The instinctive use of a personality assessment is as a hiring filter: show us the "top talent profile" and screen candidates against it. This is a bad use of the instrument. A better use — cheaper, higher-impact, and dramatically less legally exposed — is to run the assessment once the offer has been accepted, and use it to make the first ninety days faster.
Why hiring is the low-value use case
Three reasons:
- Predictive validity is modest. The best-validated personality instruments predict on-the-job performance with a correlation coefficient of around 0.2-0.3. That is meaningfully above zero, but it is well below what the same instruments achieve in other use cases, and well below what structured interviews and work samples achieve for hiring.
- Fakeability is real. When candidates know their answers will be used to screen them, they shade their answers toward what they think the ideal profile looks like. Forced-choice instruments mitigate this; they do not eliminate it.
- Legal risk is real. Using personality assessments as a screening tool exposes you to adverse-impact claims in most jurisdictions. Even where it is technically permissible, the audit trail has to be impeccable — which is expensive in time if not in money.
Why onboarding is the high-value use case
The same instrument, used the day after the offer is accepted, has none of the above problems and gains three new upsides:
- Honest answers. The new hire has no incentive to shade their profile — they already have the job. The report is more accurate as a result.
- High leverage on time-to-productivity. The first 90 days of a new role is when working-style collisions with the new team do the most damage. A report the hiring manager reads before the new hire arrives compresses the "how do we work together" learning curve from months to weeks.
- Development-friendly framing. The report is an input to the development plan, not a reason the person was or was not hired. It is much easier for the new hire to engage with honestly.
The shape of a good onboarding use
For a senior hire — head-of-function level — we recommend the following sequence:
- Offer plus 1 week. Invite the new hire to take the assessment. Explain clearly what it will and will not be used for: development, not performance management; shared with the manager and the new hire, no-one else.
- Offer plus 2 weeks. New hire completes the assessment and reviews their own report. Optional debrief call with HR.
- Before day 1. Hiring manager reads the report. HR facilitates a thirty-minute conversation between the new hire and the manager specifically on working-style fit — not on the job, on how to collaborate.
- Day 30. Light-touch review: what has been true, what has not, what to adjust.
- Day 90. Formal review including the assessment. Set development goals that reference the profile.
The budget conversation
This change often requires a budget line to move. The spend used to sit in recruitment; it now sits in onboarding or L&D. That is a surprisingly political move because recruitment budgets are closely watched and L&D budgets are the first to be cut in a squeeze. Two arguments usually carry the day:
- Retention economics. A senior hire who leaves in year one costs between 1x and 2x their salary to replace. A £129 assessment in week one of their tenure is a rounding error on that risk.
- Compliance. Moving assessment spend out of the hiring funnel reduces legal exposure. In heavily regulated sectors — financial services, public sector, healthcare — this alone is often enough.
When is it OK to use psychometrics in hiring at all
There is a narrow, defensible use case for assessment in hiring: as a late-stage input for a final candidate list of 2-3 people, used to inform the conversation with each finalist rather than to rank them, with a structured interview built around the profile for each person. It is expensive, slow, and only worth doing for roles where the cost of mis-hire is very high — C-suite, regulated leadership roles, board seats.
For everything else, spend the money after they accept the offer, not before.