Sit down with a cup of tea and a blank page, and draw a company of 1,000 people as a tree. At the top is one person. Under them go some managers, under those managers more managers, and at the bottom the people who do the actual work. The only decision you have to make is how many boxes hang under each box. That number is the span of control, and it is more powerful than it looks.

Let me show you with arithmetic, because the arithmetic is what people forget. If every manager has 4 direct reports, you need six levels to cover 1,000 people (1, 4, 16, 64, 256, 1,024 boxes, and you run out at the sixth level). With 6 reports each, you need five levels. With 10 each, you need four. Same company, same work, but the shape changes from tall to flat.

And the number of people who carry the title “manager” changes too. In a full tree, managers are roughly the total headcount divided by the span. So about 250 managers at a span of 4, about 167 at a span of 6, and about 100 at a span of 10. If a manager costs the company even a little more than an individual contributor, that difference is real money. If a manager is also a good engineer or salesperson who has been pulled away from the work, it is real lost output as well.

Why narrow feels safe and still hurts

A narrow span is what most of us secretly want when we become managers. We can know everyone well, review everything, and nothing slips. But look at what it does to the rest of the tree. Tall organisations pass information through more hands. A decision made at the top reaches the floor after five handoffs, each one adding a little delay and a little reinterpretation. Layers also create promotion ladders that are short on real change in responsibility. A “senior manager” who manages four people is often doing the same job as a “manager” one level below, with a longer title.

Narrow spans also tempt managers to hover. When you have three reports and plenty of time, micromanagement is the path of least resistance. Some of the most common complaints in exit conversations, being watched too closely or not being trusted with decisions, grow naturally in a structure where managers have spare capacity.

Why wide is not simply efficient

The opposite error is just as common, and it usually arrives under the name of “flattening”. A cost-cutting exercise removes a layer of managers, and 6 reports quietly become 15 or 20. On the organisation chart it looks neat. In practice, the manager now has twenty one-to-ones that cannot all happen weekly, twenty appraisals to write, twenty leave requests, and nobody with time to coach the two who are struggling. The weakest and the newest people, who need the most attention, get the least.

Wide spans do work in some conditions. They work when the tasks are repetitive and well defined, such as a team of customer-support agents using the same scripts, or an assembly line where the output is visible by the hour. They work when people are experienced and need little supervision. They work when the manager is not also expected to deliver individual work and when systems handle the admin: dashboards, self-service leave, automated reporting.

They fail when the work is complex, when people are junior, when the team is spread across cities and time zones, or when the manager is a player-coach with a full workload of their own.

What actually sets the right number

There is no single correct span, and anyone who quotes “seven” as a law is repeating a rule of thumb. Writers on organisation design, from the classical management thinkers onwards, have argued about this for a century without settling on one figure. What they agree on is that the right span depends on a handful of things you can actually examine.

  • How similar the reports’ jobs are. Ten people doing the same job are easier to manage than five doing five different things.
  • How experienced they are. A team of freshers in their first year needs far more manager time than a team of seven-year veterans.
  • How much of the manager’s week is their own work rather than people work.
  • How much the team has to coordinate with each other and with other teams.
  • Physical and time-zone spread. A team across Pune, Manila and Dallas costs more manager attention per head than one sitting in a single bay.
  • How good the supporting systems and team leads are.

A practical way to test a span is to count hours. A decent manager of a team doing complex work might reasonably need an hour a week per person for a real conversation, plus time for hiring, reviews, escalations and planning. Put that into a 45-hour week and you will see why a span of 15 with no team leads is a promise that cannot be kept, while a span of 5 for a manager with almost nothing else to do is a waste.

Reading the numbers in your own company

The HRMS already has what you need. Export every manager and count direct reports. Then look at the distribution rather than the average, because an average span of 7 can hide a lot of managers with 2 and a few with 18.

Three patterns are worth looking for. Managers with one or two reports are usually a historical accident: a person promoted to keep them, or a team that shrank. Managers with spans far above their peers are usually where attrition and weak engagement scores will appear, because the people there are rarely coached. And a long chain of single-report layers, where a manager manages one manager who manages one manager, is nearly always a sign that titles are standing in for pay.

Before you redraw anything, remember what a span change does outside the chart. Moving a team to a new manager resets relationships, appraisal history and trust. Removing a layer can be right, but it needs a plan for who now writes the appraisals, who handles the promotion conversations and where the displaced managers go. Otherwise you save the manager salaries and pay for them later in attrition.

Figures and names in the examples are illustrative.

Leave a Reply

Trending