The theory of constraints, applied to a service firm

The theory of constraints, set out by Eliyahu Goldratt in The Goal, says that every system's output is limited by exactly one constraint at any given time — and that improving anything other than the constraint changes nothing. For a B2B service firm, the constraint is almost always one of four things: pipeline, conversion, capacity, or cash. Find which one is binding, fix that, and ignore the rest until it moves.

That is the whole theory. Its power lies in what it forbids: the diffuse, everywhere-at-once improvement effort that keeps founders busy and firms static. It slots into the wider toolkit I have covered in the systems-thinking guide for founders; this piece applies one idea from that toolkit properly.

Why does only one constraint matter?

Picture a chain. Its strength is the strength of its weakest link — not the average of all links, and not the sum. Strengthening any other link adds weight and cost while the chain holds exactly the same load. Goldratt's insight is that businesses are chains: enquiries flow to conversations, conversations to proposals, proposals to projects, projects to invoices. The flow through the whole chain is set by its narrowest point.

The corollary is uncomfortable. Most improvement work in most firms is applied to non-constraints, because non-constraints are more pleasant to work on. A founder whose real constraint is pipeline will happily redesign the website, refine the onboarding deck and re-organise the project board — everything except starting more conversations. The firm gets tidier and no bigger.

What are the four constraints of a service firm?

In a B2B service firm, the binding constraint is typically one of these:

• Pipeline. Not enough new conversations entering the system. Symptom: capable people with slack time, and a forward book that thins out beyond next month.
• Conversion. Conversations happen but don't become clients. Symptom: a busy diary of first calls, and a proposals folder full of silence.
• Capacity. Demand exceeds delivery. Symptom: lead times stretching, work queuing behind the same people, sales being softly discouraged because "we couldn't take it on anyway".
• Cash. The firm could sell and deliver more, but cannot fund the gap between doing the work and being paid for it.

Only one is binding at a time. A firm with a conversion constraint gains nothing from more leads — it just manufactures more silence. A firm with a capacity constraint gains nothing from better marketing — it just lengthens the queue.

How do I find the constraint?

The diagnostic question is: where does work queue? Constraints announce themselves by the pile of work waiting in front of them.

1. Trace the flow from stranger to paid invoice, step by step.
2. Look for the queue. Unanswered enquiries queue in front of a follow-up constraint. Un-sent proposals queue in front of a founder-time constraint. Signed clients waiting to start queue in front of a capacity constraint. If nothing queues anywhere and everyone has slack, the constraint is upstream of the whole system — it is pipeline.
3. Check who the queue waits for. If every queue in the firm waits for the same person, you have found something more specific than a stage — and in owner-led firms it is usually the owner. I have written that case up separately in the founder as the bottleneck; the warning signs are catalogued in five signs your growth depends on you.

What do I do once I have found it?

Goldratt's five focusing steps, translated for a service firm:

1. Identify the constraint (above).
2. Exploit it. Get the maximum from the constraint as it stands, before spending anything. If the constraint is founder selling-time, strip everything else out of the founder's week. If it is capacity, stop the constraint's hours leaking into work anyone else could do.
3. Subordinate everything else to it. The rest of the firm now runs at the constraint's pace and in the constraint's service. This is the counterintuitive step: it is correct for non-constraint people and processes to have slack. Efficiency everywhere is not the goal; flow through the constraint is.
4. Elevate it. Only now do you invest — hire, buy, systematise — to expand the constraint itself. When pipeline is the constraint, this is where a mechanism such as an outbound engine belongs: it adds inflow without consuming the founder's hours. Elevating before exploiting wastes money on a constraint you weren't even using fully.
5. Repeat. The constraint moves. Fix pipeline and the constraint typically migrates to conversion or capacity within a quarter. This is success, not failure — but it means the diagnosis must be re-run, not assumed.

One caution from the systems toolkit: constraints and feedback loops interact. A capacity constraint that throttles selling creates the classic revenue oscillation, so what looks like an intermittent pipeline problem is often a permanent capacity problem wearing a disguise. Diagnose from the queue, not from the symptom that shouts loudest this month.

The discipline is simple to state and hard to keep: one constraint, one fix, then look again. Everything else is tidying.

---

Keep the thinking coming: The Format — a fortnightly letter on business and personal systems. Subscribe by email.