The Problem With Most LP Measurement
Loss prevention tends to attract one of two measurement problems. Either nobody's tracking anything — the business has a vague sense that "shrinkage is probably around 2%" based on a count someone did a year ago — or there's a dashboard full of metrics that generates a report every Monday that nobody reads.
Neither situation is actually measuring anything. The first has no data. The second has data with no attention. What makes a KPI useful isn't the number itself — it's whether someone is looking at it regularly, understanding what it means, and doing something different when it moves.
This guide covers twelve loss prevention KPIs across five categories. Each one has a formula, a benchmark, and — importantly — a note on what a bad number is actually signaling, because that's the part most metrics articles leave out. The signal matters more than the target.
Don't try to track all twelve at once. Pick three or four that address your biggest current gap. Get those running consistently. Add more once the first ones are giving you clean, reliable data. A few KPIs reviewed weekly beat twelve KPIs reviewed never.
Category 1 — Shrinkage Metrics
These are the headline metrics — the ones that give you the overall picture of how much stock or value is leaving your business unaccounted for.
This is your primary LP metric — the one everything else feeds into. It measures how much inventory value has gone missing relative to what you sold. It's the number most US industry benchmarks use, which makes it the most useful for external comparison.
Shrinkage % = (Dollar Value of Shrinkage ÷ Net Sales) × 100
What the signal means: A rising trend over three or more periods is more important than any single number. A rate that jumps suddenly in one period suggests a specific event — new staff, a process change, a timing error in a count. Gradual creep over many periods suggests something systemic.
This version measures what percentage of your actual stock value has disappeared. It's particularly useful for understanding the scale of loss relative to what you hold — and for category-level analysis where you don't have a corresponding sales figure to divide by.
Shrinkage % = ((Book Inventory − Physical Count) ÷ Book Inventory) × 100
What the signal means: Use this at category or SKU level to find where shrinkage concentrates. A store-level rate of 2% masking a 12% rate in one category is a much more specific and actionable problem than the blended number suggests.
A single shrinkage percentage tells you where you stand. The trend tells you which direction you're heading. This metric compares your current period's shrinkage rate against the same period last year or the prior period, giving you a directional signal that's more useful than any single reading.
Trend = ((Current Shrink % − Prior Shrink %) ÷ Prior Shrink %) × 100
What the signal means: A business whose shrink rate stays flat at 2% is not making progress. A business that moves from 2.8% to 2.4% to 2.1% is. Track direction, not just the absolute number.
Category 2 — Cash Handling Metrics
Cash is the easiest thing to lose and the hardest to trace back to a cause. These metrics give you visibility over cash handling without requiring an investigation every time there's a discrepancy.
This measures the frequency and size of gaps between the cash that should be in the till and what's actually there at reconciliation. It's one of the most sensitive indicators of cash handling problems — small, consistent discrepancies on specific shifts are often more telling than a single large gap.
Discrepancy Rate = (Total Cash Variance ÷ Total Cash Handled) × 100
What the signal means: Pattern matters more than total amount. Small consistent shortfalls on the same shift, same day of the week, or same cashier are the signature of deliberate taking — not random error. Track this by cashier and by shift, not just as a store-wide total.
This tracks the percentage of cash handling events — drops, reconciliations, deposits — where the required dual-control or documentation procedure was followed correctly. A location can have zero cash discrepancies and still be one unsupervised till away from a problem.
Compliance Rate = (Correctly Completed Cash Events ÷ Total Cash Events) × 100
What the signal means: Even one unchecked cash handling event is a gap. Low compliance rates on this metric often explain high cash discrepancy rates — the gap between them is where opportunity lives.
Category 3 — Employee Transaction Metrics
Internal theft almost always leaves a pattern in transaction data before it's ever detected through other means. These metrics make that pattern visible.
This measures how many refund transactions each employee processes compared to their peers. Returns are one of the most exploited transaction types in retail — phantom returns, inflated refund amounts, and returns-without-items are all common and all show up as outliers in this metric before they show up anywhere else.
Return Rate = (# Returns by Employee ÷ # Total Transactions by Employee) × 100
What the signal means: A cashier processing three times the team's average refund volume isn't necessarily stealing — maybe they work a different shift type or serve more difficult customers. But it's a flag worth reviewing. Pull their refund records and spot-check against CCTV. Most of the time it's nothing. Occasionally it's everything.
Voids and no-sales are the other high-risk transaction types. A void after a customer has paid and left is a classic method for pocketing the cash without the system flagging a refund. High void rates concentrated on one employee or one time period — especially when correlated with cash discrepancies — are one of the clearest signals in retail LP.
Void Rate = (# Voids by Employee ÷ # Total Transactions by Employee) × 100
What the signal means: A high void rate paired with a high cash discrepancy on the same shifts is as close to a confirmed signal as transaction data gets without CCTV review. This combination should always trigger a footage check on the specific shift days showing the pattern.
Sweethearting — giving friends and family unauthorized discounts or free items — shows up as a lower-than-average transaction value for the employee involved, especially during specific time windows. This metric is best compared across employees who work similar shifts and serve similar customer types.
Avg Tx Value = Total Sales by Employee ÷ # Transactions by Employee
What the signal means: Differences in transaction value across cashiers can have legitimate explanations — different product sections, different shift periods. Look for the pattern within a comparable group, or look at how one employee's average changes when specific customers visit.
Category 4 — Audit & Compliance Metrics
These metrics tell you how well your controls are being followed — not whether loss is happening, but whether the systems designed to prevent it are actually running.
An LP audit score measures how well a location is complying with LP policies and procedures at the time of audit. It's a point-in-time snapshot, not a continuous measure — but tracked over multiple audits it becomes a trend that reveals whether a location is improving, plateauing, or drifting. Critical finding categories should be weighted separately from minor ones.
Audit Score = (Items Passed ÷ Total Items Audited) × 100
What the signal means: A 92% score with a critical finding (e.g., no dual cash control) is not a 92% score in practice — it's a location with a high-risk gap. Always look at what failed, not just the aggregate number. Trend is more useful than any single audit score.
An LP audit that produces findings that are never acted on is a waste of time. This metric tracks what percentage of audit findings are resolved within the agreed timeframe. It's a measure of follow-through, not just of compliance — and it's one of the most important indicators of whether an LP program has any real teeth.
Close Rate = (Findings Resolved by Deadline ÷ Total Findings) × 100
What the signal means: A low close rate usually means one of two things: the deadlines are unrealistic, or there's no accountability for missing them. Either one undermines the entire audit process. This metric should be reviewed by senior management, not just the LP team.
Category 5 — Vendor & Receiving Metrics
Supplier-side losses are consistently underestimated by most businesses because they're rarely tracked directly. These metrics shine a light on the delivery and procurement gaps that turn into shrinkage.
This measures how often deliveries match the quantity and specification on the purchase order. Businesses that don't track this often mistake receiving discrepancies for shrinkage — spending time investigating theft that was actually a supplier short-delivery that nobody caught at the dock.
Receiving Accuracy = (Deliveries Matching PO ÷ Total Deliveries) × 100
What the signal means: A low rate from one specific vendor is a supplier problem — challenge them and adjust future orders. A low rate across multiple vendors usually means your receiving process isn't checking properly. Track accuracy by vendor to separate the two causes.
This tracks how often supplier invoices don't match the agreed purchase order price or quantity — catching overbilling and pricing drift before payment is released. Three-way matching (PO + delivery receipt + invoice) is the process; this KPI measures how well it's working.
Discrepancy Rate = (Invoices with Discrepancies ÷ Total Invoices) × 100
What the signal means: Discrepancies that always go in the vendor's favor — consistently above PO price, consistently above agreed quantity — aren't errors. They're a pattern worth challenging formally. A high overall rate across all vendors suggests your PO process or three-way matching isn't working.
All 12 KPIs at a Glance
| # | KPI | Category | Priority | Review Frequency |
|---|---|---|---|---|
| 01 | Shrinkage Rate (% of Sales) | Shrinkage | Critical | Monthly / per count |
| 02 | Shrinkage Rate (% of Inventory) | Shrinkage | High | Per count |
| 03 | Shrinkage Trend | Shrinkage | High | Quarterly |
| 04 | Cash Discrepancy Rate | Cash | Critical | Daily / per shift |
| 05 | Safe Compliance Rate | Cash | High | Weekly |
| 06 | Return / Refund Rate per Employee | Staff | Critical | Monthly |
| 07 | Void / No-Sale Rate per Employee | Staff | Critical | Monthly |
| 08 | Avg Transaction Value by Cashier | Staff | High | Monthly |
| 09 | LP Audit Score | Audit | High | Per audit |
| 10 | Audit Finding Close Rate | Audit | High | Per audit cycle |
| 11 | Receiving Accuracy Rate | Vendor | Critical | Weekly |
| 12 | Vendor Invoice Discrepancy Rate | Vendor | High | Monthly |
Start With Three — Not Twelve
If you're starting from scratch, the most useful thing you can do is pick three KPIs from the Critical column, measure them consistently for 60 days, and build from there. Data that's a month old and actually being looked at is worth more than twelve metrics being ignored.
The three I'd suggest for most businesses starting out: Shrinkage Rate (gives you the headline number), Cash Discrepancy Rate (gives you daily visibility on your highest-risk asset), and Return Rate per Employee (gives you early warning on the most common form of internal theft). Those three together cover the majority of retail loss exposure.
"A KPI that's reviewed every week and acted on is worth twenty that sit in a report nobody opens."
— PreventLoss.orgBlock 30 minutes every Monday morning for your three critical KPIs. That's it. Trend lines, outliers, anything that moved in the wrong direction. Ask why, and write down what you're going to do about it. Four weeks of that consistently beats a quarterly deep-dive that happens once and then gets forgotten.
Want to Dig Deeper Into Loss Prevention?
Explore our full library of guides on audits, shrinkage, vendor controls, and cost management — all free, no sign-up needed.
Browse All Articles →