situation are not information for the client simply because they emerged during the interview.
What the client needs is the professionally relevant assessment: what the candidate can realistically deliver, what they expect, what environment they are likely to need, where the alignment appears strong and what may still be worth exploring with the hiring manager.
For me, the quality of candidate assessment is not measured by how impressive someone sounds at the end of an interview.
It is measured by whether both the candidate and the company leave the process with a sufficiently realistic understanding of what they can expect from each other.
And sometimes the first step towards a better assessment is not asking the candidate a better question.
It is noticing what happens to our own questions when we like their answers too much.