Is diversity hiring a quota?
And what inclusive hiring means.

The difference between widening a pool and reserving a place

Most of the argument about this subject comes from the two being treated as the same thing. They intervene at different points in a process and they produce different results.

A firm that maps a market thoroughly and approaches everyone qualified is doing inclusive hiring whatever it calls the practice. A firm that works its existing network and presents whoever it already knows is doing the opposite, also unannounced.

What the law generally permits

Rules vary by jurisdiction and change, so this is the shape rather than advice on any specific market. Take local legal advice before setting policy.

• Widening the pool is generally permitted. Advertising more broadly, sourcing from a wider set of organisations, and removing requirements that screen out capable people are ordinary good practice.
• Structured, consistent assessment is generally permitted and encouraged. Applying the same criteria to everyone is the clearest defence against a claim from any direction.
• Selecting on a protected characteristic is generally restricted. Some jurisdictions permit it narrowly as a tie-break between equally qualified candidates, and many prohibit it outright.
• Targets and quotas differ in status. An aspiration to improve representation is usually lawful. A reserved allocation in hiring usually is beyond what the law allows.

The practical consequence is that the defensible version of this work happens before the shortlist rather than at the point of decision.

What actually changes a shortlist

Interventions at the decision point are contested and largely ineffective. Interventions earlier in the process are uncontroversial and do most of the work.

1. Rewrite the requirements. Long lists of desirable criteria screen out capable people who self-assess strictly. Requirements that reflect what the job genuinely needs widen the pool immediately.
2. Source beyond the obvious organisations. Most shortlists come from five companies. The sixth through twentieth contain people doing the same work with less visibility.
3. Fix the criteria before anyone is seen. Criteria agreed in advance keep the standard fixed instead of adjusted afterwards to fit whoever arrived, which is where bias does most of its work.
4. Structure the interview. Same questions, same order, scored independently before discussion. The single most evidenced improvement available, and the least glamorous.
5. Measure conversion by stage. Where a group enters the pool and disappears between first and second interview, the problem has a location and can be examined.

Why this behaves differently at senior level

Most published advice on inclusive hiring assumes an application process. Senior appointments have none, which changes which interventions matter.

• Nobody applies, so widening the advert achieves little. The pool is whoever the search firm chooses to approach, which makes sourcing the entire question.
• The pipeline argument becomes checkable. At junior level "there were few candidates" is hard to test. At senior level the qualified population is countable, and a firm can be asked how much of it was approached.
• Networks compound. A firm working its existing relationships reproduces the composition of those relationships year after year, by default rather than decision.
• Assessment is bespoke. Structured interviewing helps, and the larger risk is criteria quietly rewritten after the field arrives, to fit whoever looked most familiar.

The practical consequence is that at senior level this is a question about how thoroughly the market was mapped rather than a question about the interview. Asking a search firm what share of the qualified population it approached gets closer to the truth than any policy document.

The merit objection, taken seriously

The objection is that this lowers the bar. It deserves a direct answer rather than a dismissal, because the version of this work that does lower the bar genuinely exists.

Where an organisation sets a headcount target and leaves the process unchanged, pressure lands at the decision point and the standard bends. That is a genuine setback and it damages the people it was meant to help, who then carry a suspicion about how they were appointed.

Where an organisation widens the pool and tightens the assessment, the standard rises rather than falls, because a larger field assessed consistently produces a better appointment than a smaller field assessed loosely. The two approaches share a vocabulary and share very little else.

Marquee leads on merit for this reason, and it costs us work. Clients who want a shortlist shaped to a target sometimes go elsewhere, and we would rather that than run a process we would have to explain away afterwards.

Hiring people with disabilities

A distinct question that is frequently folded into the same conversation and behaves differently in practice.

• Most of what excludes disabled candidates is procedural rather than attitudinal: an inaccessible application system, an interview format that tests something unrelated to the job, a venue nobody checked.
• Adjustments are usually inexpensive and frequently improve the process for everyone. Sending questions in advance, for instance, produces better answers from all candidates.
• The requirement is generally a legal one rather than a discretionary one, which changes the register of the conversation entirely.

The useful question is whether the process tests the job, or tests the ability to get through the process.

Why the language around this is so contested

The vocabulary has become a fight of its own, and it is worth understanding why before choosing which words to use internally.

The phrase "diversity hire" entered common use as a description of a practice and is now used at least as often as an insult aimed at an individual. That shift does real damage: it attaches a question mark to people who were appointed on merit, and it makes the subject harder to discuss honestly inside an organisation.

• Describe the practice, rather than the person. "We widened the search" is a statement about a process. "A diversity hire" is a label applied to a colleague, and it will be heard that way.
• Talk about the pool, publicly. Reporting how many people were approached, and from where, moves the conversation to something checkable.
• Announce appointments as appointments, rather than as achievements of representation. It is well meant and it reliably undermines the person it is meant to honour.

The organisations that handle this best tend to say least about it and do most, which is the reverse of how it is usually approached.

Measuring it honestly

Headcount representation is the most quoted measure and the least useful for managing, because it moves slowly and says nothing about cause.

• Pool composition. Who was approached, before any filtering. This is the number a search firm controls and should report.
• Stage conversion. Where groups drop out. A gap between application and first interview points at screening; a gap at final stage points at the panel.
• Offer acceptance. Where offers are declined disproportionately, the issue is the organisation rather than the pipeline.
• Retention at two years. The measure that reveals whether people were hired into somewhere that works for them, and the one most often left unexamined.

Marquee reports pool composition on mandates where a client asks for it, because it is the part of the process we are accountable for. The stages after the shortlist belong to the client, and measuring them is where most of the useful information sits.