Executive Search Beyond the Rolodex

Executive summary

Executive search exists for real reasons. Senior hiring is confidential, candidate volumes are low, and the best people are often passive rather than actively applying. In that environment, networks are useful. The strongest research does not say that networks are always bad. It says something more uncomfortable: networks can improve search when they transmit genuine information about hard-to-observe capability, but they become dangerous when they substitute for evidence, narrow the field to familiar names, or entrench the preferences of incumbents.

That distinction matters because executive hiring remains unusually exposed to informal selection. The UK’s Davies review found that almost half of directors surveyed had been recruited through personal friendships and contacts, only 4% had a formal interview, and only 1% got the role through responding to an advertisement. The same report said it found no evidence that this had changed substantially by 2011. A later Parker Review consultation found that two in three minority ethnic executives believed they were not in the talent pools or networks of current directors or executive search firms.

The commercial and governance costs are no longer easy to dismiss as simply “soft” concerns. Research on 9,801 US director appointments found that 69% of new directors had professional ties to incumbent boards, even though that connected group represented only 13% of all potential candidates. In more complex firms, some connected appointments appeared useful. But when ties ran to the incumbent CEO, announcement returns and shareholder voting support were weaker. Other studies link stronger executive and director social networks to more aggressive earnings management, while work on selection methods continues to show that structured interviews, job-relevant simulations, and multi-measure assessment outperform loosely intuitive selection.

The practical conclusion is straightforward. Companies do not need to abandon networks. They need to stop letting networks make the decision. The best model is network-enabled sourcing plus evidence-based selection: wide market mapping, explicit outcome scorecards, structured interviewing, calibrated assessment, transparent shortlist criteria, and a post-hire feedback loop that tests whether selection scores actually predict business results. UK and EU reforms are moving in exactly that direction, with the UK Corporate Governance Code requiring formal, rigorous and transparent appointment procedures based on merit and objective criteria, FCA disclosure rules increasing scrutiny, and the EU Women on Boards Directive requiring clear, neutrally formulated criteria and comparative assessment where targets are not met.

The industry and the problem

Executive search is a large, global, influential industry. In 2025, AESC described itself as representing the “$20+ billion” executive search and leadership consulting profession, and its standards describe member firms as operating on a retained and exclusive basis for executive search, primarily in board and C-suite hiring. In other words, this is not a niche corner of recruitment. It is part of the governance machinery through which companies choose who gets to run them.

In the UK, the profession has also become a recognised policy lever. The revised Voluntary Code of Conduct for executive search firms has been signed by more than 100 firms, collectively accounting for the vast majority of UK board work. The Code explicitly tells search firms to broaden candidate pools, share data on track record, and give proper weight to intrinsic competencies and capabilities rather than over-valuing certain kinds of experience. That wording is revealing. It is a tacit acknowledgement that the old model has often privileged familiarity, title history and social proof over demonstrable performance.

The problem is not that executive search firms use networks. Of course they do. The problem is that many boards and clients still treat network proximity as if it were evidence of future performance. That encourages a drift from “Who has done the job well?” to “Who do we already know, who looks like a safe pair of hands, and who comes recommended by people we trust?” When the labour market at the top is already highly homophilous, that drift has predictable consequences. As McPherson, Smith-Lovin and Cook put it, homophily means similarity breeds connection and personal networks become homogeneous across many social characteristics. When hiring depends heavily on networks, access and information also become localised within those circles.

The result is a system that often behaves more conservatively than firms think it does. Board diversity data in the UK show real progress where process reform has been pushed hard, but much slower change in executive power roles. By the end of 2025, women held 43% of FTSE 350 board roles, yet fewer than one in ten CEO roles. Women represented 35.9% of FTSE 350 leadership roles overall, but only 29.3% of executive committee roles. The message is hard to miss: transparency and targets can move visible board seats faster than they move the deeper, less visible, more network-dependent pipeline into operating power.

How network-based executive hiring works in practice

Retained executive search is built around confidentiality, access and advisory judgment. AESC’s own materials describe member firms as exclusive advisers that define the search, identify and assess candidates, treat senior candidates with a high degree of confidentiality, and often use competency-based interviews, referencing, due diligence, psychometrics and broader assessments. That sounds rigorous, and often it is. But it also explains why network dependence can become structural. When a search is quiet, urgent and aimed at passive candidates, the search starts with who can be reached credibly and discreetly.

Academic evidence shows how powerful those connections are. In a large study of US director recruitment, a socially connected candidate had roughly a 28 percentage point higher probability of winning a board seat among otherwise qualified candidates. A later summary of related work reported that being socially connected to someone already on the board could increase the chances of appointment by as much as 41 percentage points. Professional, not merely social, ties matter most. Referrals in this labour market are disproportionately professional and appear to transmit information that is partly invisible on the CV.

That is why network hiring persists. It does solve real information problems. But it also creates a familiar funnel with mixed results.

Two further mechanics matter. The first is “fit”. The UK’s inclusive recruitment guidance from the CIPD warns that cultural fit is subjective and increases bias, partly because it often becomes shorthand for similarity to existing employees. The second is the industry’s off-limits convention, which protects clients from poaching but can narrow who a retained firm is willing or able to approach, especially in concentrated sectors. That is not unethical. It is part of why clients trust search firms. But it can turn the available market into a smaller and more familiar subset than boards assume.

A fair summary is this: networks are best used for sourcing and contextual intelligence. They are much weaker when used as a proxy for merit.

What the evidence says about bias and outcomes

The evidence is strongest when it avoids caricature. Networks are not purely bias. In the 9,801-appointment study, connected directors often helped boards add gender diversity, new skills and new industry background, especially in more complex firms. In the board-referral study, referred directors looked higher ability on measures designed to capture harder-to-observe quality, suggesting that referrals can reduce information asymmetry. The serious question is therefore not whether relationships have value. It is when that value outweighs the exclusion and agency risks they create.

Bias enters when a relationship-based market meets a homophilous elite. The UK’s Davies report found striking dependence on personal contacts in board recruitment. The Parker Review then documented the racial version of the same problem, reporting that two in three minority ethnic executives believed they were not in the talent pools or networks of current directors or executive search firms. The older social-science literature explains why this should not surprise anyone: ties of advice, support, work and information transfer are not randomly distributed. They cluster around people who already resemble one another.

More recent work shows the mechanism operating in elite labour markets. Lalanne and Seabright found that the earnings payoff from professional networks is larger for men than for women in European and North American executive samples, and a public summary of the work reports that so-called “female-friendly” firms appear to reward networks less, possibly because they rely more on objective criteria than on contacts. A 2025 field study in Sweden found strong gender homophily in referrals: 71% of female participants referred a female candidate and 75% of male participants referred a male candidate. Again, this does not prove malice. It shows how a seemingly neutral referral mechanism can reproduce whatever structure already exists.

The measurement problem is equally important. Selection science has spent decades comparing intuitive and structured methods. Sackett and colleagues’ updated meta-analysis found that structured interviews emerged as the top-ranked selection procedure. Follow-on work using the updated matrix found that structured interviews carried the greatest weight in validity-maximising selection combinations. Assessment-centre evidence is also material for senior hiring: a meta-analysis found a corrected correlation of .28 between overall assessment-centre ratings and supervisory performance ratings, while an AESC white paper argues that multi-measure assessment and properly designed psychometrics can support fairer and more objective comparison. None of this means executive hiring can be reduced to a test battery. It means that “I know them” and “they interviewed well” are weak substitutes for systematic evidence.

The performance and governance effects are real, though they vary by context. In the director-appointment study, connected appointments were not uniformly bad; firms with more advice needs sometimes benefited. But appointments connected to incumbent CEOs specifically were associated with lower announcement returns and lower shareholder votes. Other research links stronger external social networks of top executives and directors to more aggressive earnings management. On the diversity and governance side, studies have found positive associations between female directors and bank stability across 14 countries, and between Chair-CEO diversity and firm performance in UK listed firms. The broad “diversity always raises profits” claim remains too simple, but the narrower finding is stronger: narrower, more insider-heavy appointments can weaken independence and increase governance risk, while broader and better-evidenced appointments improve the odds of stronger challenge, oversight and resilience.

Case studies and contrasts

The first case is a UK board-market case, and it is the clearest. Before the last fifteen years of reform, UK board recruitment was heavily informal: friendships and contacts dominated, formal interviews were rare, and open advertisement was almost absent. After the Davies review, the search-firm code, stronger governance expectations and FCA disclosure rules, the process became more transparent and measurable. By the end of 2025, women held 43% of FTSE 350 board roles and 35.9% of leadership roles. Yet women still held fewer than one in ten CEO seats in the FTSE 100. The contrast is telling. Process reform widened visible board access. The remaining bottleneck is the executive pipeline, where informal sponsorship and network visibility still matter more.

The second case comes from the 9,801 US director appointments study. Here the contrast is inside the network model itself. Professional ties to incumbent boards were common and, in more complex firms, added coordination value, skills and even some diversity benefits. But when the tie ran specifically to the incumbent CEO, shareholder reaction worsened. That is the difference between network as information and network as capture. The lesson for boards is not “never hire connected people”. It is “treat CEO-linked enthusiasm as a conflict risk and demand more evidence, not less”.

The third case is an anonymised large-bank promotion system, but it maps directly onto executive pipelines. Cullen and Perez-Truglia found that greater face-to-face interaction with managers increased promotions and could explain up to a third of the gender gap in promotions at the firm they studied. Crucially, they did not find matching gains in effort or performance measures like days worked, time in the office or sales revenues. In other words, social access changed careers even when measurable output did not. If the route into senior leadership is shaped like that, executive search later inherits a pipeline already distorted by relationship capital rather than results.

A fourth, more public but less controlled contrast is useful as an illustration rather than proof. J.C. Penney hired Ron Johnson in 2011 on the strength of his celebrated Apple retail record and outsider narrative; Reuters later reported a 25% sales plunge in 2012 and a further 16.6% same-store sales decline in the first quarter of 2013. Best Buy, facing its own crisis, appointed Hubert Joly in 2012 explicitly to tap his restructuring and turnaround track record; Best Buy’s own announcement highlighted fifteen years of successful turnarounds and growth, and Reuters reported in 2014 that his turnaround efforts were showing signs of progress. This is not clean causal science, and it is not purely a network story. But it does underline a broader executive-search point: narrative prestige and prior brand halo are not the same thing as role-relevant evidence.

Why the model persists

The most obvious reason is that boards are often underprepared. Heidrick & Struggles found that CEO succession planning sat near the bottom of areas where boards had increased time post-Covid, and that 57% of CEOs and directors had little or no confidence that their company’s CEO succession planning was positioning the organisation well for the future. When succession is weak, the emergency response is predictable: ring a trusted search partner, ask for safe names, and lean harder on known networks.

The second reason is economic. Retained executive search is paid as an assignment-based advisory model. AESC says member firms charge a consulting fee or retainer for the search and are involved from definition through integration. In market practice, many firms also offer replacement guarantees measured in months rather than years. That creates an incentive structure heavily focused on successful completion and acceptable early tenure, even though the true value of a CEO, CFO or divisional leader often only becomes visible over a much longer operating cycle. Terms vary by firm, but the mismatch is real.

The third reason is that clients ask for the wrong things. Both the UK search code and the Corporate Governance Code now push back on over-valuing familiar backgrounds and emphasise objective criteria, merit, rigour and transparency. The need for that language suggests how often clients still arrive with narrow proxies for safety: “Must have been a public-company CFO”, “Must already be a listed-company chair”, “Must fit our culture”, “Must be known to the board”. Those proxies can sometimes be relevant. Too often they are just socially acceptable ways of shrinking risk perception by shrinking the candidate pool.

The fourth reason is that data are patchy. The Parker Review has explicitly noted legal and practical challenges around ethnicity data and data protection for recruitment and executive search companies. Many private-company searches remain confidential, and post-hire outcome data are not routinely disclosed in a way that would let buyers compare search firms on real quality-of-hire over time. In other words, the industry often lacks the feedback loop that would force it to learn publicly from selection quality rather than private reputation.

Recent reforms are beginning to change incentives, though unevenly. The UK now has the revised search-firm code, FCA board and executive management diversity disclosure rules, and a Corporate Governance Code that requires formal, rigorous and transparent appointment procedures based on merit and objective criteria. At EU level, the Women on Boards Directive goes further where targets are not met, requiring fair and transparent selection procedures with comparative assessment and clear, neutrally formulated criteria established in advance. That is important because it shifts the debate from quotas versus merit to process quality as the route to merit.

How to build results-based executive hiring

The practical aim is not to remove judgment. It is to discipline judgment. Boards, companies and search firms should keep using networks to surface, persuade and reference candidates. But selection itself should move onto a more auditable footing.

Start with the role, not the person. Before any names are discussed, the board or nomination committee should agree a one-page role scorecard containing five to seven measurable outcomes for the first 12, 24 and 36 months. For a CFO, that might be working-capital turns, capital allocation quality, close-cycle discipline, audit quality, financing resilience and team build-out. For a business unit CEO, it might be revenue quality, margin improvement, safety, customer retention, digital adoption and succession depth. The search brief should then state how each shortlisted candidate has produced analogous results at comparable scale and complexity.

Then separate sourcing from selection. Ask the search firm for three things, not one: a network-sourced list, a full market map, and a statement showing what proportion of the longlist comes from beyond warm relationships. Networks are allowed to open the door. They are not allowed to close the field.

Next, use structured comparison. The evidence base supports structured interviews, assessment-centre style exercises, and multi-measure assessment far more strongly than free-form panel conversation. For executive roles, that usually means a consistent interview rubric, a strategy or leadership simulation, calibrated referencing tied to specific claims, and where appropriate a psychometric or leadership-assessment component. The point is not to create false scientific certainty. It is to make candidates genuinely comparable.

Boards should also redesign mandates and contracts. If fee structures remain fully front-loaded, companies should at least add process and outcome conditions to the RFP and the statement of work. That can include requirements for market-map breadth, diverse slate thresholds, structured-assessment compliance, and post-hire review participation at 12 and 24 months. Some firms will resist. That resistance is informative.

Finally, close the loop. Every senior hire should be reviewed against the original scorecard after one year and again after two years. The board should ask a brutally simple question: did the evidence used in the selection process actually predict performance? If not, either the scorecard was wrong, the assessment model was wrong, or the board ignored it in favour of instinct. All three are fixable, but only if they are tracked.

The table below sets out a practical division of labour.

The most useful KPIs are the ones that force learning rather than simply reporting.

A sensible contractual incentive for search firms is a modest but meaningful tranche of fees linked not to raw business performance, which can be unfairly noisy, but to jointly auditable milestones: completion of market map, shortlist breadth, assessment compliance, and participation in 12-month quality-of-hire review. The firm should not control company outcomes. It should be accountable for search quality.

There are obvious objections. Some are reasonable.

Sources and limitations

This article draws primarily on AESC standards and research, UK government and governance documents, FTSE Women Leaders Review data, Parker Review materials, FCA rules, EU board-balance rules, and academic studies in finance, labour economics and personnel selection. Key sources include AESC’s Professional Practice Standards and How Executive Search Differs from Contingent Recruiters; the Davies Women on Boards review; the revised UK Voluntary Code of Conduct for Executive Search Firms; the UK Corporate Governance Code 2024; FCA Policy Statement PS22/3; the FTSE Women Leaders Review 2026; Parker Review materials and press releases; Cai, Nguyen and Walkling’s Director Appointments: It Is Who You Know; Lalanne’s work on referrals and networks in board recruitment; Cullen and Perez-Truglia’s The Old Boys’ Club; Sackett et al. on selection validity; and Hermelin, Lievens and Robertson on assessment-centre validity.

Two limitations are worth stating clearly. First, the best public evidence is richer for listed-company board appointments and internal promotion systems than for private, confidential C-suite search mandates. That is partly because executive searches are often confidential by design. Second, public outcome benchmarking for search firms remains weak, so some of the strongest recommendations here are process recommendations intended to make later outcome learning possible.