Held Over: How UK Boards Can Track the Calls They Choose Not to Make

The gap between action and judgement

Most UK boards have an action log. It is usually maintained by the company secretary, reviewed at the start of each meeting and closed out with reasonable discipline. What boards rarely have is a comparable record of the decisions they chose not to take. A matter is discussed, found to be premature, and then disappears into the minutes under a polite formulation: “Noted; to return.”

That is not an action. It is a deferred decision, and it deserves to be treated as one. The distinction matters because deferral is itself a judgement. The board has decided that the timing is wrong, the evidence incomplete or the stakeholders unready. If that judgement is not recorded with the same care as an action, the matter will either drift indefinitely or return months later without the context that made the original deferral sensible.

This is not another minutes audit. It is a question of decision hygiene: how the board preserves the reasoning behind what it has parked, and how it ensures those parked matters come back at the right moment rather than the most convenient one.

What belongs in the register

A deferred-decision register should be short, owned and reviewed. It is not a repository for every unresolved thought. It should capture only those matters on which the board has consciously decided not to decide yet.

At a minimum, each entry needs five fields:

• The decision deferred. One sentence, expressed as a decision rather than a topic. Not “site strategy” but “whether to approve the second-site capital request before the Q1 trading update”.
• The owner. A named director or executive, not a committee. Someone must be accountable for watching the trigger and bringing the matter back.
• The trigger. The observable event or date that should cause the matter to return. “When the lender confirms revised covenant headroom” is a trigger. “When more information is available” is not.
• The review date. A backstop, even if the trigger is expected sooner. This prevents a matter from sitting dormant because the trigger never quite arrives.
• The reason it was parked. A brief note on what the board was waiting for, what options were considered and what would change the judgement.

A simple table in the board pack is sufficient. The register should not become a second set of minutes. Its purpose is to make the deferred decision legible to a director who was not in the room and to the same directors six months later.

Why the reason matters more than the record

The most useful field is the reason. Deferrals tend to fall into a small number of categories: an evidence gap, a sequencing dependency, stakeholder readiness, capacity, or avoidance. Naming the category helps the board challenge itself at review.

An evidence gap is legitimate. The board may rightly wait for due diligence or customer data. A sequencing dependency is also legitimate: one decision cannot sensibly be taken before another. Stakeholder readiness may be a matter of timing with a lender, regulator, major customer or workforce. Capacity is a real constraint, though it should be named honestly rather than disguised as prudence.

Avoidance is the category boards are least likely to record and most likely to benefit from recording. If a difficult call keeps returning and being parked again, the register will show the pattern. That is not an administrative failure. It is governance information.

The cadence: review, not re-litigate

The register should be reviewed at a fixed point in the board cycle, not whenever a matter happens to resurface. A monthly or quarterly slot is usually enough, depending on the board’s rhythm. The chair should lead the review, with the company secretary maintaining the register and the owners speaking to their entries.

At each review, there are only three sensible outcomes. The matter is closed because the trigger has passed and the decision has been taken elsewhere. It is re-deferred with a new trigger and a new review date, and the reason is updated. Or it is escalated to the board agenda because the trigger has passed and the decision can no longer wait.

The danger is the fourth, unofficial outcome: the matter is re-deferred by default because the agenda is full. A register that permits silent re-deferral quickly becomes decorative. The chair should require a new trigger and a new date, and should ask what has changed since the last review.

Between meetings: the trigger test

A deferred-decision register only works if the trigger is monitored between meetings. That is the owner’s job. The trigger should be drafted so that its arrival is observable, not a matter of interpretation. “When the Q1 management accounts show EBITDA below £1.2m” is observable. “When the market settles” is not.

When the trigger is met, the owner should bring the matter forward through the board portal or to the chair, rather than waiting for the next scheduled review. The register is not a substitute for judgement; it is a prompt for it. The board still decides whether the matter is ripe, but it does so knowingly.

What good looks like

A healthy register is unglamorous. It has perhaps a dozen live entries, each with a clear owner, an observable trigger and a date. Some entries will close quickly. Some will be re-deferred once, with a better reason. A few will expose a decision the board has been avoiding.

The test is simple: could the board explain, at any point in the year, what it has chosen not to decide and why? If the answer depends on someone’s memory of a conversation in February, the board does not have a memory for its deferred decisions. It has a habit of hoping they will return. A register replaces the hope with a process, and keeps the board’s judgement visible in the intervals between meetings.