Beyond Headcount: What Boards Should Actually Measure

The question the disclosure cannot answer

I was sitting in a board composition review where the numbers, by any reasonable measure, looked good. Gender balance had improved. Ethnic diversity met the targets that mattered. The disclosure tables were compliant and the committee was rightly pleased. Then the discussion turned to a strategic pivot the company was considering, and something quieter became visible. Almost everyone at the table had reached it by the same route. Similar sectors. Similar functional ladders. Similar governance training. Similar formative decades. The people were different. The journeys were not. What was absent was harder to name but easy to feel. Nobody in the room had operated outside the industry's home markets, scaled a business through a downturn, or built anything from scratch. The table looked more diverse, but the repertoire around it had moved less than the disclosure suggested.

That observation does not undermine what boards have achieved. The Parker Review's March 2026 report showed 98 FTSE 100 companies meeting the target of at least one director from a minority ethnic group on their boards (Parker Review, 2026). FCA listing rules require diversity disclosure against defined targets (FCA, 2022). The UK Corporate Governance Code expects boards to describe how composition supports strategy (FRC, 2024). This is progress, and it is measurable precisely because regulators made it so.

But those regimes mostly answer one question. Who is in the room? They do not, by themselves, answer the question boards care about every time strategy shifts. How similar are the routes into that room still?

This is one of seven disciplines I am publishing on evidence-based leadership decisions. Each examines a different point at which the evidence and the confidence diverge.

Why boards still reproduce themselves

The sociology here is old and uncomfortably durable. Westphal and Zajac showed that board appointments favour candidates who resemble incumbent directors, in background, in outlook, and in social position (Westphal & Zajac, 1995). Westphal and Stern went further, demonstrating that ingratiating behaviour toward existing directors improved appointment prospects, particularly for candidates who lacked conventional elite credentials (Westphal & Stern, 2006). The routes into the boardroom run through networks, and networks reproduce themselves.

Granovetter explained the mechanism decades earlier. Opportunity travels through ties, and the ties that matter most are often the weak ones that bridge otherwise separate worlds (Granovetter, 1973). Burt made the structural version of the point. People whose networks span structural holes, the gaps between clusters, are where fresh information and genuinely new ideas enter a system (Burt, 2004). A board built from one cluster can be demographically varied and structurally closed at the same time.

None of that research says demographic diversity is unimportant. It says something more precise. Network-driven selection narrows the range of journeys a board contains, and demographic reporting was never designed to detect that narrowing.

What the diversity literature actually distinguishes

The construct problem was named years ago. Harrison and Klein pointed out that "diversity" is not one thing. Separation, variety, and disparity are different constructs with different effects, and collapsing them produces confused claims (Harrison & Klein, 2007). A board can improve on one construct while standing still on another.

The outcome literature reflects that complexity. Post and Byron's meta-analysis found that female board representation relates positively to accounting returns, with effects that vary by context, and much weaker direct relationships to market performance (Post & Byron, 2015). Georgakakis and colleagues showed that the effects of top-team diversity depend on which dimensions combine, in which contexts (Georgakakis et al., 2017). Woolley and colleagues showed that collective intelligence in groups tracks social sensitivity and how evenly conversation is shared, rather than the average capability of individual members (Woolley et al., 2010).

And the strategy literature adds the dimension boards feel most directly. Bantel and Jackson found that more heterogeneous top teams were associated with greater innovation in banking (Bantel & Jackson, 1989). Wiersema and Bantel found that team composition shaped openness to strategic change (Wiersema & Bantel, 1992). Repertoire matters when the strategy has to move.

Here is the honest reading of that evidence. Diversity effects are real, contingent, and construct-dependent. The literature does not license a slogan in either direction. What it licenses is a sharper question. Which kinds of difference does this board actually contain, and which has it only reported?

This is where I abstain from a stronger claim. The research does not show that any single composition mix drives performance. It shows that boards rarely measure the constructs that would let them answer the question properly.

Trajectory Divergence

Trajectory Divergence is the discipline of measuring how varied the routes to each seat actually are, alongside the demographic composition the board already reports. The claim is deliberately bounded. This is not an alternative to demographic diversity, and it is not a proxy for it. It is a second lens the disclosure regime does not capture, and it is measurable from the record of careers rather than from anyone's identity.

Four lenses make it testable:

First, career-path divergence. How many distinct routes to the boardroom does the table contain? Executive ladders inside one sector, professional services, entrepreneurial building, careers run across international markets, public-sector leadership, technical depth. A board where every seat was reached by the same ladder has one route, however varied the people on it.

Second, sector and market divergence. How many industries, geographies, and market conditions has the board collectively operated in? Growth markets and declining ones. Regulated and unregulated. Domestic and genuinely international. A strategy discussion draws on the conditions the table has actually operated through.

Third, functional divergence. Which disciplines have the directors actually run? Finance and legal are usually well covered. Technology, supply chain, digital operating models, and transformation delivery often are not. The gap becomes visible the moment the strategy depends on one of them.

Fourth, formative-context divergence. Across which decades, cycles, and shocks were the board's instincts formed? A table whose formative experiences all sit inside the same twenty years will reach for the same playbook under pressure, however diverse it looks in the annual report.

None of these lenses requires new disclosure regimes or personal data beyond the career record. Every one of them can be evidenced from what directors have actually done. That is the point. Trajectory Divergence is a claim about the record, not about identity, and it is auditable in exactly the way this series keeps demanding.

The question boards should add

What I have seen is disclosure improving while the pathway narrowed quietly underneath it. The evidence that would overturn that reading is specific. Boards showing that demographic disclosure alone reliably captures the breadth of repertoire around the table, or composition reviews that routinely measured career, sector, functional, and formative divergence and found those routes already varied. Neither has appeared.

In “What Boards Can and Cannot Attribute to a CEO”, I argued that boards should separate what they can attribute to a CEO from what belongs to context. In “Where Leadership Analytics Goes Wrong and How to Fix It”, I argued that systems making claims about leaders should publish where their claims stop. Composition deserves the same discipline. A board that reports who is in the room but cannot describe how varied the routes into the room are, is making a completeness claim its own record does not support.

Before the next composition review, the board should be able to answer both questions. Who is in the room, measured as the disclosure regime requires. And how divergent are the journeys that got each person there, measured across career path, sector exposure, functional depth, and formative context.

Written by James Nash.

First published on inBeta.io. Co-published on Substack. Summer 2026.

Series: The Seven®, by James Nash. © Copyright 2026 inBeta. inBeta, Optics, Divergence and The Seven are all trademarks of inBeta Ltd