A good Non-Executive Director adds value. A governance function does something else entirely, and most boards have quietly mistaken the first for the second.
The substitution has consequences beyond semantics. A board that believes it has a governance function because it has a capable NED cohort is exposed in exactly the way governance failures have been exposed for the past twenty years. The model produces the exposure. The cohort, however able, operates inside it.
The NED sits inside the consensus the role is meant to challenge
The NED is on the board. The board reaches decisions through consensus. Challenging the consensus that produced a decision, after sitting through the meeting at which it formed, requires both the standing to do so and the willingness to be the person who keeps raising the concern after everyone else has moved on.
This is why NED challenge is intermittent and personality-dependent rather than systematic. The NED who challenges most effectively is the one with the least to lose — the longest track record, the most alternative board roles, the highest standing relative to management. The model produces challenge as a function of individual temperament. The architecture does nothing to produce it.
Defensive decision-making explains why challenge does not happen
The behaviour has a name. Rory Sutherland and Gerd Gigerenzer call it defensive decision-making. It means choosing a knowingly inferior option because, if it fails, the blame falls on the system rather than on the individual who chose it. Nobody was ever fired for going along with the room.
Kahneman documented the mechanism precisely. Asked whether they would accept a bet with a 50% chance of a large gain and a 50% chance of a smaller loss, eight of ten division heads at one Fortune 500 firm declined, because the downside was career-terminal for them as individuals even though the aggregate bet was strongly positive for the firm. The decision that protects the institution and the decision that protects the decision-maker diverge, and in a committee the second one wins.
A NED challenging board consensus is taking exactly the bet those division heads declined. The career penalty for a persistent, unsuccessful challenge is real. The reward for a successful one is modest. Under those incentives, challenge is individually irrational, and committees converge on defensible positions regardless of quality. You cannot fix this with a better calibre of NED. The incentive sits in the structure itself, and it operates on whoever occupies the role.
The survey data shows the gap directly
The 2024 Global Board Governance Survey by Protiviti, BoardProspects and Broadridge polled 1,006 directors and C-suite executives. Its central finding is a perception gap. Directors rate the board higher than the executives who watch it work. Ninety-five percent of directors said the board is constructively engaged and asks probing questions; only 80% of the C-suite agreed. On preparedness for talent risk, 50% of directors rated the board very or extremely prepared against 29% of the C-suite. The same direction held across every threat category examined.
A board that scores its own challenge higher than the management it oversees does is describing a model that creates the experience of governance without reliably delivering the substance. The papers are reviewed. The discussion happens. The decisions are approved. The adversarial challenge that would surface the assumptions behind those decisions is structurally absent, and the people closest to the board can see it.
The Big 4 have the same problem in a different form
The alternative most boards weigh is a project engagement from EY, Deloitte, KPMG or PwC. The limitation is analogous, and it presents differently.
A Big 4 governance engagement is project-based. A team assembles, conducts the review, produces the deliverable and leaves. The institutional knowledge built during the engagement departs with the team. The next engagement, if there is one, starts from a standing start. The challenge function is episodic by design, and the understanding of the specific governance architecture does not survive the project boundary.
There is a second dynamic worth naming. Boards default to the Big 4 partly because the choice is defensible. The defensibility of the brand, more than the quality of the work, drives the decision. The same defensive logic that suppresses NED challenge also steers the procurement decision toward the safe brand. It is the institutional version of nobody ever got fired for buying IBM.
My ten years as UK Head of Fiduciary Risk Management at Credit Suisse are the reference point I draw on. The function operated continuously across the portfolio, accumulated the knowledge that makes analysis specific to the institution, and produced work that changed decisions, where a project report typically documents decisions already made. That standard of continuity and specificity sits beyond the reach of both the NED model and the project engagement. It is what an institutionalised challenge function reaches when it is built to.
What a genuine governance function requires
An adversarial governance function has four structural characteristics the conventional model lacks.
Scope: the mandate covers the governance architecture broadly, extending past the remit of any single committee. A challenge function confined to what the audit committee already reviews is a compliance function.
Continuity: the function operates across board cycles and accumulates knowledge over time. An annual review delivers a snapshot. A quarterly NED meeting delivers an update. Both leave the intervals between them ungoverned.
Independence: the challenger is structurally free from the consensus being challenged. The person doing the challenging cannot sit inside the committee whose assumptions are in question. This is a structural constraint that holds regardless of preference.
Mandate: the function is required to challenge, where the conventional model merely permits it. Optional challenge will not be exercised consistently, for the reasons the defensive decision-making evidence makes plain. Required challenge, with a defined scope and a clear reporting line, will be.
The NED model provides none of these reliably. The Big 4 model provides continuity and mandate within a project scope, but the continuity ends at the project boundary and the mandate is advisory.
From the SGaaS White Paper
Why traditional models under-deliver
The white paper chapter "Why Traditional Models Under-Deliver" develops this analysis in full, with the supporting evidence base and the four-tier SGaaS delivery architecture that addresses each of these structural gaps directly.
Owen Vallis is the founder of Marentis Labs, the firm that originated Strategic Governance as a Service. He spent ten years as UK Head of Fiduciary Risk Management at Credit Suisse and holds active board roles in the public and charity sectors. Schedule a confidential discussion.