Sub quote or own crew: 4 relevant cost checks

Key takeaways

• Remove costs you will pay either way.
• Check which fixed costs would actually stop, and when.
• Compare totals for matching scope and volume.
• Test uncertain inputs before choosing.

Why the lowest number is not always the cheapest

A sub quote lands on your desk looking clean. It is one number, one unit price, one line on a page. Your own crew is harder to read. It shows up as wages, equipment, materials, insurance, and a share of overhead, and nobody has sorted those into one tidy figure. So the natural move is to divide everything by the units, compare the result to the quote, and pick the smaller number.

That shortcut goes wrong more often than most owners expect. A sub's unit price is a complete price, but your internal unit cost is usually a mix of two kinds of cost. Some costs will be there whatever you decide. Others will disappear if someone else does the work. Only the second kind should drive the decision. In managerial accounting, costs that differ between the options are called relevant costs, and the unit cost that matters is the one built from them.

A textbook case from OpenStax shows the problem in a clean form. The business in the example, Thermal, needs lids for one of its carriers, and an outside supplier, Plato Plastics, offers to make them. The offer is specific: "Plato Plastics has approached Thermal and offered to produce the 120,000 lids Thermal will require for current production levels of the T6 carrier, at a unit price of $1.75 each." ( OpenStax)

At first glance the outside offer looks obviously better. The text says so directly: "Initially, the $1.75 presented by Plato seems like a much better price than the $2.19 that it would cost Thermal to produce the lids." ( OpenStax)

That first impression is the trap the four checks are built to catch. Work through them in order, and keep a running table as you go, because each check changes the numbers.

Check 1: Strip out the costs that do not change

Start with the full cost of doing the work yourself. Then go line by line and ask one question of each cost: if we stopped doing this work in-house, would this cost stop too? If the answer is no, the cost is not relevant to this decision, no matter how large it looks on the page.

Typical examples from a contractor's books are the overhead on the office, the salary of a supervisor you keep on staff, a yard lease you will pay either way, and the insurance on the business as a whole. These costs get spread across every job as a unit charge, which makes them look like part of the job. They are not. They are still there after the sub signs on, so they do not belong in the comparison.

Crew wages need the same test. They belong in the comparison only if the hours would really drop when the work goes out. A crew you would keep on the payroll and redeploy anyway is a fixed cost, and it should be treated like one. Be honest about this. Many owners keep paying a crew through slow weeks because they want to keep the people, and that choice is a real one, but it is a separate decision from the sub-or-crew question. For a closer look at paid time, see crew hours and overtime costs.

The OpenStax case does this step explicitly. It separates out the fixed part of the internal cost: the fixed costs of $0.51 for each unit are unavoidable and therefore are not relevant. ( OpenStax) Once those costs are set aside, the internal option looks much cheaper. The text runs the totals and finds that "it is more cost effective for Thermal to produce the 120,000 units of the T6 lid internally than to purchase it from Plato." ( OpenStax) In dollars: "By producing the T6 lid internally, Thermal can save $8,400 ($210,000 − $201,600)." ( OpenStax)

Notice what happened. The first comparison, using the full $2.19 cost, pointed toward buying. Removing the unavoidable fixed cost flipped it toward making the lids. That is the first lesson of this method: a full-cost comparison and a relevant-cost comparison can point in opposite directions, and the relevant-cost one is the right basis for the decision.

But the answer is not final yet. Check 2 can flip it back.

Check 2: Find the fixed costs you could actually avoid

Check 1 removed fixed costs that stay no matter what. Check 2 looks at the fixed costs that might go away if you subcontract. Some do. Interest on equipment you could sell, insurance on a vehicle you could return, a lease you could end, or a rental contract that would stop on the day the work goes out may all fall into this group. The test is not whether a cost is labeled fixed. The test is whether it would actually stop.

Be careful here, because the label can mislead in both directions. A fixed cost that stays on the books after you sub out the work is not avoidable, even if it is painful. A fixed cost you can cancel, sell, or release is avoidable, even if it looks like overhead. Check each one against the contracts and assets you actually have.

The OpenStax case makes this point with one change to the facts: suppose that, of the $0.51 in fixed costs for each unit of the T6 lid, $0.12 of those fixed costs are associated with interest costs and insurance expenses and thus would be avoidable if the T6 lid is purchased externally rather than produced internally. ( OpenStax)

When that $0.12 is counted as a cost of making, the answer reverses: "it is more cost effective for Thermal to buy the T6 lid from Plato, as Thermal would save $6,000 ($216,000 − $210,000)." ( OpenStax)

Look at how small the change was. A portion of the fixed cost that was previously excluded is now included, and the decision moved from making to buying. The gap that drove the result is only $0.12 for each unit, but multiplied across the full volume it decides the outcome. That is why you cannot stop at a unit-price comparison.

For each avoidable item, ask two things. Would the cost really go away once the sub starts? And when? A lease that runs past the start date is still a cost until it ends. Equipment you can sell only after a long process still ties up money in the meantime. An avoidable cost that stops on a later date is only avoidable for the period after that date, and the comparison should reflect the timing.

Check 3: Compare the totals across the full volume

A unit price is only a comparison when the volume is the same on both sides, and even then the total is what you will actually pay or spend. The sub quote covers a specific quantity. Your in-house cost covers the same quantity. Compare the totals, using the volume the job really needs.

The OpenStax case ties its numbers to a stated volume. The offer covers 120,000 lids, and the comparison is made on "the 120,000 units of the T6 lid." Each relevant unit cost gets multiplied by that volume to produce the totals that decide the case. If your volume is different, your totals are different, and the answer may be too. For help checking the calculation, see lot cost to unit cost.

Volume is not always a simple multiplier. Some fixed costs step up or down with volume. A second shift, a larger machine, or an extra inspector may only appear above a certain level of work. So a volume change can shift which costs are relevant as well as how big they are. Ask whether the cost behaves in a straight line with volume or in jumps, and build the totals accordingly.

The quote itself needs the same scrutiny. The OpenStax offer is priced "for current production levels," which means the price is tied to a volume. If your volume moves, the quote may not hold, and a new quote may come in at a different unit price. Before you compare, confirm the following on the sub's side of the comparison:

• Quantity and specification match your own scope exactly.
• Delivery terms, freight, and the schedule are included or costed separately.
• Setup, cleanup, testing, and any required inspections are in the price or on your side of the comparison.
• The quote's validity period is long enough for you to make the decision and start the work.

A lower unit price that leaves out freight or cleanup is not a like-for-like comparison. Add the missing items to the sub's side, then compare the totals.

A simple two-column table makes this easier to check. Put the in-house relevant costs in one column and the sub's total in the other, both at the same volume. Label every row, so that anyone reviewing the numbers can see which costs were kept, which were removed, and why. Most arguments about a sub quote are really arguments about a missing row, and the table makes those rows visible before the decision is made.

Check 4: Test the answer against the numbers that could move

A decision that turns on a small unit cost item is fragile. Before you commit, rerun the comparison with the inputs that are least certain. Typical inputs include your labor rate, your productivity on this kind of work, the sub's price if the scope grows, the sub's price in the next quote, your volume, delivery delays, and the share of fixed costs you can really cancel.

Ask two questions. First: how far would the sub price have to move before the decision flips? If a small price increase flips it back, the margin is thin and the decision is less settled than the first table suggested. Second: how far would your own cost have to move? If your crew's productivity dropped a little, or a fixed cost came in higher than expected, would you still make the same choice?

In the OpenStax case, the switch between making and buying comes down to a single avoidable item of $0.12 for each unit. A decision that depends on one item that small deserves the most careful look at its underlying facts before you act. Verify the number with the lease, the loan, the insurance policy, or the asset record, not with a memory of what the number was last year.

A sensitivity check is not a prediction. It does not tell you what will happen. It tells you where the decision is sensitive, and it tells you which numbers to verify before you sign. If the answer holds across the range of likely inputs, you can act with more confidence. If it flips inside that range, you know the decision is really a judgment call, and the non-cost factors below should carry more weight.

Illustrative example: Compare matching quantities

Suppose an illustrative job needs 100 units. Your crew's relevant cost is $10 each, and the sub quotes $10 each. Both options cost the same at that volume. If the sub excludes delivery, add that cost before choosing.

What the numbers leave out

Relevant cost handles the money, but some of the most important factors in a sub-or-crew decision never appear as dollars on a page. They still deserve a place in the decision.

Control over schedule. When your own crew runs late, you can often move people around and push a task forward. When a sub runs late, you may have less room to respond. Ask what a late delivery would cost you in customer relationships, and whether your contracts allow penalties or delays that would fall on you.

Quality and rework. Who fixes a defect, and who pays for it? Compare the sub's warranty and inspection terms with what you would promise your own customer. If your own crew catches problems earlier, that is worth something, and so is a sub who carries the risk of its own mistakes.

Capacity and opportunity. Handing off work may free your crew for jobs with better margins, or it may leave them idle between jobs. Both outcomes matter. Think through the next few months of work, not only this single job.

Dependence on one supplier. If the sub raises prices later, slows down, or walks away in the middle of a job, how long would it take you to rebuild the capability in-house? A sub relationship that is easy to start can be hard to undo, and the reverse is also true. Consider how much room you would have to change course.

Crew retention and skills. Sending work out can slowly erode the skills of the people you rely on. Installers and estimators who stop doing a type of work may not be available when you need them again. If the crew is part of what makes your business worth buying, protect that asset in the decision.

None of these factors has to outweigh the numbers. They should sit next to the numbers, though, and when the relevant-cost comparison is close, they are often what settles it.

What to do today

Take one current sub quote and mark every in-house cost that would stop if you accepted it. Check the least certain item against its contract or asset record.

Questions before you choose

Should allocated overhead count?

Count allocated overhead only when the underlying cost changes between the options.

What if the totals are close?

Weigh schedule, quality, capacity, supplier dependence, and crew retention alongside the cost comparison.

Does a lower unit price settle the decision?

No; compare matching scope and volume, including costs excluded from the quote.