Hiring and onboarding

Hiring is deciding who joins, against a bar defined before anyone is interviewed, through a process that measures capability rather than impression. Onboarding is the work that turns a good hire into a contributing one, and its absence is the usual reason a good hire fails. The competency is structured, calibrated evaluation rather than gut feel, and it is among the highest-leverage decisions a manager makes: a weak engineering hire costs a team a year, a weak manager appointment costs every person on that team, and a weak executive hire can cost an organisation years it does not get back. The leverage compounds with scope, and so does the temptation to skip the structure, because senior hiring feels like judgment about people rather than a process to be run. It is both. The examinations test hiring because its failure modes are so consistent: the bar lowered because the seat has been open too long, the single impressive signal that colours everything, the senior hire left to sink, and the interview that measured which tools a candidate was allowed rather than what they can do.

1. Define the bar before the first interview. What the role must be able to do, at what level, decided and written down before any candidate is seen. A bar defined during the process is defined by the candidates in it.
2. Structure beats impression, and it is also fairer. Consistent questions mapped to a rubric, scored independently before any discussion. This is one of the better-evidenced claims in management: structured interviews predict performance markedly better than unstructured ones, and they are less biased in the same stroke.
3. Assess what a candidate can do, not which tools they were permitted. The question is whether the person can reason, design, judge and explain. Whether an assistant was in the room matters only insofar as the interview reveals what the candidate understood of what it produced.
4. Hold the bar under pressure. A bad hire costs far more than an empty seat, and the pressure of the open seat is exactly when the bar is most at risk. If the process is not finding people, change the process or the sourcing, not the standard.
5. Onboarding is where the hire is activated or wasted. Explicit outcomes for the first quarter, an early real contribution, a named guide, and documented context instead of tribal knowledge. The more senior the hire, the more onboarding is integration — norms, relationships, how decisions really get made — and the more its absence is the actual cause of failure.
6. Intervene early on a hire that is not landing. Three months of thin results and a sceptical team is a signal to diagnose and support now, not to wait politely for six-month certainty or to conclude instantly that the hire was a mistake.

Hiring is unusual among the competencies in this standard: the central claim has strong empirical support, and the unit grades it accordingly.

Structured interviews research

The meta-analytic finding in Frank L. Schmidt and John E. Hunter, The Validity and Utility of Selection Methods in Personnel Psychology: Practical and Theoretical Implications of 85 Years of Research Findings (1998), across eighty-five years of selection research, that structured interviews are among the most valid predictors of job performance available, well ahead of unstructured interviews, and that combining them with a work-sample or ability measure improves prediction further. The mechanism is noise reduction: the same questions, a defined rubric, independent scoring before discussion. Structure reduces noise; it does not by itself remove bias. A rubric written by one person encodes that person's assumptions, and the remedy is deliberate design and a panel that does not all share them.

Calibration practice

The practice of aligning interviewers, and later managers, on what "meets the bar" means, so that the same capability earns the same decision regardless of who assessed it. It has no single provenance; it is the operational complement to structured interviewing and the mechanism that keeps a hiring bar consistent across teams and over time. Without it, structure produces consistent scores that mean different things to different scorers.

The halo effect research

Described in Edward L. Thorndike, A Constant Error in Psychological Ratings (1920) and replicated many times since: one salient positive trait colours the assessment of unrelated traits. In hiring it is the impressive credential, the strong first answer, or the fluent presentation that lifts every subsequent score. Independent scoring per criterion before discussion is the direct countermeasure; a panel debrief that starts with overall impressions re-introduces the effect the structure removed. Its cousin, affinity bias — favouring people like oneself — is the reason panels should not resemble the manager.

The first ninety days practice

The framing in Michael D. Watkins, The First 90 Days: Critical Success Strategies for New Leaders at All Levels (2003) of a new role as a transition with its own work: learning, early wins, building the relationships the role depends on. For engineers it becomes explicit 30-, 60- and 90-day outcomes with a first shippable task; for senior leaders it is mostly integration, and the book's central point stands for both — a transition that is left to happen usually happens badly. A plan is not a relationship, and no checklist substitutes for the support behind it.

Hiring as a system practice

At organisational scale hiring is a system with a throughput and a quality, and Will Larson, An Elegant Puzzle: Systems of Engineering Management (2019) treats it as one: sourcing, a bar that survives growth, calibration across many interviewers, and the choice between internal promotion and external hire as a portfolio decision rather than a one-off. The value of the systems view is that it makes the failure visible before it is felt: a bar that drifts one interviewer at a time is invisible per hire and obvious per year.

The role brief and rubric

Before sourcing begins, a one-page brief states what the role must be able to do, the two or three criteria that matter most, and what "meets the bar" looks like for each. The rubric is derived from the brief, not from what recent candidates happened to show. Every interviewer scores every criterion independently and in writing before the debrief.

The debrief that starts with scores

The panel meets with scores already submitted. The debrief begins with the criterion where scores diverge most, not with anyone's overall impression, and the most senior person speaks last. Divergence is the signal the process exists to surface; converging early on a strong opinion re-introduces the halo the structure removed.

Capability-first exercises

Where a technical exercise is used, it is designed so that the candidate's reasoning is visible whatever tools they use: they explain the approach, defend the trade-offs, and modify the result under a changed requirement. An exercise that can be passed by pasting output measures nothing the role needs.

The first-quarter plan

Every hire has a written plan for the first quarter: outcomes at thirty, sixty and ninety days, a first real contribution small enough to ship in the first fortnight, a named guide who is not the manager, and the documents that hold the context the team otherwise carries in its heads. For a senior hire the plan names the relationships to build and the decisions to observe before making any.

The three-month check

At three months the manager asks two questions of themselves: is this person doing what the brief said the role must do, and does the team think so? If either answer is uncertain, the conversation with the hire happens that week, with the specific gap named and the support offered. It is a diagnosis and a repair, not a verdict.

At team scope the object is an individual hire and their first quarter, and the manager runs the process personally. At organisational scope the object becomes the team's hiring bar as a system, calibration across teams so that transfers between them work, and the appointment of managers, which is a hire into a new profession and is onboarded as one. At executive scope the object is the hiring system itself — a bar that survives doubling in a year — and the senior and executive appointments where the choice between an unproven internal candidate and a proven external one has no formula and must be made consciously and resourced either way. The pattern is in How Judgment Scales; who is selected to lead, and why, is PL-6 Leading through leaders.

• Defines the bar first and writes the rubric from it, not from the candidates. team
• Runs structured, calibrated interviews and scores independently before discussion.
• Holds the standard when the seat has been open too long, and changes the sourcing instead. team
• Designs onboarding for an early real contribution, a named guide, and written context. team
• Assesses what a candidate can reason and explain, whatever tools were in the room.
• Failure mode — lowering the bar because the search is dragging. team
• Failure mode — over-indexing on a single impressive signal. team
• Failure mode — "onboarding" that is a week of reading with no real task. team
• Failure mode — assuming a strong hire needs no support. team
• Failure mode — a panel that all resembles the manager, and a rubric nobody has examined for what it assumes.
• Calibrates the hiring bar across several teams so the same capability earns the same decision. org
• Treats a manager appointment as a hire into a new profession, and onboards it as one. org
• Intervenes early and supportively on an external manager hire that is not landing at three months. org
• Failure mode — leaving an external hire to sink or swim, then concluding they "weren't a fit". org
• Failure mode — hiring bars that drift apart team by team until transfers fail. org
• Makes the internal-versus-external bet for a senior seat consciously and resources whichever is chosen. exec
• Treats senior onboarding as the true risk surface: norms, sponsorship, early direct feedback before any conclusion. exec
• Runs a hiring system whose bar survives growth. exec
• Failure mode — defaulting to the "safe" external hire and under-supporting them anyway. exec
• Failure mode — concluding a struggling senior hire is not a fit without having integrated them. exec

Speed versus standard. Every week a seat is open costs the team, and every week of pressure erodes the bar. The resolution is never to move the bar; it is to move everything else — sourcing, process latency, the brief's must-haves versus nice-to-haves — and to say explicitly what the empty seat is costing so the trade-off is visible.

Structure versus judgment. Structured interviews are more predictive, and experienced managers feel that structure suppresses what they know. Both are true. The judgment goes into designing the rubric and reading the evidence; it does not go into overriding the scores with an impression, which is precisely the noise the structure exists to remove.

Capability versus fluency. A candidate who presents well, or produces a polished artifact quickly with assistance, looks capable. The interview has to separate the two, and the separation costs time and design effort that pressure argues against.

Internal versus external. The internal candidate knows the organisation and has a visible trajectory but has not done the job; the external candidate has done it somewhere else and must integrate. Neither is safe. The discipline is to choose consciously and then fund the risk of the choice: coaching and air cover for the promotion, deliberate integration for the hire.

Fit versus sameness. A panel converges fastest on candidates who resemble it, and calls the result fit. Some of what is called fit is real, meaning shared standards; some is affinity. The structure exists to tell them apart, and a panel that never disagrees is not evidence that it has.

A head of engineering must fill a vice-president seat that leads a third of the organisation. Two candidates are in front of them. The internal candidate is a director who has run one strong team of teams for two years, is trusted by their people, and has never operated at the new scope. The external candidate has held the equivalent role at a larger company, interviewed well, and would arrive with no relationships and a very different working culture behind them. The board's chair has said, informally, that the external hire is "the safe choice".

There is no formula, and the first judgment is to refuse the framing that one option is safe. The external hire carries integration risk that is usually underestimated and usually the reason such hires fail; the internal promotion carries scope risk that is visible and usually over-weighted because it is visible. The head of engineering writes both risks down, in the same terms, and asks what each choice would commit the organisation to.

They run the process they would run for any hire, at this scope: the role brief is written first, the criteria that matter most for this seat are named — in this case, the ability to hold a strategy steady across several organisations and to develop the directors beneath — and both candidates are assessed against them with the same structure, by a panel that includes peers the seat will work with. The internal candidate is not exempted from the process because they are known, and the external candidate is not credited for scale they have not yet shown here.

The evidence favours the internal candidate on the criteria that matter most, and the head of engineering makes the promotion. The decision is then resourced as the bet it is: an explicit conversation about what the new scope demands and where the director has not yet operated, a coaching arrangement for the first two quarters, and visible air cover — the head of engineering routes questions to the new vice-president in front of the leadership team and does not answer for them. The board chair is told the reasoning, the criteria, and what the organisation is doing to make the bet succeed.

Had the evidence gone the other way, the same discipline would have applied in reverse: the external hire's first ninety days would have been treated as the risk surface, with explicit norms, a sponsor on the leadership team, and direct feedback in the first month rather than a verdict in the sixth. What the head of engineering does not do in either case is make the choice by which option is easier to defend if it fails.

• PL-6 Leading through leaders — who is selected to lead, on what evidence, and what they are owed in their first six months.
• CC-3 Inclusive practices and equitable opportunity distribution — affinity, panels that resemble the manager, and the equitable distribution of opportunity that begins at hiring.
• PL-4 Performance management — what happens when a hire is not meeting the bar: expectations, support and honest consequences.
• PL-5 Career development and growth paths — the growth path a hire is joining, and internal candidates as the product of it.
• TJ-5 AI-assisted engineering — assessing capability rather than tool access, and what AI assistance changes about the work being hired for.

1. Why are structured interviews both more predictive and more equitable than unstructured ones?
   Answer

   Because they reduce noise: the same questions, a defined rubric and independent scoring mean the assessment tracks the candidate rather than the interviewer's mood, first impression or affinity. The same mechanism that improves prediction removes much of the room for bias.
2. The seat has been open for four months and the team is exhausted. What changes, and what does not?
   Answer

   Sourcing, process latency and the brief's genuine must-haves can all change, and the cost of the empty seat should be made explicit. The bar does not change; a bad hire costs more than the empty seat ever did.
3. What is the halo effect and what specific step in the process counters it?
   Answer

   One salient positive trait colouring the assessment of unrelated ones. Independent scoring per criterion, submitted before the debrief, and a debrief that starts with divergent scores rather than overall impressions.
4. An external manager hire is three months in with thin results and a sceptical team. What do you do, and what are the two mistakes? org
   Answer

   Diagnose specifically what is not working, name the expectations, and support the correction now. The mistakes are waiting politely for six-month clarity, and concluding instantly that the hire was wrong; most senior-hire failures are integration failures nobody addressed.
5. How should an interview establish capability when candidates may use AI assistance?
   Answer

   By making reasoning visible whatever the tools: the candidate explains the approach, defends the trade-offs, and modifies the result under a changed requirement. An exercise that can be passed by pasting output measures nothing the role needs.
6. When does an unproven internal candidate beat a proven external one for a senior seat, and what does the choice commit you to either way? exec
   Answer

   When the evidence on the criteria that matter most for the seat favours them; there is no formula. The promotion commits you to coaching and visible air cover for the step up; the hire commits you to deliberate integration in the first ninety days. Neither option is safe, and the failure is choosing by which is easier to defend.
7. A technically strong senior hire from a very different company culture is clashing with their new organisation at two months. What comes before any conclusion? exec
   Answer

   Integration: explicit norms, a sponsor, and early direct feedback. It is usually an integration problem before it is a capability problem, and it cannot be judged as capability until integration has actually been attempted.

• Frank L. Schmidt and John E. Hunter, The Validity and Utility of Selection Methods in Personnel Psychology: Practical and Theoretical Implications of 85 Years of Research Findings (1998) — the validity of structured interviews and other selection methods.
• Edward L. Thorndike, A Constant Error in Psychological Ratings (1920) — the halo effect.
• Michael D. Watkins, The First 90 Days: Critical Success Strategies for New Leaders at All Levels (2003) — the transition as its own work; early wins and integration.
• Will Larson, An Elegant Puzzle: Systems of Engineering Management (2019) — hiring as a system; bars that survive growth.