Are Your Cabinet Quotes Still Profitable After the 2026 Wage Increase?
The National Minimum Wage rose to $26.44 an hour and award rates rose 4.75% from 1 July 2026. Adding that percentage to last year's quote does not protect your margin - recalculating the labour model behind it does.
The short answer
A wage increase does not tell you how much to charge. From the first full pay period on or after 1 July 2026 the National Minimum Wage is $26.44 per hour and modern award rates rose 4.75%. But a cabinet maker paid $32.00 an hour can cost close to $54 per productive hour once superannuation, workers compensation, leave and non-productive workshop time are counted - and a workshop quoting on 38 productive hours a week may only be getting 27. Rebuild the labour model from your own job data, then apply the increase to a rate you can defend.
A cabinet shop wins a kitchen at the same labour allowance it used last year. The quote looks competitive. The client accepts on the spot. The job runs, the cabinetry is good, the installer gets it done, and everybody moves on to the next one.
Twelve months of that and the business has a full order book, a busy factory and a bank balance that has not moved. Wages went up. Installation took longer than allowed. Two jobs needed return visits. A supplier delay cost a day of workshop time. None of it showed up anywhere, because the quote was never compared with what the job actually cost to deliver.
The 2026 wage increase makes that gap wider and more visible. It also gives every cabinet and joinery business a legitimate reason to do something most have been meaning to do for years.
So: when was the last time you rebuilt your labour model from actual job data, rather than adjusting last year's number?
This is operational guidance on pricing and job costing, not legal or industrial relations advice. Confirm the wage obligations that apply to your team with the Fair Work Ombudsman or your own adviser.
What changed in Australia on 1 July 2026
The Fair Work Commission handed down its 2026 Annual Wage Review decision in June, and the new rates applied from the first full pay period starting on or after 1 July 2026. That timing detail matters: the increase did not apply from 1 July itself unless a pay period happened to start that day.
| What changed | From 1 July 2026 |
|---|---|
| National Minimum Wage | $26.44 per hour, or $1,004.90 per 38-hour week |
| Casual under the National Minimum Wage | $33.05 per hour, including the 25% casual loading |
| Modern award minimum rates | Increased by 4.75%, including the Joinery and Building Trades Award 2020 |
| Superannuation guarantee | Unchanged at 12%, with no further legislated increases scheduled |
| Payday Super | Commenced 1 July 2026. Contributions now align with each payday rather than each quarter |
The two percentages are not the same number
This is the part that causes confusion in workshop conversations. The National Minimum Wage rose by roughly 5.97%, while minimum rates in modern awards rose by 4.75%. They are different figures applying to different populations of employees.
So “the wage increase” is not a single number a cabinet shop can apply across the board. Which figure is relevant depends on how each person on the floor is actually paid:
- An apprentice or tradesperson on an award classification rate moves by the award increase
- An employee on the National Minimum Wage moves to the new national figure
- An employee on an enterprise agreement moves according to that agreement, which must still not fall below the relevant award rate
- An employee on an over-award contract rate may not move automatically at all, but the rate must still clear the new award floor, and allowances, overtime and penalty rates calculated from award rates will have moved underneath it
Worth checking before anything else: which award, agreement or contract applies to each employee, and whether the rates actually paid clear the new minimum. Everything in this article is operational guidance on pricing, not legal or industrial relations advice. Confirm your obligations with the Fair Work Ombudsman or your own adviser.
The market you are quoting into also moved
Wages were not the only thing that shifted. ABS building approvals for July 2026 show total dwelling approvals fell 3.6% to 17,687 and the value of total residential building fell 4.9%, while approved non-residential building rose 14.4%.
For a cabinet or joinery business, that is not a story about volume growth or decline. It is a story about mix. Residential and commercial work carry different job sizes, different lead times, different site conditions and quite different installation labour profiles. A business drifting toward commercial work while still quoting on residential labour assumptions has a margin problem waiting to arrive.
Why adding 4.75% to your old quote is not enough
The instinct after a wage decision is to uplift the quote by the same percentage and move on. It feels responsible. It usually is not, because a percentage applied to a wrong number produces a wrong number with more decimal places.
There are three separate problems with the uplift approach.
1. The increase applies to part of the price, not all of it
If labour is 40% of a quoted job and the wage component rises 4.75%, the direct effect on the quote is about 1.9%, not 4.75%. Adding 4.75% across the whole quote overshoots. But materials, hardware, freight, insurance, power and software subscriptions move on their own schedules and are not covered by the wage decision at all, so adding 1.9% and stopping undershoots. Neither number is the answer, because neither was calculated from the actual cost base.
2. It preserves whatever error was already there
A shop quoting at $80 per hour when its true recovery requirement was $95 does not fix anything by moving to $83.80. It carries the same $15 shortfall into every job for another year, plus a little extra confidence that the rate has been “reviewed”.
3. It assumes the productive-hours ratio held
This is the big one, and it is the subject of the rest of this article. A labour rate is a fraction. The wage increase changes the numerator. Most margin erosion happens in the denominator - the number of genuinely productive hours the business gets from each hour it pays for.
| Number | What it actually means |
|---|---|
| Employee wage | The hourly rate on the payslip. The smallest of the five numbers, and the only one most people can quote from memory. |
| Fully loaded labour cost | Wage plus superannuation, leave loading, workers compensation, payroll tax where applicable, tools, PPE and training. |
| Productive labour cost | Fully loaded cost divided by hours actually spent on billable work, rather than by hours paid. Usually a large step up. |
| Charge-out rate | Productive labour cost plus workshop overhead recovery plus the target margin. What the business needs to bill per hour. |
| Final customer price | What the market and the competitive position will actually bear. Knowing the four numbers above is what tells you whether this one is a decision or an accident. |
A business that cannot separate those five numbers cannot tell the difference between a job it priced badly and a job it built badly. That distinction is the whole game.
Calculate the fully loaded cost of a cabinet-making employee
Below is a worked example. The numbers are illustrative and rounded - a hypothetical full-time cabinet maker on an over-award rate of $32.00 per hour, in a business below the payroll tax threshold. Substitute your own figures. The method is the point, not the totals.
Step 1: the annual cost of employment
| Cost | Basis | Annual |
|---|---|---|
| Base wages | $32.00/hr × 38 hrs × 52 weeks | $63,232 |
| Annual leave loading | 17.5% on 4 weeks, where it applies | $851 |
| Superannuation | 12% of qualifying earnings | $7,690 |
| Workers compensation | Illustrative 3% premium rate | $1,923 |
| Payroll tax | Only above the state threshold | Nil in this example |
| Tools, PPE, training, licences | Allowance | $1,500 |
| Total cost of employment | ~$75,200 |
The most common error in this table is counting leave twice. Annual leave, public holidays and personal leave are paid at the normal rate and are already inside the 52 weeks of wages. Adding them again as a separate cost line inflates the total. Their real effect is not on the cost - it is on the hours available, which is the next step. Leave loading is the exception: it genuinely is extra money.
Step 2: the hours you actually get
A full-time employee is paid for 1,976 hours a year. Nobody produces 1,976 hours of cabinetry.
| Hours | Detail | Annual |
|---|---|---|
| Paid hours | 38 hrs × 52 weeks | 1,976 |
| Less annual leave | 4 weeks | −152 |
| Less public holidays | Approximately 11 days | −84 |
| Less personal leave taken | Illustrative 5 days | −38 |
| Attended hours | Hours the person is actually at work | 1,702 |
| Less non-productive attended time | Illustrative 18%: cleaning, set-up, material handling, waiting, toolbox meetings, training, internal movement, rework | −306 |
| Productive hours | Hours genuinely available to be recovered in a quote | ~1,396 |
Step 3: the three rates this produces
- Cost per paid hour: $75,200 ÷ 1,976 = $38.06
- Cost per attended hour: $75,200 ÷ 1,702 = $44.18
- Cost per productive hour: $75,200 ÷ 1,396 = $53.87
An employee paid $32.00 per hour costs roughly $53.87 per productive hour - about 68% above the payslip rate. Put another way, 1,396 productive hours across 52 weeks is around 26.8 productive hours in a 38-hour week.
A workshop that builds its quotes on 38 hours is short by more than a quarter of the week, every week, for every person on the floor. No wage decision created that gap and no percentage uplift closes it.
Step 4: from cost to charge-out rate
The cost per productive hour is not the rate to quote. It still has to carry overheads and margin. Continuing the same illustration, with workshop overheads - rent, power, machinery, maintenance, insurance, software, administration and management time - recovered at $22 per productive hour and a 20% target margin:
- Cost per productive hour: $53.87
- Plus overhead recovery: $22.00, giving $75.87
- Divided by (1 − 0.20) for a 20% margin on price: $94.84 per hour
Note the margin arithmetic. A 20% margin means dividing by 0.8, not multiplying by 1.2. Multiplying $75.87 by 1.2 gives $91.04 and a margin of 16.7%, not 20%. It is a small mistake that quietly costs about 3 points of margin on every hour a business sells.
So the chain runs: $32.00 wage → $38.06 per paid hour → $53.87 per productive hour → $94.84 charge-out. Each step is defensible, and each step is invisible if the business only ever looks at the first number and the last one.
Include installation and site risk in the labour model
Most cabinet businesses estimate workshop labour with real care and then estimate installation as a round number. Two days. Three days. A week for a big one. Workshop hours are estimated per component; site hours are estimated per vibe.
That asymmetry exists because workshop conditions are controlled and site conditions are not. Which is exactly the argument for estimating site labour more carefully, not less.
Site labour that rarely appears in an estimate:
- Travel time, both ways, for every person on the crew
- Parking, loading, unloading and carrying to the work area
- Site access: stairs, lifts, narrow corridors, protection of finished surfaces
- Waiting on other trades, or on the site being handed over
- Site not ready: walls not straight, services not roughed in
- Variations agreed verbally on site and never priced
- Return visits for handles, adjustments, damage or late parts
- Rework caused by a workshop error or a transport knock
- Final clean, rubbish removal and defect walk-through
A worked illustration. A kitchen install is estimated at 16 hours. The lift is booked for the wrong day, the electrician has not finished, two panels need scribing that was not allowed for, and a return visit is needed for a late handle delivery. Actual: 27 hours. At the $94.84 charge-out rate above, that is roughly $1,043 of unrecovered labour on a single job - and if the shop runs forty kitchens a year with even half that variance, it is a five-figure hole that never appears as a line item anywhere.
The point is not that every one of those hours should have been charged to the client. Some of them were the business's own fault, and some were genuinely a variation. The point is that a job can look profitable at the moment the cabinetry leaves the factory and still lose money by the time the last return visit is done - and nobody will know unless the site hours are recorded against the job.
Installation hours are also the ones most sensitive to the residential-versus-commercial mix mentioned earlier. Commercial sites bring more trades, more waiting, more access control and more programme dependency. A labour allowance calibrated on domestic kitchens will not survive a shopfitting job unchanged.
Use estimated hours and actual hours together
Everything above is a calculation done once. This section is what turns it into something that stays true. The process is five steps and none of them are complicated:
- Estimate labour hours by stage when the job is quoted
- Record actual hours against the same stages as work happens
- Compare the variance when the job closes
- Identify the cause, not just the number
- Feed the finding back into the next estimate
The critical word is stages. A job that comes back “30 hours over” tells you almost nothing. The same job broken down tells you whether the estimator, the workshop or the site created the gap:
| Stage | Estimated | Actual | What the variance suggests |
|---|---|---|---|
| Measure and site check | 3 | 4 | Access or scope was more complex than assumed |
| Drafting and detailing | 6 | 9 | Late client changes, or an unresolved specification |
| CNC and cutting | 8 | 8 | Stable, as machine time usually is |
| Edgebanding | 5 | 7 | Tape changes not batched, or a setup problem |
| Assembly | 14 | 15 | Within tolerance |
| Finishing | 6 | 6 | Stable |
| Packing and delivery | 4 | 5 | Loading or access time not allowed for |
| Installation | 16 | 24 | Site readiness and access - the biggest single gap |
| Rework and return visits | 0 | 3 | Estimated at zero, which is never true |
Read that table as a business owner rather than an estimator. The CNC, assembly and finishing estimates are fine - leave them alone. Drafting and installation are where the money went, and rework was estimated at zero, which no workshop in Australia has ever achieved. Those are three specific, actionable findings. “We went 30 hours over” is none.
The other habit worth building is tracking the direction of variance over time. A stage that is consistently 20% over is an estimating assumption that needs correcting. A stage that swings wildly between jobs is a process problem, and repricing it will not help. It is the same diagnostic logic as finding the constraint in the workshop, described in how to increase workshop capacity without hiring.
Higher prices, better workflow, or both?
Not every margin problem is solved by charging more. Raising prices to cover an internal inefficiency makes the business less competitive while leaving the inefficiency in place. Before changing a rate, work out which of three causes is actually driving the variance.
| Cause | What it looks like | The right response |
|---|---|---|
| Pricing problem | The work was done efficiently and the scope was correct, but the hours genuinely required always exceed the hours quoted. The variance is consistent across jobs and across people. | Change the rate or the allowance. This is the case where a price rise is the honest answer. |
| Productivity problem | The same job takes different hours depending on who does it or what week it runs. Rework, waiting, searching for material and double handling show up in the variance. | Fix the workflow. Raising the price passes an internal cost to clients and hides the real issue. |
| Scope-control problem | The hours were spent on work that was never in the quote: verbal variations, site conditions, extras absorbed to keep someone happy. | Tighten inclusions, exclusions and variation handling. The rate was never the problem. |
Most businesses have some of all three, which is precisely why stage-level variance data matters. A consistent 20% overrun on drafting is a pricing problem. An erratic edgebanding time is a productivity problem. Eight extra hours on site that everyone remembers agreeing to but nobody wrote down is a scope problem.
Scope control in particular has a habit of being solved verbally and then losing its way between the quote and the site, which is the same failure described in what cabinet makers must clarify in every kitchen quote.
Build a labour review into your monthly process
A labour model calculated once a year, in the week after the wage decision, is a guess with good intentions. The businesses that hold margin treat it as a monthly management number, not an annual event.
A workable monthly review is eight lines long:
- Average estimated labour hours per job, by job type
- Average actual labour hours per job, by job type
- Margin by job type, quoted against achieved
- Rework hours as a percentage of total hours
- Installation variance, estimated against actual
- Overtime hours and what drove them
- The three jobs with the largest negative variance
- Supplier or material delays that cost labour hours
Two of those deserve a note. Rework hours are the single most useful number on the list and the one most often recorded as nothing, because rework is usually absorbed rather than logged. The three worst jobs matter more than the average, because averages hide the outliers that actually did the damage.
Then recalculate the fully loaded cost per productive hour at least annually, and immediately after anything that moves the inputs: a wage decision, a workers compensation premium change, crossing the payroll tax threshold, a significant change in overtime patterns, or a shift in the residential-to-commercial mix.
The practical test: if the labour rate lives in one spreadsheet, the actual hours live in a different system or a pile of dockets, and nobody has compared the two in the last quarter, then the business is not pricing from data. It is pricing from memory and hoping the memory is still current.
Profitability depends on the labour model behind the quote
The 2026 wage increase is a prompt, not the problem. It is a good reason to open the labour model, but the risk it exposed was there before July and will still be there next July. The practical response is short:
- Check which award, agreement or contract applies to each employee, and whether the rates paid clear the new minimum
- Recalculate the fully loaded cost of employment, without double-counting leave
- Separate paid hours from attended hours from genuinely productive hours
- Build the charge-out rate from cost per productive hour, plus overhead, plus margin - dividing by (1 − margin), not multiplying
- Estimate installation and rework as real hours, not a round number
- Compare estimated against actual by job stage, every month
- Diagnose each variance as pricing, productivity or scope before changing a price
The businesses most likely to protect their margin in 2026 are not necessarily the ones charging the most. They are the ones that can say what a job actually cost to deliver, and can show where the last estimate was wrong.
CabiPro connects quoting, job stages, production, materials and installation in one workflow, so the hours a job was estimated at and the hours it actually took sit against the same record. If your business is still calculating labour in one spreadsheet and tracking actual job time somewhere else, this is a good time to bring those numbers together.
Frequently asked questions
How much did wages increase in Australia on 1 July 2026?
From the first full pay period starting on or after 1 July 2026 the National Minimum Wage rose to $26.44 per hour, or $1,004.90 per 38-hour week, with the casual rate under the National Minimum Wage at $33.05 per hour. Minimum wages in modern awards, including the Joinery and Building Trades Award 2020, increased by 4.75%. The two figures differ because the National Minimum Wage itself rose by about 5.97% while award rates rose by 4.75%, so the increase that applies to a given employee depends on whether they are paid under an award, an agreement or a contract rate.
Can I just add 4.75% to my existing cabinet quotes?
Not reliably. A percentage uplift applied to an old charge-out rate preserves whatever error was already in it. The 4.75% applies to the award wage component, not to overheads, superannuation, workers compensation or non-productive time, and it does nothing about the assumption most likely to be wrong: how many productive hours the workshop actually gets from each paid hour. Recalculate the labour model, then apply the increase to a rate you can defend.
What is the difference between labour cost and charge-out rate?
Labour cost is what the employee costs the business per hour. Charge-out rate is what the business bills per hour. The gap between them has to cover workshop overheads such as rent, power, machinery, insurance, software and administration, plus the profit margin. A business that quotes at its labour cost is working for nothing, and a business that treats its charge-out rate as though it were its labour cost has no idea what its margin actually is.
How do I calculate the cost per productive hour?
Take the total annual cost of employing the person, including wages, leave loading, superannuation, workers compensation, payroll tax where applicable, tools, PPE and training. Then divide by the hours actually spent on billable production, not by the hours paid. Start from 1,976 paid hours for a full-time 38-hour employee, subtract annual leave, public holidays and personal leave to get attended hours, then subtract non-productive attended time such as cleaning, set-up, material handling, waiting, meetings and rework. The result is usually far fewer hours, and a far higher rate, than the paid-hours figure suggests.
Why do cabinet jobs lose money during installation?
Because installation labour is usually estimated as a round number while workshop labour is estimated in detail. Travel, parking and loading, site access, waiting on other trades, difficult stairs or lifts, incomplete site readiness, late variations and return visits are all real hours that rarely appear in the estimate. A job can show a healthy margin at the point the cabinetry leaves the factory and still lose money by the time the last return visit is finished.
Should I raise my prices or improve my workflow when margins fall?
It depends which of three causes is driving the variance. A pricing problem means the quote was too low for the work genuinely required, and the rate or allowance needs to change. A productivity problem means the hours were higher than they should have been because of rework, waiting or poor sequencing, and raising the price would only pass an internal inefficiency on to clients. A scope-control problem means the work performed was never in the original quote, and the fix is tighter inclusions, exclusions and variation handling. Comparing estimated hours with actual hours by job stage is what tells the three apart.
Sources and further reading
- Fair Work Ombudsman, Annual Wage Review 2026: fairwork.gov.au
- Fair Work Ombudsman, Joinery and Building Trades Award 2020 (MA000029): fairwork.gov.au
- Australian Bureau of Statistics, Building Approvals, Australia, July 2026: abs.gov.au
- Australian Taxation Office, Super guarantee: ato.gov.au
Sources checked on 14 September 2026. Wage rates, award entitlements, workers compensation premiums and payroll tax thresholds differ between states, territories and industries and change over time; confirm your own obligations with the Fair Work Ombudsman, your state revenue office and your workers compensation insurer.
Disclaimer: This article provides general information only and does not constitute legal, industrial relations, taxation or financial advice. All worked calculations are illustrative and use hypothetical figures. They should be rebuilt with your own business numbers before being relied on for pricing.
More from the shop floor
Find out what your jobs actually cost to deliver.
Bring one finished job to a CabiPro demo and see the estimated hours, actual hours, materials and installation time sit against the same record - so the next quote is built on what really happened.