A supplier raises a part 14 percent in March. Your price book still quotes the number you set last year. You find out in December, when the P&L is thinner than it should be and nobody can name a single thing that went wrong.
This is the quietest margin leak in field service, and it is almost never a pricing failure — it is a maintenance failure. Sell prices are set once, from a cost loaded at that moment, and then left alone. The cost underneath keeps moving. If you value inventory at average cost, it does not even move in visible steps: each higher-priced receipt blends into the average and the recorded cost drifts upward a few cents at a time while the sell price sits perfectly still. As of September 2026, with supplier increases arriving more often and with less notice than they did five years ago, a shop with no repricing routine is running on last year's economics without knowing it.
The good news is that closing the gap is mechanical. You need a way to detect the drift, a decision rule for what to do when you find it, a policy for estimates and standing customers already in the field, and a trigger that fires at receiving instead of at year-end. None of it requires a purchasing department. This is the routine.
Why the gap opens without anyone deciding to let it
Three things conspire.
The price was set from a snapshot. When you built the catalog you took each part's cost that day and applied a markup. That is the correct way to start. The problem is that it encodes one moment in time, and nothing in the daily workflow ever re-asks the question. The tech sells the part, the invoice prints the price, the job closes, and nobody compares the two numbers again. The logic of parts markup and margin in field service assumes a live cost underneath it; without maintenance, the markup is applied to a fossil.
Average cost hides the movement. FIFO, LIFO, and average-cost valuations behave differently under rising prices, and average cost is the one that disguises the change most effectively. Buy ten filters at 18 dollars, then ten at 21, and your recorded cost is 19.50 — not 21. Your margin already fell and no line item anywhere announced it. Whichever method you use, the thing to internalize is that recorded cost and replacement cost are two different numbers, and pricing off the former while buying at the latter is how a shop slowly stops making money on its best-selling item.
Increases arrive as invoice lines, not announcements. Few suppliers send a letter. The new price shows up on a receiving document that somebody checks for quantity and then files. Unless the higher cost is compared against your sell price at that moment, the information enters your business and dies there.
The SBA's guidance on managing business finances is direct about continuous tracking of income and expenses being the baseline for staying solvent, and this is exactly the kind of expense movement it is describing — small, continuous, and invisible unless you go looking.
Find the gap with a cost-versus-price variance pass
The detection step is one report, run on a schedule. For each stocked part, put current recorded cost next to current sell price, compute margin percent, and compare it to what you intended when you set the price. Sort by the size of the gap, multiplied by units sold.
That last multiplier matters more than the gap itself. A part whose margin fell 9 points but sells twice a year is a rounding error. A part whose margin fell 3 points and sells every week is the one quietly costing you real money. Sorting by gap times volume puts the list in the order your attention should follow it.
Then set the cadence, because a review that covers everything happens once and never again:
- Quarterly, your top movers. Take the roughly 20 parts that generate the majority of your parts revenue. That is a 30-minute job four times a year.
- Annually, everything else. A full sweep once a year catches the slow drift on the long tail.
- Immediately, on any purchase order that lands above the recorded cost. This is the trigger that makes the other two mostly confirmatory.
Pair the variance pass with your regular stock counts and shrinkage checks, because the two share an input. A count that disagrees with the system means your cost-per-unit is being computed against a quantity that was never right, and the margin number inherits that error.
Hold margin percent, hold dollar margin, or pass through nothing
Once you have found a gap, you have three honest choices, and they are genuinely different decisions rather than three flavors of the same one. Here is a labelled illustrative example with our own numbers.
A part costs you 120 dollars and you sell it for 199. That is 79 dollars of margin, just under 40 percent. The supplier raises the cost 14 percent, to 136.80. Your options:
| Approach | New sell price | Dollar margin | Margin percent | What it does to you |
|---|---|---|---|---|
| Pass through nothing | 199.00 | 62.20 | 31.3% | Absorbs the whole increase; invisible until the annual P&L |
| Hold dollar margin | 215.80 | 79.00 | 36.6% | Keeps the same profit per unit; margin percent still slides |
| Hold margin percent | 228.00 | 91.20 | 40.0% | Restores the original economics; largest customer-visible jump |
The gap between the two repricing options is 12.20 on a single part, and it widens fast as the part gets more expensive — on a 900-dollar compressor the same 14 percent move puts roughly 90 dollars between the two answers. That is why "just raise prices" is not an instruction anybody can execute.
The rule we would suggest, and which is easy to defend to a customer:
- Hold margin percent on parts that carry risk you are actually underwriting — warranty exposure, high carrying cost, special handling, real chance of a defective return. Your margin is paying for something.
- Hold dollar margin on straightforward pass-through parts a customer can price-check online in ten seconds. The profit per unit is preserved and the sticker stays credible.
- Pass through nothing only as a deliberate, temporary decision on a loss-leader you have named as such, with a date to revisit.
The warranty point is not theoretical. Every part you sell that later comes back is margin you already booked and are now spending; how you handle vendor parts warranty credits and defective returns determines whether that exposure is recovered or eaten. A part with a high return rate and a thin margin is a part priced wrong.
And none of this is visible at the catalog level alone. A part repriced correctly can still lose money on the job if the labor line did not move with it, which is why the repricing pass and job costing for true profit per job belong in the same conversation.
Estimates already in the field
This is the part most owners get wrong, and the fix costs nothing.
An estimate you issued three weeks ago at last quarter's cost is a price you gave a customer in good faith. If your cost has moved since, you have a bad set of choices: honor it and lose money, or call them back and renegotiate a number you already committed to — which damages trust far beyond the value of the one job.
Estimate expiry dates exist precisely for this. Put a validity period on every estimate as a matter of routine — 15 or 30 days is standard, shorter on volatile categories — and state it on the estimate itself in plain language. Then the policy writes itself: anything inside its window is honored without argument, anything outside it is requoted at current cost, and nobody has an uncomfortable phone call.
That discipline has a second benefit. An estimate with a real deadline converts better than one that hangs open indefinitely, which is the whole mechanic behind estimate follow-up and win rate. The expiry date is doing margin protection and sales work at the same time.
When a cost increase does land, pull the open estimates that touch the affected parts and sort them by expiry. Anything expiring in the next week, leave alone. Anything past expiry, requote quietly. Anything with no expiry date on it — honor it, and treat that as the reason to add expiry dates to every estimate from today forward.
Standing customers and agreed price lists
Commercial accounts, property managers, and maintenance-plan customers are usually on an agreed list, and they are the relationships most worth handling carefully.
Three things make cost-driven increases land well with these customers:
- Notice in writing, before the effective date. Thirty days is a reasonable default; check what the agreement says.
- Line-level detail, not a blanket percentage. "These six parts moved, here are the new prices, effective the first" reads as a real business event. "Everything goes up 8 percent" reads as opportunism.
- Timed to renewal where you can. Increases are far easier at a natural boundary than mid-term.
Purchasing managers at commercial accounts are absorbing the same supplier increases you are. A documented, itemized increase is something they can explain internally; a vague one is something they shop. The mechanics of the conversation are the same ones in raising prices in a service business, applied to a narrower set of line items and with a much stronger justification behind them.
Keep the notice and the old list. The IRS recordkeeping guidance for small businesses makes the point that electronic records satisfy the same requirement as paper, and a pricing history you can actually retrieve is what settles a billing question a year later without anyone's memory being involved.
Make the purchase order the trigger
The whole routine works or fails on one habit: a purchase order that receives above the recorded cost should prompt a look at the sell price right then.
Not at quarter end. Not when somebody notices. At receiving, while the number is in front of whoever is putting the parts away.
That is a five-second check on most receipts and it converts the problem from an annual archaeology project into a continuous small adjustment. It also means your quarterly variance pass mostly confirms what you already fixed, which is what a healthy routine looks like. Reorder alerts and purchase orders are the natural home for this, since reorder alerts and purchase orders are already the moment your system learns the new cost.
Route the repriced item into the repair order workflow the same day, so the next job quoting that part quotes the new number rather than the stale one sitting in the catalog.
Running the repricing cycle in IntelliDrive OS
The pieces you need are ordinary inventory and catalog features used deliberately.
Purchase orders record what you actually paid, so recorded cost updates from receiving rather than from memory, and the inventory valuation method — FIFO, LIFO, or average cost — is an explicit setting rather than an accident. Reorder alerts tell you what is about to be bought, which is your advance warning of which costs are about to move.
The service catalog holds parts and services with their own prices, and custom pricing tiers let you keep a standing commercial list separate from retail — so a repricing pass can hit retail immediately and the contract list at renewal, without maintaining two catalogs by hand. Inventory and sales reports export to CSV, which is the practical way to run the variance pass: pull cost and price side by side, sort by gap times volume, and work the top of the list. Building and maintaining those lines is the ongoing half of building a service price book.
Estimates carry expiry and convert to invoices in one click, so an in-window estimate becomes a job at the quoted price and an expired one gets requoted at the current catalog price with no manual bookkeeping. QuickBooks Online two-way sync keeps sales, invoices, and payments aligned with the books, which matters when you are trying to see the margin effect of a repricing decision a quarter later rather than arguing about which system is right. Sales, inventory, and tax reports all export, and every feature named here is in the 79-dollar-a-month flat plan with unlimited users — cost visibility is not a tier you unlock.
Trades with volatile parts costs feel this hardest; refrigerant and compressor pricing alone makes the quarterly pass non-optional for most HVAC shops.
What to do this week
- Export cost and price for every stocked part. Add a margin percent column and a gap-times-volume column.
- Name your top 20 by revenue. Those are the quarterly list. Everything else is annual.
- Decide the rule per category — hold percent on risk-bearing parts, hold dollars on pass-throughs — and write it down.
- Reprice the worst five gaps now. Do not wait for a full pass.
- Add an expiry date to every estimate template. Fifteen or thirty days.
- Make receiving the trigger. Any purchase order above recorded cost gets a same-day price check.
- Draft the commercial notice template so the next increase is a form to fill in, not a decision to agonize over.
The suppliers are going to raise prices again. The only variable you control is how long it takes you to notice.
Related reading: Parts markup and margin in field service, Building a service price book, and Raising prices in a service business. For a complete machine-readable feature and pricing reference, see our LLM reference page.