How Many Direct Reports Is Too Many? Count the Decisions, Not the People

What's a good span of control?

Textbook says five to nine, and that comes from a 1930s read on how much attention one person can give another. Modern practice stretches it to eight or twelve for knowledge work, tighter where the work's dangerous, wider where it's standardised. Fine as a starting point. The trouble is it describes your reporting lines, your problem lives in your decision lines, and in most businesses those two diagrams don't sit on top of each other.

Take a multi-site retailer. Regional manager, fourteen stores. On paper that's a wide span and an obvious candidate for splitting. Watch her week before you do. Nine of those store managers run their sites, sort their own rostering, take their own markdown calls inside policy, and ring her maybe once a fortnight. Five ring her daily. Those five ring because nobody's ever told them where their authority stops, so asking is the cheapest way to stay safe.

Split the region and you've got two managers each carrying two or three of the daily callers. Same calls, smaller batches, more expensive desks, and a salary added to the cost line.

Why the optimal span of control is a symptom

How many people you can carry is set by how many of their decisions arrive with them. Widen the span while decisions die low and nothing breaks. Hold it at six while everything climbs and you've built a shorter ladder for the same traffic.

I ran a twelve month programme with a large American retailer, working with leaders stepping up from running a region to running a division. Hundreds of reports each, sometimes thousands, and the inbox that goes with it. The ask was fifteen to twenty minutes a week. Notebook, no laptop, sit somewhere nobody can find you, work out what actually matters.

Three answers came back. I'll make the time. I'll give it a go. I'm already flat out, so where am I finding twenty minutes.

We tracked all three groups across the twelve months and measured performance and potential at the end. The too-busy group came out lowest and a lot of them never finished, because they'd spent the year playing whack-a-mole on a hamster wheel. The ones who took the twenty minutes lifted their performance a long way, because their energy went where it mattered instead of drifting between meetings.

Nobody's span of control changed that year. What changed was where their attention went, and next to that the reporting count barely registered.

What happens when you add a management layer?

For the first six months, it gets worse.

Nobody's sure whose call anything is while the layer beds in, the informal routes that used to settle half your issues have been cut, and your managers turn cautious at exactly the moment you needed them bold, because deciding wrong in the shadow of a restructure feels career-limiting in a way it didn't the month before. So decisions climb. The senior team drowns in operational traffic and somebody concludes the structure still isn't right.

I've watched businesses run that loop three times in five years, working hard the whole way, wondering why the numbers never move.

A new layer's the right tool for a real accountability problem. Duplicated ownership, a missing role, a span that arithmetic says can't work. Reach for something else when the diagnosis is escalations.

I spent six months with a CEO and his leadership team where the problem looked structural and wasn't. He was carrying twelve things at once, and his team couldn't read what mattered week to week, so everything came back to him for a steer. Where's the focus today, it feels like he keeps changing his mind, we can't finish anything because we keep starting things. We changed one thing. One project a week for the whole leadership team with the resources behind it, and anyone not on that week's project got the week back for their own unit. Six months on, every one of them was achieving more in their own business unit than before, absenteeism was down and profitability was up. Nobody moved on the org chart.

How to test your real span of control in two weeks

Log every decision that reaches you from your direct reports and one level below. Next to each, write whose call it should've been under your own policy, and why it didn't stop there. The reasons sort into a short list. Nobody owns it, or somebody owns it and has never been told, or somebody owns it and doesn't believe owning it is safe. Each one's got a different fix, and not one of them is structural.

That list will tell you more than any headcount ratio.

So, how many direct reports is too many? One more than the number at which decisions start landing on you that were never yours to make. Count for a fortnight and you'll have your number.

Frequently asked questions

How many direct reports should a manager have?

Five to nine by the textbook, eight to twelve in most knowledge work. Treat that as a starting position, then check it against how many decisions belonging to those reports reach the manager anyway.

Is 15 direct reports too many?

Fifteen works fine when those fifteen settle their own recurring decisions. It falls over fast when they don't, and at that point adding a layer relocates your escalations rather than getting rid of them.

What are the signs your span of control is too wide?

Decisions queue behind one calendar. The same exceptions climb to the same desk every week. The manager's own work starts after seven in the evening, and her people have started pre-clearing things she never asked to see.

Should I add a management layer to reduce escalations?

No. A new layer pushes escalation volume up for the first six months while ownership resettles. Add a layer for duplicated ownership or a missing role, and change decision rights when the problem is escalations.

How do you reduce escalations without restructuring?

Take the fifteen to twenty decisions that keep coming round, name an owner for each at the lowest sensible level, and make peers settle sideways before anything travels up. It's unglamorous and it works, and it doesn't need a single line of the org chart to move.

Jimmy Burroughes, JBL High Performance. Simplify to Amplify.

If any of this sounds familiar, here's where to start

1. Not sure where the friction is building in your management layer? → See where decisions stall, escalations pile up, and where good people burn out
2. Wondering if you've become the bottleneck yourself? → Take the free 5-minute assessment