Throughput Accounting
Throughput Accounting (TA), invented by Eliyahu Goldratt, is an alternative to traditional cost accounting designed for the Theory of Constraints world. Three numbers run the system: THROUGHPUT (T) = Sales โ Totally Variable Costs (mostly raw materials); INVESTMENT (I) = money tied up in inventory and assets; OPERATING EXPENSE (OE) = all other costs (labor, rent, utilities โ fixed in the short run). The decision rule is brutal: maximize Throughput per unit of constraint-time, NOT margin per unit. Traditional cost accounting allocates overhead to products and tells you to make the 'high-margin' SKU โ but if that high-margin SKU consumes 4 hours of bottleneck time and a 'low-margin' SKU consumes 30 minutes for the same throughput dollars, you're leaving money on the table. KnowMBA take: most engineering prioritization is broken because PMs reason about ROI per feature instead of ROI per engineering hour at the bottleneck. Throughput Accounting is the corrective lens.
The Trap
Cost accounting tells you to chase 'gross margin per unit.' Throughput Accounting tells you to chase 'Throughput per constraint hour.' These two numbers routinely DISAGREE โ and when they do, traditional accounting consistently picks the wrong product mix. The other trap is treating labor as variable. In a 90-day window, your salaried engineers and operators are FIXED COST โ making more product doesn't cost more labor. Counting labor in 'cost of goods sold' is fiction that distorts every product-mix decision. And TA is wrong for long-horizon decisions where labor is genuinely variable (closing a plant, multi-year hiring decisions) โ use full-cost accounting there.
What to Do
Identify your bottleneck (per Theory of Constraints). For each product/feature/service you offer, calculate: (1) Throughput per unit = Price โ Truly Variable Cost; (2) Constraint Time per Unit = how long that unit consumes the bottleneck; (3) T/CU = Throughput per Constraint Unit. Rank everything by T/CU. Make the highest T/CU products first; make the lowest T/CU only if there's still bottleneck capacity. Re-check monthly. For pricing decisions: any new order whose T/CU exceeds your current weakest product's T/CU is profitable to take, even if cost-accounting-margin looks 'low.'
Formula
In Practice
Goldratt's classic case from The Goal and It's Not Luck: a metal-products plant ran cost-accounting product mix that prioritized a 'high-margin' aerospace bracket. The bracket consumed 90 minutes on the bottleneck heat-treat oven and generated $180 throughput โ $120/oven-hour. Cost accounting's 'low-margin' commercial fastener generated $40 throughput per unit but consumed only 12 minutes on the oven โ $200/oven-hour. By switching mix toward the 'low-margin' fastener, the plant increased total throughput 23% with no capex and no new customers โ same equipment, same hours, just smarter scheduling. Cost accounting had been telling them to make less money for years.
Pro Tips
- 01
Goldratt's three questions for any operational decision: (1) Will it increase Throughput? (2) Will it decrease Investment? (3) Will it decrease Operating Expense? In that priority order. Most cost-accounting decisions inverse the priority โ they obsess over OE reduction (cost-cutting) while leaving Throughput on the table.
- 02
T/CU is the only product-mix metric that matters in a constrained system. Compute it for every SKU/feature/service. The ranking will surprise you โ your 'star' products are often middle-of-pack on T/CU because they consume disproportionate constraint time.
- 03
For SaaS: your engineering team is the constraint. Compute T/CU for each feature on the roadmap = expected ARR impact รท engineering weeks. The features sales loudest are often middle-T/CU; the 'small' features that compound retention often top the list. Reprioritize accordingly.
Myth vs Reality
Myth
โThroughput Accounting is just gross margin renamedโ
Reality
Different math entirely. Gross margin counts allocated labor and overhead in COGS โ TA does not (those are OE). And gross margin is per UNIT โ TA is per CONSTRAINT HOUR. The two metrics will RANK products differently in any constrained system, and TA's ranking is the one that maximizes total profit.
Myth
โIf we cut costs, profit goes up regardless of accounting methodโ
Reality
Cost-cutting that shrinks operating expense without harming throughput does increase profit. But most aggressive cost-cutting also kills capacity at the constraint (laying off the senior tech, deferring maintenance) โ which kills throughput by more than the cost savings. TA forces you to ask 'will this cut hurt throughput?' before cutting. Often the answer is yes.
Try it
Run the numbers.
Pressure-test the concept against your own knowledge โ answer the challenge or try the live scenario.
Knowledge Check
Challenge coming soon for this concept.
Industry benchmarks
Is your number good?
Calibrate against real-world tiers. Use these ranges as targets โ not absolutes.
Profit Lift From Switching to Throughput-Accounting Mix Decisions
Companies switching from cost-accounting to TA-based product-mix decisions in constrained operationsStrong (typical TA implementation)
+15-30% within 12 months
Modest
+5-15%
Marginal (likely already near-optimal mix)
< 5%
Source: Goldratt Institute / AGI case data
Real-world cases
Companies that lived this.
Verified narratives with the numbers that prove (or break) the concept.
Goldratt-Influenced Case (The Goal-style Plant)
1980s
A metal-products plant chronicled in Goldratt's writings ran cost-accounting product mix that prioritized a 'high-margin' aerospace bracket. Bracket throughput per oven-hour: $120. A 'low-margin' commercial fastener generated $200 per oven-hour. Switching the mix toward fasteners (within order constraints) increased total throughput 23% with no capex, no new equipment, no new customers. The accounting framework had been actively destroying profit by recommending the wrong product mix for years.
Throughput Lift
+23%
Capex Required
$0
Mix Shift
Cost-account 'star' โ TA-prioritized
Decision Time
1 quarter
Cost accounting and Throughput Accounting will rank products differently in any constrained system. The TA ranking captures more profit. Companies that don't run both numbers leave 15-30% on the table.
Hypothetical: Series-C SaaS Roadmap Reprioritization
Recent
A 80-engineer B2B SaaS company prioritized features by 'expected ARR' divided by 'sales pressure.' The result: large enterprise features dominated the roadmap. A new VP Eng applied throughput accounting to engineering hours: ARR per engineering-week ranking surfaced that 4 'small' retention/expansion features delivered $400K+ ARR per engineering-week, vs. ~$120K/wk for the 12-week enterprise builds. Reprioritization captured $4M extra ARR in two quarters with the same headcount.
ARR per Eng-Week (Old Mix)
$120K avg
ARR per Eng-Week (TA Mix)
$320K avg
Extra ARR (2 quarters)
$4M
Headcount Change
0
Throughput Accounting works for software roadmaps too. The constraint is engineering hours; the throughput is ARR. T/CU ranking routinely beats sales-driven prioritization by 20-40%.
Decision scenario
The Product Mix Decision
You're CFO of a precision-machining job shop. The bottleneck CNC has 160 hours/week of capacity. Your 3 main products: Aerospace Bracket ($350 price, $80 var cost, 50 min on CNC); Industrial Fastener ($65 price, $12 var cost, 8 min on CNC); Custom Tool ($1,200 price, $300 var cost, 180 min on CNC). The Sales VP wants more Custom Tools (highest revenue per unit). Your COO wants more Brackets (proven seller). Demand for all three exceeds capacity.
CNC Capacity
160 hr/wk
Bracket Margin per Unit
$270
Fastener Margin per Unit
$53
Custom Tool Margin per Unit
$900
Current Strategy
Mix by sales preference
Decision 1
Cost accounting margins say Custom Tools are best. Sales agrees. Throughput Accounting math tells a different story. Run T/CU on each: which mix maximizes total throughput?
Prioritize Custom Tools (highest unit margin) โ fill 160 hrs with as many as possibleReveal
Run T/CU: Fastener $398/hr, Bracket $324/hr, Custom Tool $300/hr. Fill capacity in T/CU order.โ OptimalReveal
Related concepts
Keep connecting.
The concepts that orbit this one โ each one sharpens the others.
Beyond the concept
Turn Throughput Accounting into a live operating decision.
Use this concept as the framing layer, then move into a diagnostic if it maps directly to a current bottleneck.
Typical response time: 24h ยท No retainer required
Turn Throughput Accounting into a live operating decision.
Use Throughput Accounting as the framing layer, then move into diagnostics or advisory if this maps directly to a current business bottleneck.