A Rival Read: Writing the Right to Disagree Into the Engagement
Photo: board meeting discussion challenge executives around table, via img.freepik.com
There is a particular kind of meeting that most non-executives will recognise. The recommendation is on the table, the supporting pack runs to sixty pages, and someone around the table says, reasonably, that they would like to hear a contrary view. The conversation that follows is courteous and, in substance, decorative: everything has been built, tested and priced on the assumption that the recommendation proceeds. The dissent is real, but it arrives with no machinery behind it.
Boards do not lack the appetite for challenge. They lack a mechanism that gives challenge a place in the calendar before the decision has hardened. That is a design problem, and it is fixable at the point of engagement.
The debate that arrives too late
The natural rhythm of an advisory engagement pushes challenge to the end. Analysis builds towards a recommendation; the recommendation is presented; questions are invited. By then the client team has invested months in the work, the adviser has a stake in its own conclusion, and the timetable permits exactly one meeting. Nobody behaves badly. The structure simply leaves no room for a genuinely rival reading to be developed with the same care as the primary one.
Some firms respond by asking the incumbent to argue against itself — a red-team exercise conducted by the same team, with the same data and the same instincts. It is better than nothing. It is not a second opinion.
The clause, in outline
A second-opinion provision is short enough to sit in a single schedule, and it should be agreed with the engagement letter rather than negotiated when it is first needed. Four elements matter.
Who may invoke it. Two named officeholders, not a committee. A common arrangement is the internal sponsor together with one non-executive director, with either able to trigger the provision without the other's consent. Naming individuals avoids the diffusion of responsibility that makes such rights unused.
When. A defined window rather than an open-ended entitlement: for example, not earlier than the midpoint review and not later than five working days before the final recommendation is tabled. A window protects the timetable and forces the challenger to engage while the shape of the answer is still fluid.
What it produces. A written rival read of no more than a few pages, addressed to the board rather than to the incumbent team, setting out the strongest available alternative interpretation of the evidence and the conditions under which it would be correct. Not a critique of the report; a competing read of the problem.
At whose cost. Capped, agreed in advance, and borne by the client. A cap expressed as a modest percentage of the engagement fee is usually sufficient to buy a few days of genuinely senior attention. The incumbent's fee is unaffected, which is important: the provision must not become a mechanism by which one supplier funds its own challenge, nor a penalty on the original appointment.
Making it real rather than symbolic
Three practical details determine whether the clause is used.
First, the rival read should come from outside the incumbent firm, whether a different adviser, a specialist with contrary experience, or a senior internal voice with the standing to disagree. The point is divergence of instinct, and instinct rarely diverges within one team.
Second, the incumbent should have a right of reply — a short written response within the same window. This is not symmetry for its own sake. A recommendation that survives a properly argued alternative is materially stronger, and the reply is where that strengthening happens.
Third, the board should see both documents together. A rival read that circulates to the sponsor and dies there has purchased nothing except a filing entry. The two papers should arrive as a pair, with the decision-making body given the same amount of time to read each.
The obvious objection
The objection is that this institutionalises doubt, slows decisions, and gives a minority the means to reopen settled questions. There is something in it. Any right of challenge can be abused by someone who simply wishes to delay.
Two features of the design answer this. The window is fixed and short, so the disruption is bounded. And the cost cap, borne by the client, creates a price for using the right — not a large one, but enough that it is invoked when there is a real alternative to argue rather than a general unease to ventilate.
What it changes
The clause does not make boards braver. It makes the existing courage usable, by placing it at a point in the engagement where it can still alter the outcome. A firm that can point to three or four occasions on which a recommendation was tested and either revised or confirmed on the record has something more valuable than consensus: it has a defensible decision.
That is what governance is for, and it is cheaper to buy at signature than to reconstruct afterwards.