Culture Add

What culture add is

Culture add is a selection criterion that asks what a candidate would bring to what the team already has, instead of asking how closely the candidate resembles it. It reframes the question of culture fit rather than overturning it.

The difference between the two lies in the reference point the candidate is measured against. Under fit, the reference is the team as it stands today, and the smaller the distance, the higher the score. Under add, the reference is what the team lacks among the things its work requires, and the more of that gap a candidate fills, the higher the score.

One boundary is worth setting before anything else. The subject here is the quality of a selection criterion and whether it can be reviewed. The legal treatment of discrimination in employment is a separate question with its own sources, and nothing below settles it.

What “fit” actually measures

The word fit is used in three different senses, and the interview record collapses all three into that one word:

• An unwritten requirement of the job. The role may demand working without close supervision, for example, so a candidate who cannot do that is turned down and the rejection is recorded as poor fit. The requirement is legitimate. Its only flaw is that nobody wrote it down.
• A value the organisation can name. Keeping to deadlines, for instance, or being open to challenge. These can be described in observable terms, and they belong among the competencies the role is assessed on.
• Familiarity. The candidate is close to the interviewer in ways that have nothing to do with the work: a similar manner of speaking, a shared experience, a common background.

The first two senses can be written down; the third cannot. That is the distinction the whole question turns on. A criterion that is never written cannot be reviewed, the person applying it cannot be asked about it, and its effect on a decision cannot be traced, whether it helped a candidate or counted against them. It is also one of the places where implicit bias operates, since bias does not have to be intended in order to take effect.

The working test: can it be written down?

A single question sorts the three senses. Can the criterion be written as behaviour that is observable at work, with its levels described in words that tell one level from the next?

If it can, it is a competency. It goes into the structured interview with its own question and its own rating guide, and it needs no other name. If it cannot, it is not a criterion at all but an impression, and an impression is not improved by adding a new word to the evaluation form.

A form that lists culture add as a criterion, without defining what the candidate adds and which gap it fills, keeps exactly the same impression under a newer label.

Why the pull towards similarity needs no intent

An interview is a conversation with limited time, and that alone is enough to produce the pull. A candidate who speaks the way the interviewer speaks is understood faster and needs less explaining, and the session runs more smoothly. That ease is then written up on the record in other words: clear, quick on their feet, gets on well.

It follows that the candidate an interviewer finds hard to read is not necessarily the weaker candidate. They may simply need more time to be read, and time is exactly what an unstructured interview does not have. For that reason the most effective way to reduce the effect is a change to the instrument rather than a lecture to the interviewer: the same questions put to every candidate, a rating written after each question rather than at the end of the session, and two interviewers who each rate before they compare notes.

The result is that an overall impression has nowhere to be recorded, so the interviewer has to tie each judgement back to what was said in answer to a question. This does not remove the pull. It moves it off the form and into a place where it can be seen and discussed.

One reasonable decision, and a total that is not

Take a team of ten positions in which two are refilled each year, so that all ten turn over within five years. Take a trait that has nothing to do with the work, on which the candidates in the market are split evenly in half.

If each hiring decision were neutral towards that trait, it would pick a candidate who has it half the time, and the team would settle at around five in ten. If each decision leaned towards resembling the current team and picked the trait 80% of the time, the expected figure would be eight in ten.

The sharper difference shows in the probability that the entire team ends up sharing the trait:

• Under neutrality: 0.5 raised to the power of ten, or about 0.098%, which is close to never.
• Under a lean towards similarity: 0.8 raised to the power of ten, or about 10.7%.

The probability is roughly 110 times higher, yet no single decision looks unreasonable. Each leaned eight to two, a tilt nobody would notice in an individual case. What stands out is the total after five years, when nobody remembers any of the individual decisions.

The figures are assumptions chosen to show how a small, repeated lean compounds. They are not a benchmark and are not taken from any source. The trait is deliberately abstract: the example does not name it and does not stand for any particular characteristic.

The effect is stronger where the pool of candidates is itself narrow, as in hiring through employee referrals. The candidates come from people the current staff already know, so the pool has already narrowed before selection begins, and the criterion compounds that.

What changes in the interview question

A move to culture add changes the interview question in a way that shows on the record:

• Instead of “Will they get on with us?”, the question becomes: which gap in this team does this candidate fill?
• Instead of “Are they like us?”, it becomes: what would they see in us that we do not see ourselves, and do we need it?

Both questions depend on one condition: what the team already has must be written down. A manager who does not know what experience and what perspectives the team holds cannot say what it lacks. This is the practical side of the principle that organisational culture is read in everyday decisions rather than in declared values. The written reference here is a description of what actually happens in the team, not the list of values on the wall.

Where difference stops being an addition

Not every difference is an addition. Some areas of an organisation depend on uniform behaviour, not on variety: safety rules, the way decisions are documented, the obligation to report a mistake when it happens. An employee who differs here is not contributing a new perspective. They are failing to meet a requirement.

One question separates the two. Does this difference change how the work is done, in a way that is safe and can be verified, or does it only change who in the team feels comfortable with the person? The first is a requirement to be written and measured. The second is where the question of culture add properly applies.

The step most often missed in practice is that this separation is made before the interview, not during it. An interviewer who walks in without a written list of the areas where consistency is required ends up deciding that in the middle of the conversation, which is exactly the point at which the decision slides back into familiarity.

Where culture add is misread

Culture add changes one thing: the question the selection criterion asks, and whether that question can be reviewed. The useful check on any reading of it is whether that reading makes it change anything else, and every common misreading does.

Reading it as a numerical target fails that check at once. The criterion sets no proportion of any kind and obliges nobody to reach one. Reading it as a way of relaxing the technical standard fails it too. A candidate who cannot do the work is not hired, however much they would add, and addition is weighed only among the candidates who have already cleared the technical bar, never before them. Reading it as a replacement for structured assessment fails it in the quietest way. Culture add decides what is asked about; the question still has to be put the same way to everyone and rated on one scale. Without that, it becomes a second impression layered on top of the first.

Nor is the case for culture add a promise of better results. Nothing in the reasoning above establishes that changing the criterion raises any outcome by a known amount. The argument is about whether a criterion can be written, applied and reviewed.

What undermines it in practice

• Leaving the gap undefined. An addition with no specified gap means, in practice, preferring the different candidate because they are different. That is the original criterion turned upside down, not corrected.
• Changing only the word. A form on which fit has been deleted and add written in its place, with no question and no rating guide behind it, has not changed at all.
• Leaving it to one interviewer. What a team lacks is something people disagree about, and a decision resting on one person’s view returns to familiarity by another route.
• Ignoring what happens after the hire. Someone hired because they add something, and then placed where nobody listens to them, adds nothing. A year later the experience is read as proof that the criterion does not work, which is why employee onboarding is part of the same decision rather than a separate one.

The criterion is also measured at the level of the team, not the organisation. An organisation that is varied as a whole can still consist of teams that are each uniform inside, and the work happens in the team, not in the total. The overall figure is reassuring, and it says nothing about the place where the decision is actually made.

Before the criterion goes on the form

The most useful test before writing it in is to ask three of the people who interview candidates each to write, in two lines, what they mean by the criterion. If the three answers come back far apart, the criterion does not exist; only its name does, and decisions are being made on three criteria rather than one.

Then the gap in the team is written down before the vacancy is opened, not after the candidates have been seen. Defining the gap after the interviews turns it into a description of whoever impressed the interviewer, which is the purest form of a criterion drafted to justify a decision already taken rather than to take one.

Related terms

• Structured Interview
• Competencies
• Behavioral Interview Questions
• Assessment Centre
• Employee Onboarding