How to Conduct a Spare Parts Audit and Fix Inventory Control Gaps

A technician comes to the storeroom window with a work order. The system shows a seal kit is in stock, but the bin is empty. To keep the repair moving, someone places an emergency order outside the normal reorder process, then adjusts the record later.

That single discrepancy can create more than a stockout. It can lead to duplicate purchases, emergency freight, maintenance delays, excess inventory, and a storeroom team that no longer trusts its own data.

A spare parts audit breaks that cycle by comparing physical inventory with the system of record, identifying why discrepancies exist, and fixing the workflows that let them return.

For maintenance organizations, the goal isn’t simply to make the inventory system match the shelves. It is to make sure the right parts are identified, available, accessible, and connected to the maintenance work that needs them.

This guide explains how to conduct a spare parts audit, determine what to count, classify inventory by risk, reconcile discrepancies, fix control gaps, and build a cycle-counting process that keeps accuracy from degrading again.

Spare parts audit checklist: At a glance

Before getting into the detailed process, use this checklist to structure the audit:

• Define the locations, part classes, and inventory types included in the audit.
• Identify critical, high-value, high-velocity, and high-risk spare parts.
• Decide whether you need a full physical count, targeted count, or risk-based sample.
• Establish a transaction cutoff and rules for receipts, issues, returns, and transfers during the count.
• Confirm the system-of-record baseline before you begin counting.
• Verify quantity, location, identity, condition, and availability status.
• Use blind counts where practical to reduce confirmation bias.
• Record every variance in a standardized variance log.
• Investigate high-impact and recurring variances instead of simply adjusting quantities.
• Separate physical availability from system on-hand quantity.
• Review min/max levels and reorder points after correcting inventory records.
• Assign owners, deadlines, and verification steps for control fixes.
• Connect parts consumption to maintenance work orders where possible.
• Establish risk-based cycle-count frequencies.
• Track inventory accuracy alongside maintenance and operational outcomes.

The detailed process below expands each step.

What a spare parts audit is and what it is not

A spare parts audit is an operational validation of spare parts inventory across several dimensions:

• Quantity: how many units physically exist.
• Location: where those units are physically stored.
• Identity: whether the physical part matches the item record and whether duplicate records exist.
• Condition: whether the part is serviceable, damaged, expired, or otherwise unsuitable for use.
• Status and availability: whether the part is available, reserved, quarantined, allocated, at a repair vendor, or otherwise unavailable for the maintenance work that needs it.

It can also surface questions about valuation, obsolescence, ownership, and traceability. If you expect to post inventory adjustments, write off obsolete stock, or change valuation treatment, involve finance early so approvals, cutoffs, and documentation requirements are clear.

What a spare parts audit is not

Not only a wall-to-wall count. The physical count provides evidence. The objective is to understand and correct the processes that created the discrepancy.

Not automatically a formal financial audit. An operational spare parts audit can improve inventory accuracy and maintenance execution. Financial reporting requirements may introduce additional controls around valuation, ownership, cutoff procedures, evidence, and approvals.

The distinction matters because a count that simply changes the system quantity to match the shelf can make the records look accurate without fixing the process that caused the error.

Step 1: Set the audit scope so you can finish it

Many storerooms cannot audit every stock-keeping unit (SKU) at once without disrupting maintenance. The first decision, therefore, is not “How do we count everything?” It is:

Which inventory creates the greatest risk if its quantity, location, identity, condition, or availability is wrong?

Start by defining the full audit population, then use risk to determine which inventory requires a full count and which can be sampled.

Define the audit population before counting

Define the audit population by ownership, inventory status, and physical location, not simply by what sits inside the storeroom.

Include every location where spare parts can be held, issued, staged, transferred, or awaiting disposition. Depending on the operation, this may include:

• Central and satellite storerooms
• Technician vehicles and mobile stock
• Tool cribs and point-of-use cabinets
• Staged maintenance kits, carts, and job boxes
• Production or work areas holding maintenance spares
• Project or site storage
• Repair-vendor locations where company-owned parts are held
• Receiving areas for inventory that has arrived but has not yet been put away
• Inventory in transit between locations

Before the count begins, document which locations are controlled, which are uncontrolled or informal, and which inventory populations are included or excluded.

Separate inventory by ownership and status

Do not automatically combine all physically present parts into one inventory total.

Identify at least:

• Company-owned inventory: Parts the organization owns and includes in its inventory records.
• Consignment inventory: Parts the organization physically holds but that a supplier or other third party owns. Reconcile these separately against the applicable ownership records or agreement.
• Company-owned parts at repair vendors: Reconcile these against dispatch records, vendor confirmations, repair status, and expected return dates.
• Third-party-owned stock: Identify separately and reconcile against the applicable consignment or ownership agreement.
• In-transit inventory: Parts shipped or transferred but not yet physically received at the destination.

Ownership, location, and status can overlap. A company-owned part may also be in transit or at a repair vendor; count it once and record each attribute separately.

Keep in-transit inventory as a separate category until you physically receive it. Establish a transaction cutoff for receipts, transfers, issues, and returns, then reconcile shipments and transfers against the relevant records to prevent double-counting or omission.

The objective is to establish one clearly defined audit population so that every relevant part is either counted, separately reconciled, or explicitly excluded with a documented reason.

If you operate across multiple sites, define an audit treatment for every in-scope location, whether it is fully counted, sampled, separately reconciled, or explicitly excluded.

Identify the inventory that carries the most risk

Once you define the population, identify the parts where an inaccurate record could have the greatest operational, financial, or supply-chain consequence.

Prioritize:

• Critical spares: Parts whose unavailability can stop production, create significant safety consequences, or block a critical repair.
• High-value items: Parts with significant unit cost or inventory value that warrant tighter financial and inventory controls.
• High-velocity items: Frequently issued parts and consumables that can quickly accumulate transaction errors.
• Long-lead-time items: Parts that take significant or uncertain time to procure or replace.
• Supply-constrained items: Parts with limited suppliers, difficult sourcing conditions, or few practical substitutes.
• Recurring-variance items: Parts that repeatedly show discrepancies in quantity, location, identity, or status.
• Known problem areas: Mixed bins, unlabeled shelves, unofficial overflow storage, and locations with repeated inventory errors.
• Multi-site inventory: Parts held at a site from which they cannot reach the maintenance job in time for the maintenance team that needs them.
• New or relocated inventory: Parts whose records or locations may not yet be stable.

Do not treat these factors as interchangeable. Criticality describes the consequence of a part being unavailable. Value describes financial significance. Demand or movement describes how frequently inventory changes. Lead time and sourcing difficulty describe supply risk. Historical variance indicates where existing controls may be unreliable.

Determine whether to use a full count or a sample

No universal percentage of inventory applies to every storeroom. The appropriate scope depends on the audit’s purpose, inventory risk, operational disruption, inventory complexity, and the reliability of existing records.

Use this three-tier approach:

Inventory profile	Recommended approach
Critical or otherwise defined as high-risk	Full count
High-value, long-lead-time, supply-constrained, or recurring-variance inventory	Targeted full count or expanded sample
Stable, lower-risk inventory with a reliable accuracy history	Risk-stratified sample

Full count: Count 100% of critical spares and other inventory where an inaccurate record could create significant safety, production, maintenance, or operational risk.

Targeted full count or expanded sample: Provide additional coverage for high-value parts, long-lead-time items, supply-constrained parts, recurring-variance items, and inventory tied to current maintenance bottlenecks. Depending on risk, this may mean counting the entire category or increasing its representation in the sample.

Risk-stratified sample: For stable, lower-risk inventory, select items across sites, storage locations, ABC classes, movement rates, inventory values, criticality levels, and historical variance. Avoid selecting only convenient bins because they are easier to count.

Set the sampling rules before you count

If the audit uses sampling, define the rules before the first count. Document:

• Which inventory populations are eligible for sampling
• Which items require a full count regardless of sampling
• How sampled items will be selected
• Which dimensions will be used for stratification, such as site, value, movement, criticality, and variance history
• The discrepancy tolerance that will trigger additional testing
• Who can approve a scope expansion

For formal audits that require quantified confidence and error tolerance, use an appropriate statistical sampling method rather than an arbitrary percentage.

For operational audits where formal statistical precision isn’t required, a risk-stratified sample can provide practical coverage without a wall-to-wall count.

Define when the sample must expand

Don’t treat a sample as sufficient just because you have counted the planned number of items.

Before starting, define the discrepancy level that will trigger additional testing. For example, an organization might set an operational rule that if the observed discrepancy rate exceeds its predefined tolerance, it expands the affected inventory group to a larger sample or a full count.

The threshold should reflect the organization’s risk tolerance, inventory value, criticality, and audit purpose. It is an internal control rule, not a universal industry standard.

Use this decision path:

Critical inventory → Full count

High value, long lead time, supply constraint, or recurring variance → Targeted full count or expanded sample

Stable, low-risk inventory → Risk-stratified sample

Observed discrepancy rate exceeds the predefined tolerance → Expand the affected sample or move to a full count

The objective is not to find a universal inventory percentage to count. It is to apply enough verification to provide reasonable confidence that the inventory records supporting maintenance decisions are reliable.

Classify criticality before counting

Not every spare part carries the same operational risk. Classify parts based on the consequences of unavailability when needed, not purchase price or usage alone.

A practical criticality assessment can consider:

Factor	Low	Medium	High
Consequence of unavailability	Minimal disruption	Local maintenance delay	Major production, operational, or service disruption
Safety impact	Little or no safety impact	Additional controls required	Significant safety consequences possible
Production or operational impact	Easily worked around	Reduced capacity or localized delay	Critical operation may stop
Lead time	Readily available	Moderate lead time	Long or uncertain lead time
Substitutability	Easy substitute available	Limited alternatives	No practical substitute
Sourcing difficulty	Multiple reliable sources	Limited suppliers	Specialized or difficult to source
Equipment dependency	Few assets depend on it	Several assets depend on it	Critical equipment depends on it

This framework is illustrative, not a universal industry standard. Organizations should define their own scoring thresholds based on equipment criticality, operating risk, supply conditions, and maintenance requirements.

Turn the assessment into a practical score

For a simple internal framework, assign each factor a score from 1 to 3:

• 1 = Low
• 2 = Medium
• 3 = High

Assess operational criticality separately from supply risk. Rate safety and operational consequences first, then assess lead time, substitutability, and sourcing difficulty. Use both assessments to set inventory control priorities. The organization can then define its own bands for low-, medium-, and high-criticality parts.

For example, consider a specialized hydraulic component used on a critical excavator:

Factor	Score
Consequence of unavailability	3
Safety impact	3
Production/operational impact	3
Lead time	3
Substitutability	3
Sourcing difficulty	2
Equipment dependency	3
Total	20/21

A score this high indicates that the component deserves high-priority inventory controls, such as tighter count frequency, verified storage location, clear availability status, and closer monitoring of replenishment risk.

The organization sets the exact score bands. The important point is to make the classification repeatable and document why a part receives its assigned criticality.

Keep demand frequency separate

Failure or demand frequency helps with inventory planning, but it should not automatically determine criticality.

A frequently consumed part may need higher stock levels because demand is high. A rarely used part may still be highly critical if its absence can stop a critical asset and replacement takes months.

Use demand or failure frequency as a planning modifier alongside criticality:

• High criticality + low demand: Consider controlled strategic stock and tighter availability checks.
• High criticality + high demand: Combine tight controls with appropriate replenishment planning.
• Low criticality + high demand: Prioritize stock availability based on consumption without automatically treating the part as operationally critical.
• Low criticality + low demand: Simpler controls may be appropriate where the consequences of failure are limited.

This distinction prevents inventory value, consumption, and operational criticality from being treated as the same thing.

Use ABC and risk-based classification

ABC analysis and criticality classification answer different questions, so treat them as separate dimensions.

ABC analysis typically ranks inventory using a quantitative measure such as annual consumption value, calculated by multiplying annual usage by unit cost. The organization then sets its own thresholds for A, B, and C categories based on its inventory distribution.

For example:

• A items: Represent the highest share of annual consumption value and typically warrant tighter inventory controls.
• B items: Fall into the middle range of annual consumption value and generally receive standard controls.
• C items: Represent the lower share of annual consumption value and may be managed with simpler controls where operational risk is also low.

Do not treat these categories as a measure of operational importance. A low-cost C item can still be critical if its absence can stop equipment or create a safety concern, or if replenishment has a long and uncertain lead time. 

Layer ABC classification with separate factors such as:

• Criticality
• Lead time
• Sourcing difficulty
• Availability of substitutes
• Equipment dependency
• Historical inventory variance

The result is more useful than value alone. ABC analysis shows where inventory value is concentrated, while criticality and risk show where an inventory error could cause the most operational damage.

For example, a $20 seal used frequently across a fleet may be an A item because its annual consumption value is high. A $2,000 specialized component used only once every few years may be a C item by consumption value but still require tight control because replacing it takes six months and its absence can keep critical equipment offline.

The key distinction is that ABC prioritizes controls by inventory value, while criticality prioritizes them by operational consequence.

Step 2: Lock down the baseline so the count stays trustworthy

A physical count is a snapshot. If inventory keeps moving while you count, the snapshot becomes difficult to reconcile.

Before starting, define the operating rules for the audit window.

Establish a transaction cutoff

Decide how you will handle:

• Receipts
• Withdrawals
• Returns
• Transfers
• Scrap
• Emergency issues
• Parts being moved between bins
• Inventory arriving during the count

Document when the cutoff begins and ends. Save a timestamped baseline of recorded quantities by SKU, location, and unit of measure. Record each count’s time and reconcile intervening receipts, issues, returns, and transfers to the same cutoff.

Keep emergency maintenance moving

An audit should not prevent a critical repair.

If parts must be issued during the count, use a controlled process such as an emergency issue slip or transaction log. Record the issue when it happens and reconcile it within a defined deadline.

The goal is not to freeze operations. It is to prevent emergency work from becoming an undocumented exception.

Assign clear roles

At minimum, define:

1. Counters
2. Reviewer/reconciler
3. Inventory or master-data owner

Where practical, avoid having someone count the bins they normally manage. Separation of duties can reduce confirmation bias and make discrepancies easier to challenge.

Confirm the location hierarchy

Use a consistent structure such as:

Site → storeroom → aisle → shelf → bin

If unofficial locations exist, document them instead of pretending they don’t.

Parts on an unrecorded overflow shelf are still inventory. 

Step 3: Count to expose quantity, location, identity, and status errors

Storeroom drift is rarely only a math problem.

The right part may be in the wrong bin. The wrong part may be in the right bin. Two item records may represent the same physical part. A physically present part may also be damaged, expired, reserved, or awaiting inspection.

Use both counting directions where practical 

Trace selected records to physical stock to test existence, and trace selected physical stock back to records to test completeness.

Location-first/bin-to-item

Start with a physical location and record what is actually there.

This works well when:

• Location records are unreliable
• Unofficial storage is common
• Parts are frequently misplaced
• You suspect duplicate or unidentified inventory

Item-first/item-to-bin

Start with the system item list and verify each item’s physical location.

This works better when:

• Location data is mature
• Bin structures are controlled
• You are validating compliance with an established inventory structure

Build controls into the count

Use blind counts where possible. Counters should record what they see without first viewing the system quantity. Compare the physical result to the system during reconciliation.

Standardize the unit of measure. Make sure “each,” “box,” “pack,” “case,” and other units are defined consistently.

Flag unknowns instead of forcing a match. If you can’t confidently identify a part, record an identity exception.

Define recount triggers. Require recounts for missing critical spares, large discrepancies, unidentified parts, or other predefined exceptions.

Don’t count “on hand” as automatically “available”

This distinction is critical.

Imagine the system reports:

• 10 units on hand
• 4 reserved for upcoming work
• 2 quarantined
• 1 awaiting inspection
• 3 available

The storeroom has 10 physical units, but only 3 are currently available for new maintenance demand.

An audit should therefore distinguish physical quantity, condition, allocation status, and available-to-issue quantity.

Inventory accuracy vs. Available-to-issue share 

A spare parts record can be numerically accurate and still fail maintenance.

Suppose the system says a storeroom has five replacement motors. You physically verify five motors, so the recorded quantity is accurate.

But if:

• Two are reserved for existing work orders
• One is damaged
• One is awaiting inspection
• One is serviceable and available to issue

then the maintenance team does not have five motors available for new demand.

Before measuring availability, define the inventory states clearly:

Inventory state	What it means
Book quantity	Quantity recorded in the inventory system
Physical quantity	Quantity physically verified during the count
Serviceable quantity	Physically present inventory that meets the organization’s condition requirements for use
Reserved or allocated quantity	Serviceable inventory committed to a specific work order, project, or other demand
Available-to-issue quantity	Serviceable inventory that is not already allocated and can be issued to new demand
Blocked or quarantined quantity	Inventory physically present but temporarily unavailable because it requires inspection, disposition, or another control action
In-transit quantity	Inventory that has been shipped or transferred but has not yet been physically received at the destination

This distinction matters because physically present does not always mean available for new maintenance demands.

Do not automatically treat reserved inventory as unusable. A reserved part may be fully serviceable and ready for the work order it was allocated to, but it is not available for a new requirement.

Measure usable availability separately

A practical operational measure is:

Usable availability = units meeting the defined available-to-issue criteria ÷ total physically verified units x 100

Using the example above:

1 available-to-issue unit ÷ 5 physically verified units = 20% available-to-issue share 

The organization should define the exact meaning of “available-to-issue.” It may require the part to be physically present, correctly identified, serviceable, accessible, and not already allocated to another demand.

Keep this measure separate from inventory accuracy. An organization can have 100% quantity accuracy but only 20% available-to-issue inventory if most of its stock is reserved, damaged, awaiting inspection, quarantined, or otherwise unavailable for new demand.

That distinction helps maintenance leaders avoid a common mistake: assuming that an accurate inventory record automatically means the required part will be available when a technician needs it.

Step 4: Reconcile discrepancies with a variance log, then diagnose the cause

Reconciliation turns physical findings into corrective action.

Avoid ad hoc adjustments made bin by bin without documenting what changed and why.

Your variance log should capture:

• Item ID
• Manufacturer part number, where available
• Recorded quantity
• Counted quantity
• Unit of measure and conversion factor
• Baseline and count timestamps
• Recount result
• Counter and reviewer
• Adjustment reference and approval
• Recorded location
• Found location
• Site and storeroom
• Variance type
• Condition/status
• Initial reason code
• Owner
• Next action
• Approval, where required

Use a consistent variance taxonomy

Variance type	What it usually means	What to check first
Quantity	More or less stock than expected	Receiving, withdrawals, returns, scrap
Location	Stock exists but is elsewhere	Putaway, transfers, point-of-use storage
Identity	Physical part doesn’t match record	Labels, duplicates, substitutions, kitting
UOM	Quantity is represented differently	Pack sizes, purchasing UOM, issue UOM
Status/availability	Part exists but cannot be issued to the intended demand.	Quarantine, damage, inspection, reservation
Condition	Part exists but is not serviceable	Damage, expiration, storage conditions
Ownership	Physical stock belongs to another party	Consignment or vendor-owned stock

Don’t simply adjust the system to match the shelf

This is one of the most important audit controls.

If the system says 10 and the count says 8, the first question should not be:

“How do we change 10 to 8?”

It should be:

“Why are two units missing?”

Possible explanations include:

• Unrecorded issue
• Incorrect receipt
• Misplaced stock
• Scrap not transacted
• A return recorded before the part was physically received 
• Transfer not recorded
• Duplicate record
• Parts consumed at point of use

Validate the discrepancy through recounts and transaction checks before posting an approved adjustment. If the root cause remains unresolved, correct the verified quantity with an exception record and keep the investigation open. 

Add root-cause analysis for high-impact variances

Your variance log tells you what is wrong. Root-cause analysis tells you why it keeps happening.

Prioritize:

• Critical-spare discrepancies
• Repeated variance items
• Large financial exposure: Discrepancies involving significant inventory value, potential write-offs, or material financial impact.
• Variances that block maintenance
• Errors occurring across multiple locations
• Errors indicating a broken workflow

A simple 5 Whys analysis might look like this:

• Problem: A critical spare is repeatedly found in the wrong bin.
• Why? Putaway is not consistently recorded.
• Why? Employees place received or returned parts in temporary overflow locations when the assigned bin is full.
• Why? Those temporary locations are not represented in the inventory location structure.
• Why? No formal process exists to request and approve new or overflow inventory locations.
• Why? The location hierarchy is not governed through a defined inventory-control or management-of-change process.
• Root cause: The organization has no controlled process to create and govern inventory locations as storage needs change.

The corrective action should address both the immediate problem and the underlying control:

• Create controlled overflow locations and include them in the inventory location hierarchy.
• Require formal requests, approval, labeling, and system entry for every new or changed location.
• Assign an owner for location master data and review location changes periodically.

The goal of the 5 Whys is not to force exactly five questions. Keep asking why until the analysis reaches a controllable underlying cause, rather than stopping at the first visible symptom.

How to prioritize audit findings

A large audit can produce hundreds of discrepancies. Not all of them deserve the same response.

Use a simple prioritization model based on operational impact, recurrence, and financial impact.

Finding	Operational impact	Recurrence	Suggested priority
Critical spare missing	High	Any	Immediate
Critical spare mislocated	High	Any	Immediate
High-value quantity discrepancy	Medium–High	Any	High
Repeated UOM error	Medium	Repeated	High
Duplicate low-value SKU	Low	One-off	Low–Medium
Obsolete low-value item	Low	One-off	Low
Point-of-use stock not recorded	Medium–High	Repeated	High

The exact thresholds should be defined by your organization, but the principle is simple:

Prioritize discrepancies by the operational consequences they create, not just the number of units involved.

What a variance log looks like in practice

A variance log does not need to be complicated.

Item ID	Mfr. part #	Recorded qty	Counted qty	Recorded bin	Found bin	Variance	Reason	Owner	Next action
SP-1042	ABC-778	10	8	A-02-04	A-02-04	Quantity	Suspected unrecorded withdrawal	Storeroom lead	Investigate WO issues and adjust after approval
SP-1187	DEF-421	4	4	B-01-02	B-03-01	Location	Mis-putaway	Storeroom lead	Correct bin and review putaway process
SP-2011	XYZ-991	6	6	C-02-01	C-02-01	Status	Quarantined	Maintenance lead	Update availability status

Use actual item IDs and location names in the working version. Consistency matters more than complexity.

Step 5: Fix the inventory control gaps that create drift

Once you’ve categorized discrepancies and identified high-impact causes, implement corrective actions for the workflow gaps you found. 

Where drift starts	What it looks like	Control to implement
Receiving and putaway	Received stock is not placed in the recorded bin	Requires bin assignment and controlled putaway
Parts withdrawal	Inventory falls without traceable consumption	Link issues to work orders where possible
Returns	Returned parts never re-enter inventory	Record return, condition, status, and location
Transfers	Stock appears at the wrong site	Record outbound and inbound movement
Repairables/rotables	Parts disappear while at vendors	Track repair status, location, and expected return
Item master	Duplicate records or mismatched UOM	Establish master-data ownership and change control
Point-of-use inventory	Technicians hold unrecorded stock	Treat point-of-use areas as controlled locations
Obsolete/excess inventory	Stock remains after asset retirement	Review and route disposition decisions

Your audit deliverable should include both:

1. Corrected inventory records
2. Control changes with owners, deadlines, and verification steps

Otherwise, the audit becomes a bulk adjustment rather than a control improvement.

Step 6: Connect spare parts to maintenance work orders

Inventory accuracy becomes much more useful when it is connected to the maintenance activity consuming the inventory.

The operational chain should look like this:

Asset → Work order → Required part → Issue → Inventory update → Maintenance completion → Asset history

Consider a pump requiring a replacement seal.

Without a connected workflow:

1. Technician identifies the part.
2. Storeroom issues the seal.
3. Someone manually adjusts inventory later.
4. The work order may not show the actual part consumed.
5. The asset history may not clearly capture the replacement.

With a connected workflow:

1. The work order identifies the asset.
2. Required or used parts are associated with the work.
3. The system issues the part from a known location.
4. Inventory quantity updates as the transaction occurs.
5. The maintenance record retains the part consumption.
6. The asset’s service history provides context for future work.

That connection creates a much stronger audit trail.

When teams also record failure codes, parts-related delay reasons, and purchasing exceptions, linked records can help answer questions such as: 

• Which assets consume the most spare parts?
• Which parts are repeatedly used for the same failure?
• Which work orders are delayed because a part is unavailable?
• Which assets are driving emergency purchases?
• Which spare parts should be stocked at particular sites?
• Are parts being consumed against the correct assets and work orders?

This is where spare parts management moves from a storeroom activity to part of the maintenance operating system.

Step 7: Review stocking thresholds and replenishment rules

An accurate count tells you what you have. It doesn’t tell you whether the quantity matches future maintenance demand.

After reconciliation, review:

• Historical consumption
• Demand variability
• Equipment criticality
• Lead times
• Supplier reliability
• Minimum and maximum levels
• Reorder points
• Safety-stock requirements
• Current maintenance plans

However, spare-parts demand is often intermittent rather than steady. A part may be used several times in one month and not at all for the next six months. This makes conventional stocking formulas less reliable when their assumptions don’t match the actual demand pattern.

For high-usage parts with relatively stable demand, conventional reorder-point or safety-stock methods may provide a useful starting point. For slow-moving or intermittent parts, teams should evaluate the assumptions behind those calculations rather than treating the resulting number as automatically appropriate.

For low-usage, high-criticality parts, the key question may be whether to hold the part at all. If a missing component could keep critical equipment offline for weeks or months, the cost of carrying an infrequently used spare may be justified by the potential cost of downtime. Conversely, holding an expensive part with a low likelihood of use may not make sense when a practical substitute or reliable expedited supply option exists.

For these parts, consider factors such as:

• Consequence of equipment downtime
• Probability and pattern of demand
• Part cost
• Lead time and lead-time variability
• Supplier reliability
• Availability of substitutes
• Equipment population and commonality
• Remaining useful life of the equipment

When standard formulas don’t adequately represent these conditions, organizations can use criticality-based, probabilistic, or other intermittent-demand approaches to inform stocking decisions. The appropriate method depends on the quality of the available demand and supply data.

The goal is not to maximize spare-parts inventory. The goal is to hold enough inventory to manage the operational risk of unavailability without creating unnecessary carrying costs or obsolete stock.

Review replenishment decisions after the audit

A recurring variance may reveal that the stocking policy itself needs attention.

For example, if a critical component repeatedly reaches zero before replenishment arrives, review whether the issue comes from inaccurate records, excessive demand, insufficient lead-time assumptions, an inappropriate reorder point, or a deliberate decision not to hold enough stock.

Do not automatically increase the stocking level. First determine whether the problem is inventory accuracy, demand, supply risk, or the stocking policy itself.

For critical spares, appropriate controls may include higher safety stock, a second supplier, a defined emergency-procurement path, a substitute part, or a decision to hold a dedicated unit. The right response depends on the operational consequence of being without the part.

Step 8: Keep accuracy from decaying with cycle counting

A one-time audit establishes a baseline. It does not keep the baseline clean.

After reconciliation, move to risk-based cycle counting.

Count high-risk inventory more frequently

Prioritize:

• Critical spares
• High-velocity items
• High-value items
• Repeat variance items
• Items with long lead times
• Inventory at high-risk locations

Count stable inventory less frequently

Lower-frequency counts may be appropriate for:

• Slow-moving items
• Low-risk inventory
• Stable locations
• Items with a strong history of accuracy

Document:

• Count schedules
• Variance tolerances
• Recount triggers
• Reason codes
• Escalation rules
• Ownership
• Closed-loop corrective action

A well-run cycle-counting program can reduce the need for disruptive wall-to-wall counts in some operating models, subject to applicable financial or external audit requirements.

Maintenance KPIs affected by spare parts accuracy

Don’t measure inventory accuracy only as a storeroom KPI.

Maintenance teams care about spare parts accuracy because inventory problems can affect equipment availability and work execution.

Track inventory measures alongside maintenance outcomes such as:

Maintenance delays caused by unavailable parts

Measure how often planned or corrective work is delayed because a required part cannot be found or issued.

Mean time to repair

If technicians spend less time waiting for parts or searching for inventory, spare parts controls can shorten repair cycles. Define whether your repair-time metric includes waiting for parts. Track parts-related waiting time separately so you can distinguish changes in active repair time from total restoration time.

Stockout-related downtime

Identify downtime events where the root cause includes an unavailable, mislocated, unidentified, or unusable spare.

Emergency purchases

Track purchases made outside normal replenishment because the expected inventory was unavailable.

Expedite and emergency freight costs

An inaccurate inventory record can turn a normal replenishment into an expensive expedited shipment.

First-time fix performance

Where appropriate, track whether technicians have the required parts when they start work.

Maintenance schedule compliance

If planned maintenance is repeatedly delayed because parts are unavailable, inventory control may be part of the underlying problem.

The goal isn’t to claim that better inventory accuracy automatically improves every maintenance KPI. Instead, connect the measures so you can identify where inventory failures affect maintenance execution.

Step 9: Measure whether the audit is improving performance

Once cycle counting is established, use metrics to determine whether accuracy is improving and whether the same problems are returning.

Define each metric clearly so the calculation cannot change from one reporting period to another.

Inventory line accuracy

Formula:

Number of counted inventory lines within the defined tolerance ÷ total inventory lines counted x 100

This measures inventory line accuracy: the percentage of counted item-location lines whose physical quantity falls within the organization’s predefined tolerance of the recorded quantity.

A line-level measure treats each counted line as one observation. It therefore does not show the magnitude or financial value of each discrepancy. A one-unit shortage and a 5,000-unit shortage can both count as one inaccurate line.

Depending on the purpose of the audit, organizations may also track:

• Quantity variance: counted quantity minus recorded quantity, expressed in the item’s defined unit of measure. 
• Inventory value variance: quantity discrepancies valued using an agreed unit-cost basis. 
• Absolute variance: the total magnitude of discrepancies, without letting positive and negative variances offset each other.

Use the measure that matches the question you are trying to answer. Line accuracy helps evaluate record reliability across inventory locations, while unit and value measures show the size and financial significance of discrepancies.

Location accuracy

Formula:

Lines found in the correct recorded bin ÷ total lines counted

A line refers to an SKU-location record.

Usable availability

Formula:

Units meeting the defined available-to-issue criteria ÷ total physically verified units

Define “available-to-issue” according to your organization’s inventory controls. It may require the part to be physically present, correctly identified, serviceable, accessible, and not already allocated to another demand.

Do not automatically treat reserved inventory as unusable. A reserved part can be fully serviceable and committed to a specific work order while remaining unavailable for new maintenance demand.

Track usable availability separately from inventory accuracy. This helps distinguish a quantity-record problem from a situation where inventory exists but maintenance can’t issue it when needed.

Stockouts by cause

Separate:

• True stock shortages
• Record errors
• Mislocated inventory
• Incorrect status
• Unusable condition

This distinction helps identify whether the problem is purchasing, inventory control, or maintenance planning.

Cycle-count completion rate

Formula:

Items or bins counted as planned ÷ items or bins scheduled

Recurring variance tracking

Track how frequently the same items, locations, or processes generate discrepancies.

Maintenance impact

Where data is available, also track:

• Work orders delayed by parts
• Stockout-related downtime
• Emergency purchases
• Expedite costs
• Repair delays caused by inventory issues

Segment metrics by site, part class, criticality, and usage behavior. Aggregate metrics can hide the risks that matter most.

Where EAM or CMMS software can help

Your system of record may be an enterprise asset management (EAM) system, computerized maintenance management system (CMMS), ERP inventory module, or a combination of systems.

The key question isn’t whether the software stores inventory quantities.

Ask whether it links inventory transactions to the maintenance workflow.

Where EZO EAM can support spare parts control

Disclosure: EZO publishes this article and develops EZO EAM.

For maintenance and storeroom teams, EZO EAM can support several of the controls described in this audit process:

• Track spare parts and consumables by location so teams can see stock quantities across sites and storerooms.
• Configure low-stock and location-based thresholds and enable the corresponding alerts to flag quantities below defined levels.
• Transfer inventory between locations while maintaining a record of the movement.
• See available inventory at a technician’s location when adding inventory to a maintenance work order.
• Record parts used against maintenance work orders, connecting inventory consumption with maintenance activity.
• Link purchase orders to open work orders so teams can track maintenance-related procurement. 
• Combine inventory, purchasing, vendor, receipt, and usage information to give maintenance and operations teams more context on parts availability and replenishment.

These capabilities can connect spare-parts records with maintenance execution, but they don’t replace audit controls. Teams still need defined counting procedures, ownership, variance tolerances, criticality rules, and corrective-action processes.

Note: Work order functionality requires an eligible CMMS-enabled subscription. Inventory, purchasing, and transfer capabilities also depend on plan and configuration.

Use the audit as a software evaluation checklist

The following questions are vendor-neutral evaluation criteria, not a list of EZO EAM capabilities. When evaluating any EAM or CMMS platform, ask the vendor to demonstrate how the system handles:

• Finding a part by site and storage location
• Verifying current stock and availability
• Recording a part withdrawal against a maintenance work order
• Updating inventory after a part is consumed
• Recording transfers between locations
• Handling returned, damaged, quarantined, or otherwise unavailable inventory
• Maintaining traceability between parts, work orders, and assets
• Reporting on inventory discrepancies and parts-related maintenance delays

This distinction matters because inventory-control requirements and software capabilities are not the same thing. A process may require a specific control even when a platform doesn’t automate it natively.

Close the audit with a control check

Before closing the audit, confirm that the findings have been converted into corrective actions rather than simply inventory adjustments.

Use this short closeout checklist:

• Reconciled: Investigate and resolve discrepancies in material quantity, location, identity, condition, and status.
• Root cause identified: Significant or recurring variances have an identified underlying cause.
• Ownership assigned: Each control gap has an owner and a target completion date.
• Records corrected: Inventory, location, unit-of-measure, and ownership records have been updated and approved. 
• Controls strengthened: Implemented required process, system, labeling, training, or governance changes. 
• Follow-up defined: High-risk findings have a scheduled recount, cycle count, or control review.
• Results measured: Monitor accuracy, availability, stockouts, and maintenance-related outcomes after corrective actions.

The audit is ready to close when the organization can explain the material discrepancies, document what changed, and show how the new controls address the causes and how their effectiveness will be verified. 

Closing the loop

A spare parts audit creates value when it changes what happens after the count. The most important findings should lead to corrected records, stronger controls, assigned ownership, and follow-up checks that verify the problem has actually been addressed.

The goal is not a perfect count once a year. It is a controlled inventory process that gives maintenance teams confidence they can identify, locate, and issue the parts they need when required.

That is where connected asset and maintenance workflows can help. When spare-parts records are tied to maintenance work, asset history, purchasing, and inventory movements, teams have more context to understand both inventory discrepancies and their operational impact.