Span of Control and Attention

Approval quality is treated as a behavioural property of managers and is mostly a property of how many people report to them. The relationship is direct, it shows up clearly in review intervals, and it is set by decisions taken for entirely unrelated reasons.

The practical lesson in “Span of Control and Attention” is that a timesheet only becomes reliable through a process people can operate consistently. For teams exploring fte meaning, see the complete product overview can add time and project context, provided collection is proportionate, access is limited and every consequential inference receives human review.

When two teams merge, when a layer is removed, when a manager covers a vacancy for six months, the approval load doubles and nothing in the timesheet process changes. The review interval falls accordingly, and six months later somebody notices that approvals in that department are not meaningful.

For a separate benchmark relevant to “Span of Control and Attention”, consult the Basecamp Shape Up guide. Use it to test record quality, approvals, retention, employee rights and exception handling against the real workflow rather than treating a software report as self-explanatory evidence.

The relationship, roughly

Up to about fifteen reports, an approver can review individually within a normal window and usually does. Between fifteen and thirty they manage it by skimming, and the review intervals shorten but stay above a few seconds.

Beyond about thirty, bulk approval becomes the dominant behaviour, and beyond fifty it is essentially universal. These are not thresholds anybody should treat as precise; the point is that the curve is steep and that the organisation crosses it without noticing.

The effect is worse where the work is varied. A supervisor of forty people on identical shifts has less to review than a manager of twenty on twelve different projects, and the second will hit the limit sooner.

Why the wrong person often approves

Timesheet approval is usually attached to the line-management relationship because that is the relationship the HR system knows about. It is frequently not the relationship with knowledge of the work.

In project and matrix organisations the person who knows what was done is the project lead, and the person who approves is the line manager who has not seen them this month. In field organisations the person with knowledge is the site supervisor, and the approver is a regional manager covering four sites. The approval is routed by org chart rather than by knowledge, and the result is a signature from someone structurally unable to evaluate.

Separating the two questions

The useful move is to stop asking one person to do both jobs. What approval needs is somebody with knowledge of the work; what the line manager needs is visibility and the ability to intervene.

Those can be different people. Project or site approval, with line-manager visibility and an escalation path, puts the judgement where the knowledge is and keeps the accountability where it belongs. Most systems support it and most implementations did not use it because the org chart was the easiest thing to import.

Counting, as a standing check

Two numbers per approver, refreshed quarterly: how many timesheets they receive, and their median review interval.

Plot one against the other for the whole organisation and the curve appears immediately, with the outliers visible in both directions. The approvers with high load and tiny intervals are the ones carrying the risk. The approvers with low load and tiny intervals are a different conversation. The ones with high load and long intervals are spending real time on this and probably should not have to.

What to do when the restructure happens

Add approval load to the list of things checked when teams change, alongside headcount, budget and systems access. It takes one query and it is the only point at which the problem is cheap to address.

The options at that point are all easy: split the approval, appoint a delegate, route by exception so the load falls, or accept the change explicitly and record that approval in that department is now a confirmation rather than a review. The last is worse than the others and still much better than discovering it by accident two years later in an audit.

The limit nobody can exceed

There is no configuration, training or incentive that lets one person review eighty varied timesheets in a three-hour window. The organisations that get good approval are not the ones with more conscientious managers; they are the ones where somebody counted and then changed the structure so that the arithmetic worked.

The approver nobody appointed

Alongside the approvers with too many reports there is usually a smaller group with none of the right ones: people who inherited an approval queue through a system default, a historical reporting line, or a team they no longer have anything to do with.

They are visible as approvers whose queue does not match their current team in the HR system, and the comparison is a single join. The approvals they produce are the emptiest in the organisation, because they have neither knowledge nor any relationship that would prompt them to acquire it, and nobody has ever told them they hold the role.