Standard Costing and Variance Analysis, Explained: How Price and Quantity Standards Turn Cost Overruns Into Variances You Can Act On

Your October close shows production cost $18,400 over budget, and the meeting to explain it goes exactly nowhere. Purchasing insists lumber prices spiked and nobody could have bought cheaper. The shop floor insists the lumber it got was warped and knotty, so of course scrap rates climbed. Both sides are describing the same $18,400, and your books cannot tell you who is right — because actual-cost bookkeeping records only what you spent, never what you should have spent. That missing benchmark is what standard costing supplies: a predetermined cost for every unit you make, so that every overrun arrives pre-sorted into a price problem, a quantity problem, or both, each with an owner and each with a different fix.

This guide walks through how to set standards for materials, labor, and overhead, how to compute each variance with worked numbers, what favorable and unfavorable really signal, and how to turn a monthly variance report into decisions instead of arguments.

What Standard Costing Actually Is​

Standard costing replaces "what did it cost?" with "what should it have cost?" as the starting point of your cost records. Before the period begins, you set a standard cost for one unit of product: the expected quantity of each input multiplied by its expected price. When production happens, you charge inventory at standard and capture every deviation in a variance account. The variances — not the raw totals — become the monthly conversation.

A standard has two halves for every input: a price standard (what one unit of input should cost) and a quantity standard (how much input one unit of output should consume). That split is the whole point. When costs run over, the price half answers "did we pay too much per unit?" while the quantity half answers "did we use too many units?" Those are different failures with different owners — a buyer negotiates prices, a floor supervisor controls usage — and fixing the wrong one wastes a month.

Standards come in two philosophies. Ideal standards assume perfect conditions: no waste, no downtime, no defects. They look rigorous and function as demoralizers, because nobody ever meets them and every variance report glows red. Attainable (practical) standards build in normal spoilage, realistic efficiency, and scheduled maintenance. Use attainable standards. A standard nobody can hit teaches your team to ignore variance reports, which deletes the entire value of the system.

Standard costing fits any business that makes the same thing repeatedly — a cabinet shop, a candle maker, a bakery, a small electronics assembler. If every job is custom, job costing serves you better; standards need repetition to be worth maintaining.

Setting the Three Standards: A Worked Example​

Follow a concrete example through the rest of this guide. Imagine you make hardwood cutting boards. Your attainable standards per finished board are:

October's actual results: you produced 1,000 boards, bought and used 5,200 board-feet of oak at $4.10 per foot, and logged 530 labor hours at an average of $24.50 per hour. Actual variable overhead was $3,350; actual fixed overhead was $5,100 against a $5,000 budget.

Total actual production cost: $21,320 in lumber, $12,985 in labor, $3,350 variable overhead, $5,100 fixed overhead — $42,755 against a $40,000 standard. You are $2,755 over. Variance analysis now decomposes that number into causes.

Note the term standard quantity allowed (SQ): the input you should have used for your actual output. For lumber that is 1,000 boards times 5 feet = 5,000 board-feet — not 5,200, the amount actually used. Every quantity and efficiency variance measures actual usage against SQ, so volume changes never masquerade as efficiency changes.

Materials Variances: Price vs. Quantity​

Materials split into two variances:

• Materials price variance (MPV) = (Actual Price − Standard Price) × Actual Quantity purchased. It isolates what you paid per unit, holding quantity constant. Compute it at purchase time, on the full amount bought, so purchasing sees the consequence of its buying immediately.
• Materials quantity (usage) variance (MQV) = (Actual Quantity used − Standard Quantity allowed) × Standard Price. It isolates how much you used, valued at standard price so purchasing's price swings do not contaminate the floor's efficiency score.

Run October's lumber:

• MPV = ($4.10 − $4.00) × 5,200 feet = $520 unfavorable. You paid ten cents too much per foot on every foot bought.
• MQV = (5,200 − 5,000) × $4.00 = $800 unfavorable. You burned 200 feet beyond standard.

That settles the opening argument with arithmetic: purchasing owns $520 of the lumber overrun, the floor owns $800, and the total lumber variance of $1,320 ties exactly to actual ($21,320) minus standard allowed (5,000 × $4.00 = $20,000). A variance report that does not foot to the total difference has an arithmetic error — always reconcile.

The two variances often trade off against each other, and that linkage is where the real insight lives. If purchasing buys cheaper, lower-grade lumber, the price variance turns favorable while scrap climbs and the quantity variance turns unfavorable — possibly by more than was saved. Investigate the pair together before praising a favorable price variance or punishing an unfavorable quantity variance.

Labor Variances: Rate vs. Efficiency​

Labor follows the same price/quantity logic under different names:

• Labor rate variance (LRV) = (Actual Rate − Standard Rate) × Actual Hours. It captures pay differences: wage increases, overtime premiums, or a different crew mix than planned.
• Labor efficiency variance (LEV) = (Actual Hours − Standard Hours allowed) × Standard Rate. It captures productivity: how long the work took versus how long it should have taken.

October's numbers:

• LRV = ($24.50 − $24.00) × 530 hours = $265 unfavorable. Average pay ran fifty cents over standard — perhaps overtime to cover an absence.
• LEV = (530 − 500) × $24.00 = $720 unfavorable. The crew took 30 hours longer than the 500 standard hours allowed for 1,000 boards.

Labor variances interact just like materials ones. Replacing two experienced woodworkers with three cheaper trainees can produce a favorable rate variance and an unfavorable efficiency variance that dwarfs it. Overtime shows up as rate variance; rushing shows up as efficiency or as scrap in materials quantity. Read the labor pair alongside materials quantity before deciding what happened on the floor.

Overhead Variances: The Four-Variance Approach​

Overhead cannot be traced to a unit the way lumber and labor can, so its variances compare actual spending and actual activity against a flexible budget. The full treatment is the four-variance approach:

• Variable overhead spending variance = Actual variable overhead − (Actual hours × Standard variable rate). Did you spend more per hour of activity than expected? October: $3,350 − (530 × $6.00) = $3,350 − $3,180 = $170 unfavorable.
• Variable overhead efficiency variance = (Actual hours − Standard hours allowed) × Standard variable rate. This is the labor-efficiency story repriced at the overhead rate — extra hours consume extra power, blades, and glue. October: (530 − 500) × $6.00 = $180 unfavorable.
• Fixed overhead budget (spending) variance = Actual fixed overhead − Budgeted fixed overhead. Rent and depreciation rarely move, so a large variance here usually means an unbudgeted fixed cost or a misclassification. October: $5,100 − $5,000 = $100 unfavorable.
• Fixed overhead volume variance = Budgeted fixed overhead − Applied fixed overhead (standard hours allowed × fixed rate). October: $5,000 − (500 × $10.00 per labor-hour... equivalently 1,000 boards × $5.00) = $0. Volume matched budget exactly.

The volume variance deserves a warning because it is the most misread number in the set. It does not measure spending — you cannot "fix" an unfavorable volume variance by cutting costs. It measures capacity utilization: when production falls short of budget, each unit absorbs less fixed cost, and the unabsorbed remainder surfaces here. Treat it as a signal about demand and scheduling, not about waste.

Add all six variances — $520 + $800 + $265 + $720 + $170 + $180 + $100 + $0 — and you land exactly on the $2,755 total overrun. Every dollar of the gap now has a name, a formula, and an owner.

Favorable Is Not Always Good​

Favorable means actual cost came in below standard; unfavorable means above. Those labels describe profit impact, not performance quality, and some of the most expensive mistakes in manufacturing hide inside favorable variances:

• A favorable materials price variance from bargain-grade inputs routinely causes an unfavorable quantity variance (more scrap), an unfavorable labor efficiency variance (harder to work with), and warranty claims months later.
• A favorable labor efficiency variance can mean rushed work, skipped inspections, or deferred maintenance that returns as a breakdown.
• A favorable fixed overhead budget variance from postponing equipment maintenance is not a saving; it is a repair bill with interest.
• An unfavorable variance can be good news wearing a scary label: overtime to fill a rush order at a premium price, or premium materials for a higher-margin product line.

Judge every variance by its cause, not its sign. The report tells you where to look; only investigation tells you whether to celebrate or intervene.

Turning the Report Into Decisions​

A variance report nobody acts on is an expensive ritual. Three practices keep it a management tool:

Investigate by exception with written thresholds. Decide in advance what triggers a follow-up — say, any variance over $500 or 5% of standard, whichever is larger — and investigate only those. Chasing every $40 variance costs more in meeting time than it can ever recover, and it trains everyone to treat the report as noise.

Assign each variance to exactly one owner. Price variances go to whoever buys, rate variances to whoever staffs and schedules, efficiency and quantity variances to whoever runs the floor, spending variances to whoever approves the spend, volume variance to whoever forecasts demand. A variance with two owners has none.

Revisit standards on a schedule, not in a panic. Prices drift, processes improve, and a standard set three years ago eventually measures history instead of performance. Review price standards when contracts renew and quantity standards annually or after any process change. But resist revising standards just to make variances disappear — a standard rewritten every quarter to match actuals is not a standard, it is a diary.

Booking Standards and Variances in Your Ledger​

If you track production in plain-text books, record inventory at standard and let variances fall out as their own postings, so the ledger itself carries the analysis. October's lumber purchase at the example's numbers looks like this:

2026-10-15 * "Oak lumber purchase — 5,200 board-feet at $4.10, standard $4.00"
  Assets:Inventory:Raw-Materials        5200 LB_OAK @ 4.00 USD
  Expenses:Variances:Materials-Price      520.00 USD
  Liabilities:Accounts-Payable         -21320.00 USD

Inventory enters at the $4.00 standard ($20,800), the $520 price variance posts to its own expense account, and the payable reflects the $21,320 actually owed. At month-end, small immaterial variances close to cost of goods sold; large ones get prorated across inventory and cost of goods sold so the balance sheet stays honest. The Beancount documentation covers inventory booking mechanics in depth, and Fava's reports make the variance accounts easy to watch month to month without building a custom report.

Mistakes That Break the System​

• Setting standards once and freezing them. Stale standards generate variances that measure inflation and process change rather than performance, and the team learns to ignore all of them.
• Using ideal standards to "motivate" people. Permanent red ink does not motivate; it teaches that the report is fiction. Attainable standards with genuine stretch produce variances worth reading.
• Investigating dollars instead of percentages, or percentages instead of dollars. A 40% variance on a $200 line is $80 of noise; a 2% variance on a $400,000 line is $8,000 of signal. Threshold on both.
• Reading one variance in isolation. Price and quantity, rate and efficiency move as pairs. Any conclusion drawn from half a pair is a guess.
• Treating the volume variance as a cost problem. It is a capacity and demand signal. Cutting maintenance to "fix" it converts an accounting variance into a real breakdown.

Your next month-end is the place to start: pick your highest-volume product, set attainable price and quantity standards for its top two inputs from last quarter's actuals, and compute just the four materials and labor variances. Four numbers, four owners, one short meeting about causes — that single cycle will teach you more about your production costs than a year of staring at actual-cost totals.