How Cabinet Makers Can Increase Workshop Capacity Without Hiring More Staff

Why hiring is not the default answer in 2026

Most Australian cabinet makers who feel over capacity in 2026 are not short of work. The Housing Industry Association reported a median kitchen renovation of around $35,000 in its 2025 Kitchens and Bathrooms report and expects renovation spending to strengthen through 2026. The order book is fine. The problem is getting jobs through the shed.

The obvious fix is another pair of hands, and for many workshops that door is closed. Jobs and Skills Australia counts roughly 30,500 cabinetmakers nationally, most of them already working full-time hours. Infrastructure Australia's 2025 market capacity report projects a shortfall of more than 300,000 construction workers by mid-2027, and apprentice completions lag starts by three to four years, so the people who signed up this year will not ease anything before 2027 or 2028.

Even when a hire is available, the wage is only the visible part of the cost.

A first-year apprentice costs in the order of $35,000 a year before incentives, and a qualified tradesperson considerably more, before they have produced a single carcass. That spend is only justified once the existing team is working on productive work most of the day. In most workshops we see, it is not, and that is the opportunity.

The rest of this guide is about recovering that capacity: finding the step that limits the whole workshop, keeping it running, and removing the rework, waiting and paperwork that eat hours nobody scheduled. Our earlier piece on protecting margins in the 2026 joinery squeeze covers the cost side; this one is about output.

Capacity is throughput at the bottleneck, not headcount

A workshop's capacity is not the sum of everyone's hours. It is the rate of its slowest step. Eliyahu Goldratt made that point in The Goal in 1984 and it has not stopped being true: at any moment one operation limits how many jobs leave the building, and improving anything else adds work-in-progress, not output.

In a cabinet shop the constraint moves with the job mix. A week of two-pack kitchens puts it in the spray booth. A run of laminate wardrobes puts it on the edgebander. A shop that has just bought a nesting CNC often finds the router is idle while someone finishes the drawings and programs, because the constraint has moved upstream into design. Whatever it is this month, three rules apply.

• An hour lost at the constraint is an hour lost for the whole workshop. It cannot be made up later by a step that has spare time.
• An hour saved anywhere else is a mirage unless it frees the constraint or removes work from it.
• Releasing more work does not create more output. It creates queues, lost parts and jobs started but not finished.

Finding your constraint

You do not need software to find it. Walk the floor at 10 am and look for where cut parts, trolleys and job folders are waiting. The step with the longest queue in front of it and idle capacity behind it is the candidate. Then confirm it with a five-day block log: a clipboard at that machine or bench where the operator writes down every stoppage, how long it lasted and the cause. Four causes cover most entries: missing information, missing material, a quality problem, or a tooling or setup issue.

Write those four numbers on the whiteboard every Friday. Nothing in this article works without a baseline to compare against.

Stop feeding rework into the schedule

Rework is the cheapest capacity to recover because it is work you have already paid for once and are paying for again. The US National Institute of Standards and Technology puts scrap and rework at 5% to 30% of total manufacturing cost across industry, and published research attributes up to 80% of quality defects to human error. Web-CAB, a woodworking software vendor, estimates the figure for cabinetry at 20% to 40% because a mis-machined panel cannot be melted down and reused. Treat that last number as a vendor estimate, but the direction is not in doubt.

The visible cost is the sheet and the hour. The expensive part is the disruption: the schedule gets reshuffled, the installer waits on site, the person who should be programming the next job is on the phone working out what went wrong. None of that appears on a job cost report, and all of it comes out of throughput.

The pattern is the same in every row: rework happens when an assumption stands in for information. The fix is a checkpoint, not a better tradesperson. If your rework hours are above 8% to 10% of production hours, this section alone is worth more capacity than any machine you could buy.

Only release complete jobs to the floor

Manufacturers call it kitting. A job is released to production only when everything it needs is on hand, checked and staged, so the person who opens the job packet can start work rather than start searching. It sounds obvious. Most small workshops do the opposite: they start the job to keep people busy and sort the rest out as they go, which is how a machine ends up waiting for a box of hinges that is still on a supplier's truck.

Before a job goes on the production board, someone confirms:

• Drawings are at the latest revision and every question from the job review is answered.
• Board, edge tape and benchtop material are physically in the rack, not just ordered.
• Hardware is picked into a labelled tub per job, with quantities counted against the cut list.
• The cut list or nest is generated from the approved drawing, not retyped from the quote.
• The site is ready and the install date is confirmed, so finished cabinets do not sit in the way.

The rule that makes this stick is simple: if the kit is not complete, the job does not start, and the reason is written down. That turns an unexplained Tuesday afternoon stoppage into a kitting exception with a cause you can act on. After a month you will know whether your real problem is supplier lead times, drawings arriving late, or purchasing that happens after production starts.

Kitting also fixes the second-most common capacity leak: jobs that are 90% complete. A carcass waiting on one door or one panel occupies floor space, gets damaged, and is handled three more times before install. Complete jobs leave; partial jobs linger.

Protect the constraint machine

Once you know which machine or bench is the constraint, the goal is brutally simple: it should never wait, and it should never do work that something else could do. Across industry a typical machine runs at 40% to 60% overall equipment effectiveness, and only a handful of plants reach the 85% considered world class. A small shop does not need an OEE meter to know its edgebander spends most of the shift not edging.

An illustrative day on the edgebander

Those numbers are made up to show the shape, but most operators will recognise them. Recover one hour a day and the constraint goes from 3.5 to 4.5 productive hours, which is roughly 28% more edgebanding capacity with the same machine and the same person. Here is where that hour comes from.

• Pre-stage the next job. Cut parts for the next job should be on a trolley beside the machine before the current one finishes. The block log from earlier tells you how often it is waiting.
• Batch by tape, not by job. Swapping tape of the same thickness takes under a minute; changing thickness or material and re-setting the machine takes ten to fifteen. Run every white interior part from three jobs in one pass, then every coloured exterior part, and you trade a dozen changeovers for two.
• Take non-constraint work off the operator.An apprentice or the person with the least loaded bench does the offloading, stacking and trolley runs. The operator's hands stay on the machine.
• Run it through breaks. A relief operator for smoko and lunch adds up to an hour a day of machine time. Nowhere else in the shop is that trade worth making.
• Ten minutes of daily care. Glue pot, pressure rollers, scraper blades and dust extraction checked every morning by the operator. Fifteen minutes on a Monday is cheaper than a Thursday afternoon breakdown.

Then do the same exercise at the next constraint, because as soon as the edgebander keeps up, something else will not. That is not a failure. It is how you know the last change worked.

Standardise the construction, then let the library do the thinking

Custom joinery does not mean every carcass is a fresh decision. The customer cares about the doors, the finish, the layout and the fit. They do not care whether the back is 12 mm or 16 mm, where the shelf pin holes start, or how the kickboard is fixed, and neither should your team have to decide it job by job.

A written construction standard and a cabinet library built on it remove hundreds of small decisions from every project. The person drawing the job picks a base cabinet with a drawer stack and the width; the construction rules produce the parts, the machining, the hardware and the edge tape. That does three things for capacity.

• Design stops being the constraint. Once a CNC is in the shop, programming and drawing is usually the slowest step. A library turns a day of drawing into an hour of configuring.
• Rework drops. A rule that is right once is right every time. A rule that is wrong gets fixed once for every future job.
• The quote becomes the cut list. If the quote was built from the same components, nobody retypes dimensions from a PDF into the nesting software. Our guide to parametric quoting explains how those rules work on the estimating side.

Start with the ten cabinets that make up 80% of what you build. Agree the construction, write it down, and make it the default. Keep genuine one-offs as one-offs; they will be a much smaller share of the week than they feel like today.

Outsource the bottleneck component, not the craft

Woodworking Network has argued for years that buying in doors, drawer boxes and even carcasses can double a shop's capacity without new machinery or new staff, and that the main obstacle is pride rather than economics. Both points hold in Australia. Vinyl wrap and two-pack door makers, drawer box suppliers and benchtop fabricators run dedicated lines that do one thing faster and cheaper than a general workshop can, and they scale with your order book instead of your payroll.

The mistake is outsourcing by habit or by price alone. Outsource the component that is your constraint. If the spray booth holds up every job, buy in finished doors and keep making carcasses. If the edgebander is the constraint and you build your own drawer boxes, drawer boxes are the first thing to buy in. Use a scorecard rather than a gut feel.

When the constraint moves to sales, which is what happens when production stops being the limit, you have a different and much better problem. Keep the craft that defines your work. Buy the parts that only consume capacity.

Take the admin off the tools

In a workshop of four to eight people the owner or leading hand is usually the best tradesperson in the building and the one spending the least time on the tools. Quoting, retyping the accepted quote into a cut list, chasing suppliers, hunting for the current drawing, and answering "where is my job up to?" calls can consume two or three days a week. That is a skilled cabinetmaker's capacity, and it costs nothing to recover except a change of system.

The tell-tale signs are familiar:

• The same job details entered into a quote, a spreadsheet, the nesting software and an invoice
• Materials ordered after production starts, because nobody knew what was short until the saw stopped
• Drawings in email, on a USB stick and printed on the bench, each a different revision
• Installers leaving without the hardware list, the site contact or the variation the client agreed to last week
• Job costs reconstructed months later from invoices, if at all

A single job record that carries the quote, the approved drawings, the materials to order, the purchase orders, the production stage and the install date fixes most of this. Whether that record lives in dedicated software or a disciplined folder structure matters less than the rule that there is only one of it. We compared the two approaches in cabinet maker software versus spreadsheets, and our guide to managing multiple cabinet projects covers the production board that sits on top of it.

The capacity test is straightforward. Count the hours the leading hand spent on the tools this week. If it is under half, the cheapest machine you can buy is the one that gives those hours back.

A 30-day capacity plan

None of this needs a consultant or a capital budget. It needs one person to own it for a month and a whiteboard. The structure below borrows from the throughput programs used by larger shops, scaled to a workshop where the owner is also the estimator.

A frequently cited case study from the US National Institute of Standards and Technology's manufacturing program reports a shop halving its lead time and more than doubling daily output through exactly this kind of constraint-focused work. Your first cycle will not look like that, and it does not need to. A 10% lift in jobs out the door with the same crew is a hire you did not have to make.

Where CabiPro fits

CabiPro does not run your edgebander or measure machine utilisation. What it does is remove the re-entry and chasing around each job that the previous sections describe. A quote built from configured components becomes the job; the job produces the materials to order and the purchase orders; the production schedule, install date and invoice sit on the same record. Our cabinet maker project management software page shows the workflow.

The capacity that gives back is the leading hand's afternoon, the job that was not released with missing hardware, and the panel that was not cut from last month's drawing. It will not find your constraint for you. It will make sure that when you do, the constraint is not waiting on paperwork.

If you want to test the idea, bring one live job to a demo and count how many times its details are currently typed in.

Frequently asked questions