Capacity planning before you turn work away or overcommit

5 days ago
3 min read

Updated: 2 days ago

Introduction

Two failures follow from not knowing your capacity, and most businesses commit both in the same quarter. They decline work they could have delivered, and they accept work they cannot, discovering the second one around the third week of the job.

The reason is that capacity feels like a single number and is not. It is whichever resource runs out first, and that resource is frequently not the obvious one. A workshop assumes it is machine hours when it is actually the one person who can programme the machine. A consultancy assumes it is consultant days when it is the partner who has to review every deliverable. Identifying the real constraint is the whole exercise.

1. Capacity planning before you turn work away means finding the constraint

One resource, not an average.

Work through your process and ask which step has the least spare capacity. Everything else has slack by definition. Plan against the constraint and the rest of the schedule resolves itself.

2. Expect the constraint to be a person

Usually a specific one.

The estimator, the qualified engineer, the person who signs off, the owner. Skills concentrated in one individual are the most common real limit in small businesses, and they are the least visible because that person absorbs the pressure quietly.

3. Measure available hours honestly

Not contracted hours.

Subtract holiday, sickness, training, travel, admin, meetings and the interruptions that fill a working day. Realistic available time is frequently sixty to seventy per cent of contracted time, and planning at ninety guarantees failure.

4. Convert your pipeline into constraint hours

The comparison that matters.

For each quote and each live job, how many hours of the constraint resource does it consume? Summed by week, against available hours, this is the only forecast that tells you whether you can say yes.

5. Plan at eighty-five per cent, not a hundred

Utilisation and reliability trade against each other.

A schedule planned to full capacity has no absorption for illness, overrun or an urgent job, so it fails weekly. Eighty to eighty-five per cent produces better delivery performance and, usually, more completed work in the year.

6. Know your options before you need them

Decide in the quiet period.

Overtime, subcontracting, a temporary hire, deferring internal work, renting equipment. Each has a cost and a lead time. Knowing them in advance turns a difficult decision under pressure into a straightforward one.

7. Say no properly rather than late

The expensive mistake.

Declining at the enquiry stage costs an enquiry. Accepting and then failing costs the customer, your reputation and the other jobs it disrupted. A clear no with an alternative date is respected; a missed programme is not.

8. Watch for the constraint moving

It relocates as soon as you relieve it.

Add a second estimator and the limit becomes the installation team. This is normal and it means the constraint has to be re-identified periodically rather than assumed once and planned around for years.

9. Track a small number of honest measures

Enough to see the picture.

Constraint utilisation, delivery-to-promise performance, and the value of work declined. The third is the one nobody records and the one that tells you whether the constraint is worth investing in relieving.

Be careful about relieving a constraint by hiring during a busy period. Recruitment takes months, the busy period may not persist, and the new fixed cost stays after the work has gone.

Conclusion

Identify the resource that runs out first, because that is your capacity.

Expect it to be a specific person rather than equipment, measure realistically available hours rather than contracted ones, convert your pipeline into hours of the constraint resource, plan to about eighty-five per cent so the schedule survives normal disruption, decide the cost and lead time of your expansion options in advance, decline early and clearly rather than failing late, re-identify the constraint after you relieve it, and record the value of work you declined.