Slow Yes, Fast No: Diagnosing Decision Latency in UK Management Teams

Ask an executive team why the organisation struggles to adapt and the answer is usually that the strategy needs sharpening. In most cases the strategy is fine. Everyone in the room can recite it. The real constraint is that the organisation takes eleven weeks to decide anything, and by the time it decides, the conditions that framed the choice have moved.

This is an argument for treating responsiveness as a governance output in its own right, measured and managed alongside financial performance.

What decision latency is

Decision latency is the elapsed time between the point at which a decision could reasonably have been taken and the point at which it was taken and communicated to the people who must act on it. 'Could reasonably have been taken' matters: it is the moment the evidence was sufficient and the owner identifiable, not the moment someone first had the idea.

In practice the interval breaks into four segments:

• Queue time: from a ready proposal to its first agenda slot.
• Deliberation: from first appearance to a decision.
• Ratification: from decision to formal sign-off.
• Broadcast: from sign-off to the affected teams knowing what has changed.

In most UK management teams, deliberation is short and queue and ratification are long. The meeting where the matter is settled may take twenty minutes. The wait either side of it takes two months. Optimising the meeting is therefore often the least valuable intervention available.

Why the distinction matters

Latency is not the same as deliberation, and speed is not the same as haste. Some decisions deserve extended consideration: irreversible commitments, material capital, anything with safety or regulatory consequence. The problem is that organisations rarely differentiate. A £15,000 software renewal and a market entry decision travel the same route, and both wait for the same quarterly cycle.

The cost of that uniformity is cumulative. Options expire. Good people conclude that initiative is not rewarded. Decisions that were once straightforward become urgent, and urgent decisions are made with fewer options and weaker analysis. Latency does not remove risk; it relocates it to the end of the process, where there is least time to think.

Measuring it without buying anything

A first measurement can be done by a capable business manager in a fortnight.

Select between twenty and thirty decisions concluded in the past six months, deliberately mixing scale and type. For each, reconstruct five dates: when the proposal was ready, when it first appeared on an agenda, when it was decided, when it was formally signed off, and when the affected teams were told. The reconstruction will be approximate, and it will be revealing.

Report the median, then report the eightieth percentile. The tail is where the damage sits. An organisation with a four-week median and a fourteen-week tail has a governance problem concentrated in a minority of cases, and those cases are usually the ones that matter.

Segment by decision type: capital, people, commercial, regulatory and risk. Most organisations find that one category carries almost all of the tail, frequently capital or anything requiring cross-functional sign-off.

Then instrument forward. A decision log with five fields, maintained by the secretariat and reviewed monthly, turns an anecdote into a management metric within a quarter.

Four structural causes

The first is undefined decision rights. Papers are circulated to everyone and owned by no one, so consensus becomes the de facto requirement and any single objection resets the clock.

The second is committee layering. The same seven people sit on three forums, each of which must consider the matter, and each of which adds a cycle rather than a perspective.

The third is information asymmetry. The person holding the data is two levels below the person holding the authority, and the gap is bridged by briefing documents that take a fortnight to produce.

The fourth is risk aversion expressed as process. A single failure prompts an additional sign-off. It is never removed, because removing it would require someone to accept a risk in writing.

Changes that make a measurable difference

Name the owner and the type. Every decision has one accountable individual, a short list of people who must be consulted and a longer list who are merely informed. One line in the paper, agreed before the paper is written.

Publish thresholds and let them operate. Delegated limits for spend, headcount and contract terms, reviewed annually and available to all managers. Most organisations have them; far fewer have managers who know what they are.

Redesign the meeting, not just the agenda. Require every paper to state the decision requested, the date by which it is needed and the consequence of not deciding. Circulate pre-reads 48 hours ahead and refuse items without them. Separate information items from decision items entirely. Cap attendance, and make clear that observers do not speak.

Set defaults. Where a decision has not been taken by a stated date, a pre-agreed outcome applies. Used selectively for low-stakes, reversible matters, this is the single most effective tool available, because it converts inaction into a decision with consequences.

Handle reversibility explicitly. Reversible decisions should be delegated downward and taken quickly. Irreversible ones should be escalated and given proper time. Most organisations apply the same weight to both, which is why they are simultaneously slow and reckless.

A reasonable target

For an organisation of any size, a defensible starting ambition is a median of ten working days for reversible operational decisions and six weeks for capital decisions above threshold, with the eightieth percentile reported and tracked separately. Assign the metric to a named executive. What is measured in a board pack eventually improves.

Some will argue that faster decisions will be worse decisions. The evidence tends to run the other way: delay does not usually buy additional analysis, it buys additional circulation. The analysis was done early; the waiting is administrative.

A strategy is a document. Responsiveness is a capability. The distance between the two is measured in weeks, and it can be counted.