Why approval chains get longer as businesses grow, not shorter

A £2,000 purchase order that once needed one signature now needs four. Nobody decided that on purpose, it happened one approval at a time, each one added for a good reason, until the process became something nobody would design from scratc

This is one of the quietest problems I see in scaling businesses, not a lack of governance, too much of the wrong kind.

Where the layers come from

Every approval step started life as a fix for something, a supplier was paid twice, so a second sign off was added to finance. A project ran over budget, so a review stage went into procurement. A decision caused friction between departments, so a committee was formed to prevent it happening again.

Each addition made sense at the time. The problem it solved was real and the fix was proportionate to the risk in that moment, what rarely happens is the removal.

The team that caused the original issue moves on, the systems change. The controls that made sense for a twenty person business stay in place at two hundred people, now applied to decisions that carry a fraction of the original risk.

Nobody reviews approval steps with the same energy they use to add them. So they accumulate, quietly, until a routine purchase needs input from four people who have no real stake in the outcome.

The cost that doesn't show up on a dashboard

Slow approvals rarely appear as a line item anywhere.

What shows up instead is a marketing campaign that launches two weeks late because the creative needed sign off from three departments, a supplier relationship that sours because payment terms slipped through no fault of the finance team, purely process delay, a talented manager who stops proposing improvements because getting anything approved takes longer than doing the work itself.

People adapt to slow systems in ways that hurt the business, they batch requests to avoid going through approval twice. They quietly work around processes rather than through them. They stop bringing forward the marginal improvements that, collected together, would have made a real difference.

None of this looks like a crisis. It looks like normal operational friction. That is exactly why it survives so long unchallenged.

The belief that doesn't hold up

There is a comfortable assumption behind most of this: more approvals mean more control. It rarely works that way in practice.

Long approval chains obscure accountability rather than sharpen it. When five people sign off on a decision, nobody feels fully responsible for it. Risk does not disappear because more people looked at it. It gets diluted across a group, which often means it gets missed rather than caught.

Long chains also change behaviour in the wrong direction. People start optimising for getting past the approval rather than for the quality of the decision itself. Compliance replaces judgement. That is a worse outcome than the one the approval was meant to prevent.

Genuine oversight and excessive process are not the same thing. Confusing the two is what allows bureaucracy to accumulate under the banner of good governance.

Designing approval that actually protects the business

The fix is not stripping out governance. It is being deliberate about where it earns its place.

Start by asking a simple question of every existing approval step: what specific risk or value does this provide, right now, at our current size? Not what it solved originally. What it is actually doing today. Some approvals will clearly still matter. Anything involving significant spend, legal exposure, or irreversible commitments deserves proper scrutiny. Keep those, and make sure the right people are the ones giving it.

Many others will not survive the question. They exist out of habit rather than need. Those are candidates for removal or simplification.

From there, build workflows around risk rather than treating every decision the same way. Routine, low value, repeatable decisions should move fast, often without a human in the loop at all. Exceptions, high value cases and anything genuinely unusual should get proper attention from people with the authority to give it.

Visibility does a lot of the work that approval used to do. If spending, project status and decisions are visible in real time, you catch problems through observation rather than through a queue of sign offs. That shifts oversight from a bottleneck to something closer to a live picture of the business, and it tends to catch issues faster than a chain of approvals ever did.

Where this leaves you

Approval chains rarely get long through bad intent. They get long through good intentions applied without review.

The businesses that stay agile as they scale are not the ones with the least governance. They are the ones who keep asking whether each control still earns its place.

Take one of your most common approval processes this week. Walk through every stage and ask what specific risk or value it protects against today. If nobody in the room can answer that clearly, you have found friction dressed up as oversight.

Fix that one process first. The pattern you use to simplify it will show you where to look next.

If you're rethinking how oversight and speed work together in your organisation, I'm glad to talk it through.