Inventory Analytics

Inventory Days on Hand by SKU:
The Metric Every Manufacturing Ops Team Is Missing

Your ERP reports 45 days of inventory on hand. That number is accurate and useless. Somewhere in your warehouse right now there are bearings with 700 days of supply three aisles from a control relay four days from a stockout. Here is how SKU-level DOH changes every inventory decision you make.

10 min read Inventory management Working capital

Why the aggregate DOH number lies to you

Every ERP system will give you a total inventory days on hand figure. It is calculated from total inventory value divided by average daily cost of goods sold, and it tells you approximately how many days your current stock will last at the current consumption rate. The number sounds useful. It is not.

The problem is aggregation. A manufacturer with 2,000 active SKUs will have some of those SKUs dramatically over-stocked and some dramatically under-stocked at any given moment. When you average all of them together, the extremes cancel out. The slow movers inflate the aggregate. The stockout risks get buried. The single number you see in your ERP report is an average of situations that require completely opposite responses, telling you to do nothing when half your items need immediate attention.

The operations teams that manage inventory well do not look at aggregate DOH. They look at DOH by SKU, sorted by the ones furthest from their target range. That view surfaces the slow movers driving capital costs, the stockout risks threatening service levels, and the imbalances across plants that represent transfer opportunities. None of that is visible in the aggregate.

The rule of thumb: in a typical manufacturing environment, 15 to 20 percent of active SKUs account for over 80 percent of the working capital tied up in excess inventory. Those SKUs are invisible in any aggregate report. SKU-level DOH is what finds them.

How inventory days on hand is calculated at the SKU level

The calculation itself is straightforward. For any individual SKU, days on hand is the quantity currently on hand divided by the average daily usage rate. The question that matters in practice is which usage rate to use as the denominator, because the answer changes significantly depending on the choice.

Three options for average daily usage

Trailing 13-week consumption: Uses actual warehouse issues and production consumption over the past 13 weeks, annualized to a daily rate. This is the most responsive to recent demand shifts. It picks up trend changes quickly and is the right choice for items with relatively stable or trending demand. The limitation is volatility for seasonal items: if you pull 13-week usage during the slow season, you will over-count DOH and miss the peak demand exposure.

Trailing 52-week consumption: Smooths out seasonality by using a full year of actual consumption data. Better for items with known seasonal cycles, but it lags trend changes. If demand for a particular component dropped 40 percent three months ago following a customer loss, the 52-week average will undercount your DOH for several more months.

Forward demand from open orders: Uses open production orders and customer orders to calculate committed future consumption. This is the most accurate approach for make-to-order and configure-to-order manufacturers where open orders represent actual committed demand rather than a statistical forecast. It is less useful for make-to-stock environments where the open order backlog does not represent total future consumption.

For most discrete manufacturers, the best practice is a blended approach: use forward order demand where available and reliable, fall back to trailing 13-week consumption for items where open orders do not fully represent future demand, and flag trailing 52-week usage for review on items known to be seasonal.

The three DOH zones and what each one costs you

Once you have DOH at the SKU level, the first analytical step is classifying every item into one of three zones. The zone boundaries are not universal and should be calibrated to each item's supplier lead time and demand variability, but the framework below applies as a starting point to most discrete manufacturing environments.

Inventory DOH zones · discrete manufacturing baseline
Stockout Risk
<15 days on hand
  • Below typical replenishment cycle for most suppliers
  • Any demand spike or supplier delay triggers a stockout
  • Line stops are possible with no warning time
  • Premium freight already being spent or about to be
Cost: premium freight + line downtime + customer penalties
Healthy Coverage
15–90 days on hand
  • Sufficient buffer for normal replenishment cycles
  • Demand variability absorbed without emergency orders
  • Working capital cost is proportionate to lead time need
  • Upper bound should be set per item based on lead time
Target zone · calibrate per item by lead time
Slow Mover / Excess
>90 days on hand
  • Capital tied up earning nothing and costing carrying costs
  • Obsolescence risk increases with every passing quarter
  • Often the result of a demand drop or forecast error
  • Write-down candidate if DOH exceeds 180 days
Cost: carrying cost (typically 8–12% annually) + obsolescence

The middle zone boundary deserves more nuance. An item with a 60-day supplier lead time should target 75 to 90 days on hand to maintain a healthy buffer. An item available domestically in two weeks can safely target 15 to 20 days. Using a single upper boundary for all SKUs is a common mistake that either overstates the amount of excess inventory or fails to flag true slow movers among items with long lead times. The right approach is to set the healthy zone ceiling individually per item based on lead time plus a safety stock factor calibrated to demand variability.

What SKU-level DOH looks like in practice

The table below illustrates the kind of spread that is present in almost every manufacturing warehouse, invisible in the aggregate. Eight items from the same plant. Four of them need immediate attention in opposite directions. The aggregate DOH across these eight items comes out to 86 days on hand, which sounds acceptable. The reality is that two items are days from a stockout and three are consuming working capital at a cost of roughly $11,000 per year in carrying costs alone.

SKU-level inventory DOH · Plant A · sample of 8 items
SKU Description On Hand Avg Daily Usage DOH Zone $ On Hand
SKF-6205 Bearing, Deep Groove Ball 2,400 48 50 Healthy $18,000
SS-304-2MM Steel Sheet 304, 2mm 180 36 5 Stockout risk $4,320
WD-HSGK-440 Washer, Hardened 440C 84,000 120 700 Excess $63,000
O-RNG-VT-112 O-Ring, Viton 112 22,000 110 200 Excess $11,000
RLY-24V-CTRL Control Relay 24V 12 3 4 Stockout risk $840
ALU-6061-T6 Aluminum Bar 6061-T6 940 28 34 Healthy $23,500
FLX-NCLN-1L Flux No-Clean 1L 1,200 4 300 Excess $6,000
MTR-SVO-3A Servo Motor 3A 8 0.8 10 Stockout risk $12,000

Two things stand out in the table that are completely invisible in any aggregate view. First, the servo motor (MTR-SVO-3A) has a 10-day DOH and a 42-day supplier lead time. That item is already past the point where a standard purchase order can prevent a stockout. The purchasing team needs to know this today, not at the next weekly inventory review. Second, the 440C hardened washer (WD-HSGK-440) has 700 days of supply. At 8 percent annual carrying cost, that single SKU costs approximately $5,000 per year to hold. A purchasing pause plus a review of the minimum order quantity would eliminate that cost within one replenishment cycle.

The multi-plant dimension: imbalance as opportunity

For manufacturers running multiple plants, DOH by SKU reveals a third category of problem that is invisible even in plant-level reports: inventory imbalance across sites. A SKU at stockout risk at one plant and with excess stock at a sister plant is not two separate problems. It is a transfer opportunity. Identifying it requires a unified item master so that the same physical part is recognized under the same identity across all ERP systems, and a cross-plant DOH view that surfaces the mismatch automatically.

Without a unified item master, this comparison is impossible. The same part might be coded as RLY-24V-CTRL at Plant A and 24V-CTRL-RELAY at Plant B. The purchasing teams at each plant do not know the other plant has the same item. The Plant A buyer expedites an order. The Plant B buyer sits on 150 units with no home. Both decisions are rational given what each buyer can see. Both are wrong given what a system-level view would show.

Cross-plant DOH comparison · same SKU, three plants
RLY-24V-CTRL Control Relay 24V · Supplier lead time: 28 days · Unit cost: $70
Midwest Precision
Epicor
4
days on hand · 12 units
Stockout risk
Valley Fabrication
Sage 100
54
days on hand · 162 units
Healthy
Summit Components
Dynamics BC
148
days on hand · 444 units
Excess
IQ Recommendation: Transfer 90 units from Summit Components to Midwest Precision. Estimated freight cost: $380. Avoids emergency order premium freight of $2,100 and supplier expedite fee of $850. Brings Midwest to 34 days DOH. Reduces Summit to 118 days DOH, still above healthy threshold but meaningfully lower.

This kind of cross-plant signal requires three things to work: a unified item master that recognizes the same part across all three ERPs, a cross-plant DOH calculation that runs continuously rather than requiring manual extraction from each ERP, and an alerting layer that surfaces the imbalance when it becomes actionable rather than waiting for a weekly review meeting to discover it.

The working capital math: slow movers as a continuous interest payment

Inventory sitting above 90 days of supply is not a neutral condition. It is an active cost. Every dollar of excess inventory is a dollar that was borrowed from somewhere, whether from cash reserves, a revolving credit facility, or opportunity cost against alternative investments. At a typical manufacturing company, the carrying cost of inventory runs 8 to 12 percent annually when you account for capital cost, warehousing, handling, insurance, and obsolescence risk.

Expressed differently: $1M in excess inventory costs $80,000 to $120,000 per year to hold. That is before any write-downs when the excess eventually becomes obsolete.

Inventory working capital snapshot · Plant A · 2,847 active SKUs
Total inventory value
$4.8M
across 2,847 active SKUs
Healthy zone (15–90d)
$2.6M
54% of total · 1,541 SKUs
Slow movers (>90d)
$1.7M
35% of total · 847 SKUs
Stockout risk (<15d)
$0.5M
11% of total · 459 SKUs

The working capital release opportunity in that snapshot is not $1.7M. The realistic reduction target is the portion of the $1.7M that represents true excess above what the lead time and safety stock requirements actually need. For most manufacturing environments, 40 to 60 percent of slow-mover inventory value is reducible within 12 to 18 months through purchasing pauses, order quantity adjustments, and cross-plant transfers. The rest is structural safety stock that is classified as excess by a simple DOH threshold but serves a real function that a more nuanced per-item target would recognize.

Getting to that per-item target requires the same data that DOH analysis provides: actual usage rates by SKU, supplier lead times, and demand variability. Once the analysis is running at the SKU level, the investment prioritization calculates itself.

Continuous monitoring versus periodic inventory reviews

Most manufacturing operations run an inventory review cycle of some kind: weekly or monthly, sorted by value or by some aging bucket. The problem is that inventory situations move faster than review cycles. A SKU that was at 45 days two weeks ago can be at 8 days today if a large production order consumed more than expected or a supplier shipment arrived short. By the time the weekly review finds it, the stockout may already be locked in.

The alternative is continuous monitoring with threshold-based alerts. Rather than reviewing all 2,000 SKUs weekly and hoping to spot the critical ones, you configure alert thresholds per item zone: alert when any SKU crosses below 15 days, alert when any SKU climbs above 90 days, alert when cross-plant DOH imbalance on any shared item exceeds a threshold that makes a transfer economic. The alert goes directly to the buyer responsible for that item category, with the context they need to act immediately.

This is what Inventory IQ does through the IQ Insights monitoring layer. The alerts are configured per threshold, not per review cycle. The system calculates DOH for every SKU daily using the most current on-hand quantities and demand data from the ERP, classifies each item against its target zone, and routes exceptions to the right person automatically. The weekly review meeting stops being an exercise in finding problems that already became emergencies and starts being a confirmation that the alert layer caught everything it should have.

The working capital connection: Days Inventory on Hand is one of the three components of the cash conversion cycle (DSO + DOH − DPO). Reducing DOH by 10 days across a $4.8M inventory base releases approximately $480,000 in cash. See the CFO reports framework for how DOH connects to the broader working capital dashboard your board wants to see.

How Inventory IQ surfaces SKU-level DOH from your ERP

Inventory IQ connects directly to your ERP, calculates days on hand for every stocked SKU using the appropriate demand denominator (forward orders where available, trailing consumption as a fallback), classifies each item against its target zone, and delivers the SKU-level view your operations and procurement teams need to act before problems become crises. For multi-plant environments, it runs the cross-plant comparison automatically once the item master is normalized across all sites.

Marquis IQ · Inventory IQ module
SKU-level DOH, slow-mover identification, and cross-plant transfer alerts from your existing ERP data.

Inventory IQ calculates days on hand for every active SKU daily, classifies each item against a configurable target zone based on its lead time and demand profile, quantifies the working capital cost of excess, and alerts the right buyer when any item crosses a threshold. In multi-plant environments, cross-site imbalances surface automatically once the item master is unified across ERP systems.

Daily DOH calculation for every SKU using forward demand and trailing consumption
Configurable zone thresholds per item based on lead time and demand variability
Cross-plant inventory imbalance detection with transfer opportunity quantification
IQ Insights alerts that route stockout risks and slow-mover flags to the right buyer automatically

FAQ

Common questions about inventory days on hand

Questions about calculation methods, target ranges, multi-plant analysis, and how DOH connects to working capital.

How is inventory days on hand calculated?

Inventory days on hand is calculated by dividing the quantity on hand by the average daily usage rate. The average daily usage is typically derived from one of three sources: trailing 13-week actual consumption, trailing 52-week consumption annualized to a daily rate, or forward demand from open production and customer orders. Trailing 13 weeks is most responsive to recent demand shifts but volatile for seasonal items. Trailing 52 weeks smooths seasonality but lags trend changes. Forward demand is most accurate for make-to-order environments where open orders represent committed future consumption. For most manufacturing environments, a blended approach using forward order demand where available and trailing 13-week consumption as a fallback produces the most actionable DOH signal.

What is a healthy inventory days on hand target for manufacturers?

There is no universal answer because the right DOH target varies by item type, supplier lead time, and demand variability. A reasonable starting framework for discrete manufacturers: items with supplier lead times under two weeks can target 15 to 30 days on hand. Items with lead times of four to eight weeks typically require 30 to 60 days on hand to maintain service levels. Items with lead times beyond eight weeks may require 60 to 90 days. Items above 90 days on hand should be reviewed regardless of lead time, because excess beyond this level typically indicates a demand change, a forecast error, or a purchasing pattern problem. The right approach is to set a target DOH range per item based on its lead time and demand variability rather than applying a single company-wide target.

What causes inventory days on hand to spike for a specific SKU?

The most common causes of DOH spikes at the SKU level are: a demand drop the purchasing team has not yet responded to, a minimum order quantity that creates periodic over-purchasing relative to actual demand, a forecast error at a recent planning cycle, a customer order cancellation that left committed inventory without a home, a product substitution or engineering change that made existing stock obsolete, or a safety stock policy set when demand was higher that has not been updated. In a multi-plant environment, a DOH spike at one plant sometimes reflects a transfer opportunity from a plant where stock is adequate rather than true system-level excess.

How is days on hand different from days of supply?

Days on hand and days of supply are often used interchangeably and the calculation is identical: on-hand quantity divided by average daily usage or demand rate. The terminology difference is usually contextual. Days on hand tends to be used for raw materials and components, describing how long current stock will last before a replenishment order is needed. Days of supply tends to be used for finished goods, describing how many days of forward customer demand the current stock can cover. In both cases, the metric is most useful at the SKU level rather than in aggregate, because aggregate figures blend short SKUs and long SKUs in a way that hides the actual risk and excess profile of the inventory.

How do multi-plant manufacturers manage inventory days on hand across sites?

Multi-plant manufacturers need two levels of DOH visibility: the individual plant level, which drives local purchasing and production decisions, and the cross-plant system level, which reveals transfer opportunities between sites. A SKU at stockout risk at one plant and with excess stock at a sister plant is not two separate problems. It is a transfer opportunity that requires a unified item master and a cross-plant DOH view to identify automatically. Without a unified item master, the same physical part appearing under different item codes at different plants is invisible as a transfer candidate, and both the excess and the shortage persist simultaneously while the system-level solution goes unseen.

See SKU-level inventory DOH running on your ERP data

We'll walk through what Inventory IQ surfaces from your ERP: slow movers, stockout risks, cross-plant imbalances, and the working capital release opportunity. Bring your ERP and 30 minutes.