CSCP Learning System 2024 · Study summary · Inventory and warehouse performance

Inventory and Warehouse KPIs, Made Practical

What each KPI really tells you, how to calculate it, a worked example, what the CSCP says and does not say, and six what-if calculators to test your own numbers.

Source base: APICS/ASCM CSCP Learning System 2024 (v5.2), Modules 1, 2, 4, 5, 6 Plus: labeled industry sources Format: study summary in original wording
Back to Supply Chain Home
The short version

Six ideas that organize everything below

01

Two jobs for inventory KPIs. Cut holding, ordering, and transport cost, and hit service targets for quality, availability, and on-time delivery.

02

Cash is the common currency. Turnover, days of supply, and cash-to-cash all describe how long money sits in stock.

03

Accuracy comes first. No forecast, ABC class, or reorder rule works if you do not know what is physically on hand.

04

Forecast error sets safety stock. Bias, MAD, MAPE, and tracking signal tell you whether to fix the forecast or buy buffer.

05

Warehouse utilization is not one number. Position, floor, and cube differ, and average differs from peak.

06

Few KPIs, tied to strategy. The CSCP suggests staying around 20 or fewer, each linked to a strategic objective.

0
How to read this page

Every KPI carries a source label

The CSCP is rich on inventory KPIs but thin on warehouse operating KPIs. To keep that honest, each item below shows where it comes from.
CSCP
Defined or quoted in the CSCP modules (module number shown).
CSCP·R
Formula reconstructed from CSCP text, definitions, or worked examples because the PDF formula was an image. Verify against the PDF before formal use.
PRACTICE
From industry sources outside the CSCP (WERC, practitioner guides). Mostly secondary summaries.
CONVENTION
A common formula that the CSCP does not state. Useful, but your company definition prevails.
Important: examples and default valuesWorked examples marked module example come from the CSCP text. All other numbers in examples and in the calculators are illustrative placeholders, not benchmarks and not data from any company. Replace them with your own data before drawing conclusions.

Six rules for choosing KPIs (from the CSCP)

  1. You get what you measure. Decide objectives first, then success criteria, then metrics, then challenging but feasible targets, set with the people doing the work. CSCP · M4
  2. All KPIs are metrics, not all metrics are KPIs. Keep the KPI set small (the module suggests no more than 20) and link each one to a specific strategic objective. CSCP · M4
  3. Balance the scorecard. Business process, customer, financial, and growth and learning. Set a baseline and a target for each, using benchmarks. CSCP · M4
  4. Avoid single-silo KPIs. Holding cost at one warehouse or cost of one shipping leg does not show whether the whole chain benefits. CSCP · M4
  5. Pair utilization with inventory. Pushing utilization up builds inventory unless demand exists for it. CSCP · M4
  6. Data must be sufficient, accurate, and regularly collected. A metric with unreliable data misleads. CSCP · M4
1
Inventory health KPIs

How much stock, how fast it moves, what it costs

Inventory can tie up 40 to 50 percent of a supply chain's invested working capital, so these KPIs are really about cash and risk. CSCP · M4

Inventory turnover

CSCP · M2

Answers: How many times a year does the money in stock return as sales?

FormulaAnnual cost of goods sold ÷ average inventory
Example (module)$21 million ÷ $3 million = 7 turns. Higher is preferred: less money tied up, less obsolescence risk, more liquidity.

Watch out: "Good" depends on the business. The module cites about 365 for baguettes, under 1 to 4 for slow spare parts, and 6 to 26 for most manufacturing plants. Retail uses value at selling price; commodity sellers use units.

Average inventory

CSCP · M2

Answers: What is the typical stock level that turnover and costs should be measured against?

Formula½ × average lot size + safety stock (steady demand and lots), or the average of several observations, such as twelve month-end balances
Illustrative arithmeticLot 400 units, safety stock 50 → ½ × 400 + 50 = 250 units

Watch out: Averaging only two balance-sheet dates can mislead. Using more points removes the effect of seasonality.

Inventory days of supply (DOS)

CSCP · M2

Answers: If nothing arrived, how many days would current stock last?

FormulaUnits: on-hand units ÷ daily usage
Value: inventory value ÷ average daily COGS
Example (module)2,000 units ÷ 200 per day = 10 days. Cross-check (computed): 365 ÷ 7 turns ≈ 52 days.

Watch out: One DOS figure hides slow items. Check it by ABC class and by item type.

Cash-to-cash cycle time

CSCP · M2

Answers: How many days is working capital tied up between paying suppliers and collecting from customers?

FormulaInventory days of supply + days sales outstanding − days payables outstanding
Example (module)60 + 40 − 45 = 55 days. Reducing inventory directly shortens it.

Watch out: The module calls it more useful than basic turnover because it includes receivables and payables. Adjust for VMI, consignment, or vertical integration before benchmarking.

Carrying (holding) cost rate

CSCP · M4

Answers: What does it cost per year to keep stock?

FormulaInventory value × carrying cost rate. The module says the rate is typically 15% to 40% of value per year and is built from storage, capital, and risk costs.
Illustrative (25% placeholder)$3,000k × 25% = $750k per year. A 10% stock cut would save $75k a year and release $300k of cash.

Watch out: Do not drop the risk part. The module's example assumes a 2% value loss per day from obsolescence, which makes stock older than 50 days worthless.

In-transit inventory

CSCP·R · M4

Answers: How much stock is on the road or at sea that never shows in the warehouse count?

FormulaAverage annual in-transit inventory depends on transit days and annual demand: annual demand × transit days ÷ 365
Illustrative36,500 units a year × 30 days ÷ 365 = 3,000 units in transit at any time

Watch out: Counting only in-stock inventory can omit a significant share of the total. Cut transit time to cut this stock.

Obsolete and slow-moving stock

CONVENTION

Answers: How much of the stock is unlikely to be used or sold?

Formula (convention)Value of obsolete or non-moving stock ÷ total inventory value. The CSCP lists "obsolete inventory" and "inventory by ABC class" as asset metrics but gives no formula.
Illustrative$180k obsolete ÷ $3,000k total = 6%

Watch out: Define "non-moving" (for example no issue in 12 months) before measuring, and link it to the write-off and reserve policy.

Inventory shrinkage

CSCP · M4 formula: CONVENTION

Answers: How much stock disappears through scrap, theft, deterioration, or evaporation?

Definition (CSCP)Reductions of actual quantities of items in stock, in process, or in transit.
Common formula(Book value − physical value) ÷ book value
IllustrativeBook $3,000k, counted $2,970k → 1.0%

Watch out: Tracking systems (WMS in storage, TMS in transit) make losses visible and valuation accurate.

Reading these togetherTurnover and DOS say the same thing in different units (about 365 ÷ turns = days). Cash-to-cash adds the customer and supplier side. Carrying cost turns days into money. Use one or two for the executive view and the others to diagnose.
2
Inventory control and accuracy KPIs

Do the records match the shelf?

No amount of forecast accuracy, ABC analysis, secure storage, or automated ordering adds up to good inventory management without an accurate, current count. CSCP · M4
Warehouse worker scanning a pallet in a racking aisle
Illustration: counting and scanning keep records and stock aligned. Visual generated with Microsoft Copilot.

Inventory accuracy

CSCP · M4

Answers: For what share of items is the recorded balance within the allowed tolerance of the physical count?

FormulaItems within tolerance ÷ items counted. The module says targets are usually 95% to 99%, depending on item value.
Illustrative962 of 1,000 items within tolerance = 96.2%, inside the 95–99% band.

Watch out: Use tighter tolerances for high-value (A) items. A mismatch between record on hand and physical on hand needs an inventory adjustment.

Location error rate

CSCP · M4

Answers: How many storage locations hold something different from what the system says?

FormulaStorage locations with errors ÷ total storage locations
Illustrative18 ÷ 1,200 = 1.5% of locations in error

Watch out: This is an error rate, so lower is better, unlike the item-level accuracy percentage above.

Cycle counting plan

CSCP · M4

Answers: How many items must be counted each day so every class is covered on schedule, without shutting the site?

FormulaItems in class ÷ working days in the counting cycle = items to count per day
Class (module example)ItemsPolicyCount per day
A1,000Monthly (20 working days)1,000 ÷ 20 = 50
B3,500Quarterly (60 days)3,500 ÷ 60 ≈ 58
C5,500Semiannually (120 days)5,500 ÷ 120 ≈ 46
Total10,000about 154 per day

Smart triggers: count just before a replenishment arrives (stock is low), when stock hits zero, or when the system shows a negative balance. Root causes are easier to find when errors are caught in real time. The what-if lab lets you change the policy.

3
Service and delivery KPIs

Did the customer get what they ordered, when they needed it?

Too little inventory brings long lead times, broken orders, and lost customers. Too much brings write-offs and cash strain. Service KPIs show which side you are on. CSCP · M4

Fill rate (customer service ratio)

CSCP · M6

Answers: Of what customers asked for, how much did we supply?

FormulaItems shipped ÷ items ordered. Variants: unit fill rate, line-item fill rate, and monetary-value fill rate.
Example (module)A customer requests 100 items and only 92 are available → fill rate 92%.

Watch out: 99% sounds strong, but the module notes it can mean 9,000 customers getting late orders on one day. Show real numbers of people affected next to the percentage.

On-time in full (OTIF)

CSCP · M6

Answers: What share of deliveries arrived complete and by the delivery date?

FormulaDeliveries both on time and in full ÷ total deliveries (percentage against a target goal)
Illustrative88 of 100 deliveries complete and on time → 88%

Watch out: The definition of "on time" must come from the customer, and partial deliveries do not count.

Stockout frequency and backorders

CSCP · M6

Answers: How often, how long, and how costly are stockouts?

Measures (module)% of items in stockout · monetary value of items in stockout · average age of stockout · average time to recover an item · orders shipped complete · backorders (unfilled orders or commitments)
Illustrative40 of 2,000 active items out of stock → 2% stockout frequency

Watch out: Weigh stockouts by item criticality, not just count.

Perfect order fulfillment

CSCP·R · M2, M6

Answers: What share of orders got everything right?

FormulaPerfect orders ÷ total orders. An order is perfect only if all seven Rs are right: product, quantity, condition, place, time, customer, cost, with complete and accurate documentation.
Illustrative920 perfect of 1,000 orders = 92%. The module cites AMR research: a 3% rise in perfect orders is linked to a 1% rise in profit.

Watch out: Errors compound across steps, and data is hard to collect across systems. The what-if lab shows how 98.5% line accuracy can become a much lower order-level rate.

Order fulfillment cycle time and dwell

CSCP·R · M2

Answers: How long from order to acceptance, and how much of it is customer-caused waiting?

FormulaAverage of actual order cycle times (order receipt to customer acceptance). Process time = cycle time − order fulfillment dwell time.
Illustrative6-day cycle with 2 days of customer-created dwell → 4 days of process time

Watch out: Idle or non-value-added time is not dwell. Only waiting created by the customer is removed.

Lead time consistency

CSCP · M6

Answers: Can customers rely on the quoted lead time?

Measures (module)Variance in lead times · percentage delivered within quoted lead time
Illustrative85 of 100 orders within quoted lead time → 85%

Watch out: The module values consistency over raw speed, because variable lead times force customers to hold safety stock.

4
Forecast accuracy and safety stock

Error you cannot remove, you must buffer

Measuring the extent of forecast deviation tells you whether to improve the forecast or rely on safety stock to meet service goals. CSCP · M1

Forecast error, accuracy, and MAPE

CSCP · M1

Answers: How far off was the forecast, in percent?

FormulaAbsolute percentage error (APE) = |actual − forecast| ÷ actual
Forecast accuracy = 1 − error %
MAPE = average APE over the periods
Example (module numbers; % computed)Actual 29, forecast 33.51 → error 4.51 → APE ≈ 15.6%, accuracy ≈ 84.4%. The module's twelve-month example gives MAPE 17.2%.

Watch out: MAD is an absolute number and means little without scale. MAPE adds the ratio to demand. Review any SKU or family with MAPE above a set limit.

Bias and tracking signal

CSCP · M1

Answers: Is the forecast consistently high or low?

FormulaBias = Σ(actual − forecast); any non-zero result is bias
Tracking signal = cumulative error ÷ MAD
Rule of thumb: stay within ±4, else review the method
Example (module numbers; sums computed)Actual 34, 29, 13 (total 76) vs forecast 34.4, 33.51, 12.05 (total 79.96) → bias −3.96. Negative means actual demand was below forecast.

Watch out: A continually negative tracking signal means over-forecasting; continually positive means under-forecasting. Ideally it oscillates around zero. Track by SKU, monthly.

MAD and the link to safety stock

CSCP · M1

Answers: What is the typical size of the miss, and how much buffer does it justify?

FormulaMAD = Σ|actual − forecast| ÷ n
Standard deviation ≈ 1.25 × MAD
Example (module)1.0 MAD × 1.25 = 1.25 units of safety stock at the 84.13% service level, the same as 1.25 standard deviations × a factor of 1.00.

Watch out: Use the MAD-to-SD rule as an approximation and review it when the error pattern changes.

Safety stock and safety stock usage

CSCP · M4 formula: CSCP·R

Answers: How much buffer do we hold, and is it being used as intended?

MethodsFixed amount · coverage (days of demand plan) · statistical (service level via standard deviations).
Statistical formSafety stock = z × demand standard deviation per day × √lead time days (z from the service level)
Example (module, coverage method)10 units/day × 30 days = 300 units. Drivers: order frequency, demand variability in lead time, lead time length, forecast accuracy, and industry requirements.

Watch out: Set a target for how often safety stock is used. More often than target means too little cycle stock; never used means too much buffer. For slow (C) items the module prefers safety lead time over safety stock. The curve is steep: going from 95% to 98% order completion needs far more stock than reaching 95%.

5
Warehouse KPIs

Space, accuracy, productivity, speed, and cost

Here the CSCP lists the metrics but rarely gives formulas. Where it is silent, the sources are labeled PRACTICE or CONVENTION.

Space and capacity

Warehouse capacity utilization

listed · M2 formulas: PRACTICE

Answers: How much of the usable storage space is in use?

Formulas (practice; measure one at a time)Position occupancy = occupied pallet positions ÷ usable positions
Floor utilization = occupied storage footprint ÷ usable storage floor area
Cube utilization = stored load volume ÷ usable storage volume
Examples from a practitioner guide4,500 of 5,000 positions = 90% · 600 of 800 m² = 75% · 1,200 of 2,000 m³ = 60%. A warehouse can show high position occupancy and low cube use when many positions hold short or partial loads.

Watch out: Exclude non-storage space (offices, receiving, staging, charging, aisles you cannot store in) and blocked locations from "usable" capacity.

Average vs peak capacity used

PRACTICE

Answers: Are we sized for the usual level or for the busiest day?

Why two numbersWERC reports average and peak warehouse capacity used as separate metrics. The CSCP sizes space to average (mode) inventory divided by a target utilization, and says fixed-location storage needs the sum of every item's maximum.
IllustrativeAverage 78% but peak 92% of positions: the peak-to-average ratio is 1.18. A target of 80% on the average leaves little room for the peak.

Watch out: Seasonal or project-driven stock can make the peak far above the average. Calibrate the target to your own ratio. The what-if lab does this.

Space per pallet position

CONVENTION

Answers: How efficiently does the building turn floor area into storage?

FormulaTotal building area ÷ pallet positions
Check on the CSCP example12,375 m² ÷ 11,000 bays ≈ 1.1 m² per position (computed). Practitioner guides cite roughly 1.3 to 2.8 m² for selective racking including aisles, depending on lift truck and what is counted.

Watch out: The CSCP example sits at the low end of practitioner ranges, so calibrate the multipliers from your own existing sites, as the module itself recommends.

Storage density and accessibility

concept · M2, M5

Answers: Are we using cubic space well without blocking access?

Concepts (no formulas in the module)Cube utilization comes from vertical racking. WMS exception reports track storage density. The ideal is 100% accessibility: every pallet reachable without moving another.
Practice noteBlock stacking and drive-in systems trade accessibility for density and suffer honeycombing, empty slots that cannot be used by another SKU.

Watch out: Filling a building with single selective racks may maximize storage but congest workflow and hurt throughput.

Accuracy, quality, and productivity

Picking, shipping, and document accuracy; damage

CSCP · M2

Answers: How often is the work done right the first time?

Formula (module)Accuracy = number of times the activity was done correctly ÷ total number of times it was done. Also: damage frequency (total or by area: warehouse, loading, transportation), document and invoicing accuracy, returns, credit claims.
Illustrative9,850 correct of 10,000 pick lines = 98.5%

Watch out: Damage by area is hard to tabulate because it is often reported only when the goods reach the customer. Line accuracy also compounds into a lower order-level rate.

Labor productivity

CSCP · M2

Answers: How much output do we get per person or per labor dollar?

Measures (module)Labor productivity in total and by area (including warehousing) · units shipped per employee · units shipped per labor dollar · equipment downtime. General form: actual output ÷ actual input.
Illustrative12,000 units shipped ÷ 20 employees = 600 per employee; ÷ 30,000 labor cost = 0.4 units per labor dollar

Watch out: Labor is a major expense, so analyze by area, hours, and headcount. Compare with schedule, budget, history, or competitor benchmarks.

Utilization, efficiency, and rated capacity

utilization · M4 others: CSCP·R

Answers: How much work can a crew or machine realistically do, and how hard is it being used?

UtilizationHours actually worked ÷ available hours
Example (module): 90 ÷ 120 = 75%
EfficiencyStandard hours produced ÷ actual hours worked
Example: 110 ÷ 100 = 110%
Rated capacityAvailable time × utilization × efficiency
Example: 160 × 0.85 × 0.90 = 122.4 standard hours

Watch out: The module warns that traditional efficiency and utilization measures are often poorly correlated with organizational performance. Pair them with throughput, inventory, and operating expense so that high utilization does not just build stock.

Speed, flow, and cost

Dock-to-stock time

PRACTICE formula: CONVENTION

Answers: How long from the truck arriving until the goods can be picked?

Formula (convention)Average of (time stock becomes available − time received at dock)
Source noteWERC's DC Measures report lists dock-to-stock cycle time in hours among its five most popular metrics, with on-time shipments, average and peak capacity used, and backorders as % of lines. The CSCP does not define it.

Watch out: Agree where the clock starts (arrival, check-in, unload) and stops (put-away confirmed in system).

Warehouse cost KPIs

CSCP · M2, M5 ratios: CONVENTION

Answers: What does it cost to run and stock the building, and are we cutting cost in one place only to raise it elsewhere?

CSCPTotal supply chain management cost % = (ordering + acquisition + inventory carrying + SC finance/planning + IT cost) ÷ product revenue. The total cost concept: choose the option with the lowest total logistics cost at equal service. Storage cost components: rent, taxes, handling, equipment leases, depreciation, power.
ConventionCost per pallet position = total storage cost ÷ positions; cost per order = total cost ÷ orders
Module exampleMore warehouses can cut transport cost but raise inventory, floor space, and overhead.

Watch out: Judge any warehouse cost change against total logistics cost and service, not the warehouse budget alone.

6
What-if lab

Change a number and watch the KPIs move

Six calculators run in your browser. Nothing is sent or stored. Each one starts with example values so you can see how it behaves, then replace them with your own.
Two logistics planners reviewing a warehouse floor plan
Illustration: planners testing scenarios. Visual generated with Microsoft Copilot.

1 · Turnover, days of supply, cash-to-cash, and carrying cost

Enter your baseline, then a scenario change. Days of supply = average inventory ÷ (COGS ÷ 365). CSCP

Scenario
—Cash released (+) or absorbed (−)
—Turnover after scenario
—Cash-to-cash after (days)
KPIBaselineScenarioChange

The module C2C preset sets average inventory so that days of supply is 60 (with DSO 40 and DPO 45), reproducing the module's 55-day example.

2 · Safety stock, reorder point, and the cost of variability

Safety stock = z × σ × √lead time CSCP·R. Reorder point = demand during lead time + safety stock. If you only know MAD, σ ≈ 1.25 × MAD CSCP · M1. Assumes normal demand and a fixed lead time.

Scenario
—Safety stock after (units)
—Yearly carrying cost change on safety stock
—Reorder point after (units)
MeasureBaselineScenarioChange

3 · Forecast accuracy: MAD, MAPE, bias, and tracking signal

Error = actual − forecast, as in the module. Enter up to six periods; blank rows are ignored. CSCP · M1

PeriodActualForecastError (A − F)APE
—MAD
—MAPE
—Bias (cumulative error)
—Tracking signal
Reading—

Tracking signal = cumulative error ÷ MAD; the module's rule of thumb is to review the method outside ±4. With only a few periods the signal is unstable; the module's own example uses 12 months.

4 · Cycle counting plan and accuracy check

Items per day = items in class ÷ working days in the cycle. Defaults are the module example (King Hospital Supplies).

Accuracy check (illustrative sample)
—Items to count per day
—Inventory accuracy
—vs 95–99% target band
ClassItemsCycle daysCount per dayCounts per year*

*Counts per year assume 240 working days (a placeholder you can mentally rescale). The module example gives 50 + 58 + 46 ≈ 154 per day.

5 · Warehouse position utilization: average, peak, and headroom

Utilization = occupied ÷ usable positions PRACTICE. Time to threshold = ln(threshold level ÷ current level) ÷ ln(1 + growth), assuming steady compound growth CONVENTION.

—Average utilization
—Peak utilization
—Peak ÷ average
CheckResultMeaning

6 · Line accuracy, order-level accuracy, productivity, and cost of errors

Order-level accuracy = line accuracy ^ lines per order. Perfect-order estimate = order-level accuracy × damage-free rate × on-time rate, assuming independent errors CONVENTION (the SCOR definition counts any failure as a failed order).

Scenario
—Perfect-order estimate after
—Yearly error cost change
—Lines per picker-hour after
MeasureBaselineScenarioChange
7
How the KPIs connect

A map of cause and effect

KPIs are not independent. Forecast error drives safety stock, safety stock drives inventory and cash, and accuracy in the warehouse drives service. Read the map left to right.
COST AND CASH SIDE Forecast accuracyMAPE · bias · tracking signal Safety stockand reorder point Average inventorylevels and mix Turnover · days of supplyhow fast stock moves Carrying cost Cash-to-cash WAREHOUSE SIDE Capacity utilizationaverage vs peak Productivity · dock-to-stockflow and congestion SERVICE SIDE Inventory accuracycycle counts · pick accuracy Fill rate · stockoutsavailability OTIF · perfect orderdelivery reliability Customer serviceand profit Conceptual relationships drawn for this summary; they are not an exhibit from the CSCP. Dashed line: weaker or indirect link.
Use the map to avoid trade-off trapsCutting safety stock lowers carrying cost but can lower fill rate. Raising utilization can lower cost per position but lift peak congestion and slow productivity. Improving forecast accuracy is the one move that helps both sides.
8
KPI packs by warehouse type

Start with a small pack that fits the site

The CSCP advises a small, strategy-linked KPI set. Below is a suggested starting pack for three common storage types.
Technician selecting a spare part from shelves in a maintenance storeroom
Illustration: an MRO storeroom. Visual generated with Microsoft Copilot.
Status of this tableThe packs are a synthesis for this summary, built from CSCP statements and the labeled practice sources. They are suggestions, not CSCP prescriptions. Choose by what decisions the KPI will drive.
Storage typeWhat the sources say about itSuggested core KPIsWatch-outs
MRO and spare parts storeMRO is expensed, often held in dispersed locations, with risks of obsolescence, buying items already in stock, and pilferage. Fixed-location storage is common for slow spares. Spare-part turnover is under 1 to 4. CSCPInventory accuracy (tight on A items) · location error rate · critical-spare fill rate and stockout frequency · obsolete % · days of supply by ABC class · duplicate purchases of in-stock itemsLow turnover is normal; do not chase turns at the cost of critical-spare availability. Size space to maximums when fixed locations are used.
Laydown, yard, and bulk storageBlock stacking suffers honeycombing and trades space against aisle area; shared (random) policy uses space better than dedicated lanes. PRACTICEArea (floor) utilization, average and peak · dwell and age of stock · blocked or relocation moves · inventory accuracy of location · disposition of aged itemsCube utilization is less meaningful where stack heights vary. Use positions or area, and measure peak, not just average.
Distribution or finished-goods warehouseWERC's top metrics: on-time shipments, dock-to-stock, average and peak capacity used, backorders % of lines. PRACTICE CSCP lists picking and shipping accuracy, damage, and labor productivity. CSCPOTIF and fill rate · picking accuracy and damage · dock-to-stock · position and cube utilization (average and peak) · units per labor hour or per labor dollar · cost per orderLine accuracy compounds to a lower order-level rate. High utilization can congest and slow throughput.
9
Targets and benchmarks

Numbers you can quote, and where they come from

Only a few numeric targets appear in the CSCP. Treat everything else as a reference to calibrate against your own history.
KPIFigureSource and caveat
Inventory accuracy95% to 99%, depending on item valueCSCP · M4
Tracking signalWithin ±4 means the method is workingCSCP · M1 rule of thumb
Carrying cost rateTypically 15% to 40% of inventory value per yearCSCP · M4
Inventory turnoverBaguettes ≈ 365; manufacturing plants 6 to 26; slow spare parts under 1 to 4CSCP · M2 illustrations by industry
Perfect orders and profitA 3% rise in perfect orders is linked to a 1% rise in profitCSCP · M6 citing AMR research
Inventory share of working capitalAs much as 40% to 50%CSCP · M4
Target warehouse utilization (CSCP example)80% in the worked example, set by policy from strategic and risk analysisCSCP · M5 example value, not a benchmark
Dynamic utilization targetMost warehouses aim for 80% to 85%; higher rates create congestionPRACTICE practitioner guide
Efficient operating range70% to 85% (and 75% to 90% with peak headroom)PRACTICE consulting blog
Best-in-class capacity used90% or higher on average; earlier medians near 85%PRACTICE secondary report of WERC DC Measures; full data is paid
CautionThe practice figures come from vendor and consultancy summaries, are not the same definition across sources, and are not guaranteed to fit your operation. Use them as a conversation starter, not a target.
10
Gaps and limitations

What this page cannot tell you

The CSCP gives no formulas for core warehouse KPIs

Dock-to-stock, put-away time, pick rate, occupancy, storage density, and cost per order or per pallet are not defined with formulas in the modules searched. They are covered here as practice or convention.

Some formulas were images

Several formulas in the PDF (turnover variants, cash-to-cash, efficiency, rated capacity, supplier index) are images. Where reconstructed from text and examples, they are tagged CSCP·R. Verify against the PDF before using them in an exam or report.

Reading coverage

Modules 1, 2, 4, and 6 were searched or read for KPI content. Module 5 was read for the logistics, warehouse design, and capacity sections, not every page. Modules 3, 7, and 8 were not searched in depth and may hold further KPIs.

No MRO-specific service KPIs

The CSCP does not define spare-part service level, criticality-weighted availability, or non-moving ratios. The pack for MRO stores relies on general inventory KPIs and the module's MRO challenges.

Calculators are simplified

They assume normal demand, fixed lead time, constant growth, and independent errors. Real data has seasonality, lumpy demand, and correlated failures.

Your definitions come first

Define "on time", "in full", "non-moving", "usable space", and the clock start and stop for time KPIs before you set targets. A good KPI is one everyone calculates the same way.

11
KPI dictionary

All KPIs in one searchable table

KPIAreaDefinition or formulaSource