Your Service Business Has Inventory. It Just Doesn’t Sit in a Warehouse.

Walk through a factory carrying too much inventory and the problem is difficult to miss.

Parts occupy shelves.

Half-finished products wait between processes.

Finished goods take up space while waiting for someone to buy them.

There is weight to the problem.

Now walk through a service business carrying too much unfinished work.

Nothing appears to be there.

The office looks the same.

The laptops take up no more space.

The project-management software can hold another thousand tasks without anyone needing a larger building.

And yet the business may be full.

Not full of physical stock.

Full of unfinished obligations.

A proposal waiting for approval.

A project that has started but cannot move because one decision remains open.

A client change that has been accepted verbally but not priced.

Creative work waiting for feedback.

An invoice waiting for someone to confirm whether the extra work was actually billable.

A vendor waiting on information from the client.

A founder carrying seven unresolved decisions in their head.

None of these appear on a warehouse floor.

But operationally, they behave surprisingly like inventory.

They occupy capacity.

They age.

They create uncertainty.

They require people to remember where they are.

They eventually have to move, change, be written off or be completed.

And the longer they remain inside the system, the more expensive they can become.

Manufacturing makes unfinished work visible

One of the ideas at the heart of the Toyota Production System is deceptively simple.

Toyota describes Just-in-Time as making only what is needed, when it is needed, in the amount needed—and specifically says that goods and information should not be allowed to become held up during production.

In a physical production environment, excess work-in-progress is visible.

You can count it.

You can see where it is accumulating.

You can watch one process continue producing while the next process is already overloaded.

That visibility changes the conversation.

The unfinished work cannot pretend it does not exist.

Service businesses have a harder problem.

Their work has almost no physical mass.

Lean practitioners have long recognised this. The Lean Enterprise Institute defines inventory broadly enough to include materials and information sitting between processing steps, and has specifically noted how difficult excess work becomes to see in office environments because digital and knowledge work occupies almost no physical space.

So the idea that knowledge work has WIP is not new.

But service businesses often underestimate something more specific.

Their inventory is not merely unfinished tasks.

Very often, it is unfinished commitments.

And those are commercially more dangerous.

An open client promise is inventory

Imagine an agency wins a new account.

The scope is agreed.

Work begins.

Then the client asks for something adjacent.

It is not quite part of the original scope, but it is small enough that nobody wants to create friction.

The team starts working on it.

Nobody formally says no.

Nobody formally says yes.

Nobody changes the fee.

The additional work now exists inside the delivery system.

What is it?

It is not quite an approved deliverable.

It is not quite out of scope.

It is not yet an invoice.

It is not a dispute.

It is simply unfinished commercial reality.

And someone is carrying it.

Perhaps the account manager is remembering that the client owes an approval.

Perhaps the delivery lead is trying to accommodate it without moving the timeline.

Perhaps finance does not know it exists.

Perhaps the founder will discover it three weeks later when the client wants another change.

That unresolved commitment is effectively inventory.

The business has consumed capacity before completing the commercial decision attached to the work.

Do this once and almost nothing happens.

Do it across twenty accounts and the delivery organisation starts carrying a warehouse nobody can see.

Digital systems can hide how full the warehouse has become

Physical inventory eventually gets in the way.

Digital inventory does not.

That is one reason it can become dangerous.

A project-management tool does not complain when another task is added.

Slack will accept another unresolved thread.

Email will happily preserve another unanswered request.

A spreadsheet can add another line.

The storage cost is almost zero.

The operating cost is not.

Every unfinished item carries some combination of:

context that someone must remember;

a decision that will eventually have to be made;

a person who cannot fully close the work;

a customer waiting for something;

a commercial question whose answer becomes harder as time passes;

or an exception capable of resurfacing at the worst possible moment.

The problem is not simply that people have “too much to do.”

It is that too much work has entered the system without leaving it.

That distinction matters.

Because the usual response to excessive workload is to add capacity.

Hire another person.

Add a freelancer.

Create another project-management layer.

Move responsibilities around.

But if the underlying problem is uncontrolled work-in-progress, more capacity may simply allow the organisation to start even more things simultaneously.

The warehouse gets larger.

Starting work feels productive. Finishing work creates capacity.

This is one of the stranger incentives inside service organisations.

Starting is visible.

A request comes in and somebody responds.

A brief arrives and somebody begins.

A founder asks for something and the team picks it up.

A client wants a change and the account lead gets moving.

Responsiveness looks like progress.

Finishing is less dramatic.

Closing the outstanding approval.

Removing obsolete work.

Deciding not to pursue something.

Getting the client to approve the change.

Completing the handoff to billing.

Resolving the exception so nobody needs to remember it again.

These actions often feel administrative.

But they return capacity to the organisation.

There is a useful operating question here:

How much of the organisation’s apparent workload consists of work being performed—and how much consists of work being carried?

Those are different things.

Work being performed consumes effort.

Work being carried consumes attention.

And attention is often the scarcer resource.

The founder frequently becomes the overflow warehouse

This becomes especially visible in founder-led businesses.

When ownership is unclear, unresolved work often travels upward.

A client exception reaches the founder.

A pricing question reaches the founder.

A hiring decision waits for the founder.

A vendor problem waits.

A proposal sits in their inbox.

Two departments disagree and the issue waits until the founder joins the conversation.

The organisation may describe this as a founder bottleneck.

But sometimes the founder is not actually creating the bottleneck.

They are simply the place where the organisation stores everything it has failed to resolve elsewhere.

They become the overflow warehouse for ambiguity.

That distinction changes the diagnosis.

Telling the founder to “delegate more” does little if the work simply returns whenever the established process reaches an exception.

The real question is where unresolved work is allowed to accumulate before that happens.

Follow the queues

To understand a service business quickly, do not begin by counting meetings or reading SOPs.

Look for the queues.

Where is work waiting?

Where are clients waiting?

Where are employees waiting for decisions?

Where are invoices waiting for confirmation?

Where are proposals waiting for internal approval?

Where is completed work waiting for the next function to accept it?

Where are changes waiting to become either authorised work or a polite no?

The queues reveal something the org chart often cannot.

They show where the operating system stops flowing.

Sometimes the constraint is capacity.

Sometimes it is capability.

Sometimes it is unclear authority.

Sometimes the process itself is unnecessarily complicated.

And sometimes there is a perfectly rational reason for maintaining a buffer.

That last point matters.

Zero inventory is not the objective

It would be easy to turn this argument into another simplistic operating doctrine:

Eliminate WIP.

Finish everything immediately.

Keep no queues.

That would be wrong.

Even lean systems deliberately use appropriate buffers where variation or upstream capability requires them. Lean guidance explicitly cautions against simply lowering inventory without first addressing variability and process capability because doing so can damage the ability to serve the customer.

The same applies to services.

A creative team may intentionally maintain several active projects because client feedback arrives unpredictably.

A law firm cannot complete every matter sequentially.

An agency needs enough active work to accommodate pauses outside its control.

A leadership team needs some decisions to mature rather than forcing immediate closure.

The issue is not the existence of work-in-progress.

It is uncontrolled work-in-progress.

Work whose owner is unclear.

Work whose commercial status is unclear.

Work whose next decision has no deadline.

Work that remains technically “active” long after anyone intends to finish it.

Work that cannot be closed because somebody upstream never completed their part.

That is the inventory worth finding.

The four questions behind every unfinished obligation

A useful discipline is to take a sample of open work and ask four things:

Who owns its next movement?

Not who owns the project overall. Who is responsible for moving this specific item forward?

What is it waiting for?

A client response? Internal approval? Information? Capacity? Judgment?

When does waiting become an exception?

If there is no threshold, work can age indefinitely while still appearing normal.

What closes it?

Completion? Approval? Rejection? Billing? Cancellation? A formal decision not to proceed?

If those answers are difficult to obtain, the work is probably costing the organisation more than the task itself suggests.

Because somebody is carrying the ambiguity.

Capacity sometimes hides inside unfinished work

Growing businesses understandably look outward when capacity becomes tight.

Who should we hire?

Which tool should we buy?

What can we automate?

Should we restructure the team?

Those can all be correct questions.

But before adding capacity, there is another question worth asking:

How much capacity would return if the organisation became better at finishing, closing and refusing work already inside the system?

That is not an argument against growth.

It is an argument for understanding the load before increasing the size of the machine carrying it.

Toyota’s Just-in-Time principle is often summarised around inventory reduction.

The more interesting idea is flow.

Toyota itself frames the system around synchronising processes, preventing goods and information from being held up, and shortening the time between demand and fulfilment.

A service business should care about the same thing.

Not because an agency should pretend to be a factory.

But because every organisation eventually pays for work that enters faster than it can leave.

In manufacturing, the evidence may be sitting on a pallet.

In a service business, it may be sitting in someone's inbox, project board—or head.

The absence of a warehouse does not mean there is no inventory.

Sometimes it only means the organisation has become very good at hiding it.