awareretailgroup.com

How Do You Prioritize Loss Prevention Audit Findings?

A loss prevention audit finding is a specific gap the walkthrough turns up, like a backroom door that won’t latch or a cash office running overrides with no second signature. Prioritizing findings means weighing each fix’s cost against the loss it realistically prevents, scoring findings that depend on an associate’s behavior toward another person in a separate top tier, and giving every fix a named owner, a due date and a re-check.

How do you rank people-safety findings that don’t have a dollar value?

Put every finding that depends on an associate’s behavior toward another person into its own top tier, and score it by severity and likelihood instead of dollars. NIST’s computer security handbook offers a parallel from outside retail: security should be proportionate “to the severity, probability and extent of potential harm” (NIST SP 800-12). A dead camera and a policy telling cashiers to physically stop a shoplifter both look free on a cost sheet, and they belong nowhere near the same tier. The British Retail Consortium found that theft and violence have become increasingly interlinked, “with confronting thieves a major trigger of violence and abuse.” NIOSH’s guidance for gas stations and convenience stores reports that active employee resistance to a robbery is significantly associated with employee injury (NIOSH). That research covers robberies, not shoplifting stops, but it points the same way as the BRC finding. NIOSH’s guidance also lists a history of robbery or other violent incidents, and no robbery prevention training for employees, among the store characteristics associated with robbery and violence at gas stations and convenience stores (NIOSH). It notes that retail trade workers face a higher risk of violence and homicide than workers in other industry sectors. OSHA’s late-night retail guidance recommends that employers require workers to report all assaults or threats to a supervisor or manager (OSHA 3153), so the safest fix routes an associate to a manager. A fix that relies on an associate confronting an offender is never a quick win, however little it costs on paper. Score that kind of finding by what it risks to a person, and leave the savings column blank. A no-chase, no-touch policy rests on the same logic.

Cap a fix’s cost at the loss it will realistically prevent

A walkthrough built on a security survey checklist produces a stack of findings fast. Deciding which ones get fixed first is the harder part. FEMA’s building-security framework starts by analyzing the threat, the asset’s value and the vulnerability itself (FEMA 426), then selecting the mitigation measures with the greatest benefit relative to cost. NIST’s computer security handbook says the same thing in budget language: costs and benefits should be examined so a control’s cost doesn’t exceed what it’s expected to return. Both frameworks come from building and computer security, not retail, but the cost logic carries over. Translate that into the audit log with a formula instead of a guess. Call the finding’s annual frequency F, the loss per incident L, the share of that loss the fix will realistically stop R, and the years the fix stays useful Y. The fix clears the bar when its cost is under F × L × R × Y. A fix on a high-frequency loss point can clear that bar even at a real price. A fix guarding an item that almost never disappears can fail the test even when the hardware is cheap. Run a store’s own numbers through the shrink calculator before signing off on either one.

Tier What drives the score Cost rule Who closes it
Safety Severity and likelihood of harm to a person, not dollars Scored in non-monetary terms, funded ahead of Loss tier Named owner, due date, verify-by date, safety fixes first
Loss Frequency times loss per incident times the share stopped times fix life (F x L x R x Y) Cost must stay under F x L x R x Y Named owner, due date, verify-by date
Friction-sensitive Same as Loss, plus the sales a locked or gated fix will cost Cost plus expected lost sales must stay under F x L x R x Y Named owner, due date, verify-by date

Weigh customer friction as part of a fix’s real cost

A fix’s cost isn’t only what facilities pays for hardware and a key. Numerator’s shopper survey found that 27% of shoppers say they will switch retailers or abandon their purchase when merchandise gets locked up (Numerator survey). That figure comes from a survey of more than 5,000 consumers (Numerator survey). Kearney partner Michael Brown called locking up merchandise a last resort, pointing instead to cameras, extra staffing and security tags (Kearney via Customer Experience Dive). Before approving cameras or tags as the alternative, name who reviews the footage and who logs each alarm. A finding that recommends locking up a category should carry the lost-sale math next to the loss-prevented math, in the same line item, not in a separate note that gets dropped when the finding is approved. See why some security cameras and tags fail to deter theft.

Test that a fix worked before closing the finding

A finding isn’t fixed until someone checks the fix by its verify-by date. The POP Center’s shoplifting guide is blunt about the state of the evidence: information on what actually works is “severely limited because few of the common preventive practices have been evaluated.” That’s a reason to distrust any single severity score until it’s checked against what happens at the register. The same uncertainty shows up in the field: NIOSH reports that robberies fell significantly after Houston’s convenience-store CPTED ordinance took effect, especially in stores that implemented all of its elements, but the study could not determine the effectiveness of any single element (NIOSH). Build that second measurement into the worksheet as a verify-by date, not a courtesy follow-up added later. A finding that gets marked closed on paper but never re-measured is functionally still open, whatever the audit log says.

Close every finding with a named owner, a due date and a verified re-check

For workplace violence hazards, California turns the follow-through into statute. Labor Code 6401.9 requires a covered employer’s workplace violence prevention plan to list the “names or job titles of the persons responsible for implementing the plan,” and if there is more than one, their roles “shall be clearly described” (Cal. Labor Code 6401.9). The plan has to be reviewed at least annually, when a deficiency is observed or becomes apparent, and after a workplace violence incident. Inspections follow the same trigger logic: when the plan is first established, after each workplace violence incident, and whenever the employer is made aware of a new or previously unrecognized hazard. Records of workplace violence hazard identification, evaluation and correction have to be kept for at least five years (Cal. Labor Code 6401.9). The same law requires training on job-specific violence hazards, the corrective measures the employer has implemented, how to seek assistance to prevent or respond to violence, strategies to avoid physical harm, and a chance to ask questions of someone who knows the plan (Cal. Labor Code 6401.9). None of that depends on a dollar figure attached to the finding. A chain running stores outside California can borrow the structure without the legal mandate: assign an owner by name, set a due date tied to the finding’s tier, and put a re-inspection on the calendar before the file gets closed. Among its exemptions, the statute excludes places with fewer than 10 employees working there at any given time that are not accessible to the public, if they comply with Section 3203 of Title 8, but a retail sales floor full of shoppers would rarely qualify (Cal. Labor Code 6401.9). See California’s SB 553 requirements.

Frequently asked questions

What is the importance of an audit report in retail loss prevention?

An audit report earns its value only when every finding carries a named owner and a deadline attached to it. A list of problems with no one accountable can sit unrevisited, while a tracked log with assigned owners makes each gap someone's job to close.

Which audit fixes look cheap but put associates at risk?

A fix that has an associate confront, chase, or physically stop someone needs no capital spend, so it looks free on a cost sheet. NIOSH's convenience-store and fuel-station guidance ties active resistance during a robbery to worker injury. Robberies differ from shoplifting stops, yet that evidence warns against pricing such a fix at zero.

What remediation deadlines are defensible for a loss prevention audit?

No single deadline fits every finding. For workplace violence hazards, California's Labor Code 6401.9 requires procedures to correct hazards in a timely manner and sets event triggers: an annual plan review, a review when a deficiency shows up, and inspections after each incident or when the employer learns of a hazard it hadn't spotted before.

Should customer friction from a security fix count as a cost?

Yes. A fix that blocks a sale still carries a cost, even with no invoice attached to it. Numerator's shopper survey put a percentage on that friction, and that figure belongs in the same cost column as labor and hardware rather than off to the side as a separate customer-experience concern.

How do you estimate the loss a fix will actually prevent?

The worksheet multiplies frequency by loss per incident, then by the portion a fix is expected to stop, then by the fix's useful life, and checks whether the fix's price falls under that total. A high-value item that rarely disappears can still fail this test, even with low-cost hardware.

Sources

  • fema.gov https://www.fema.gov/pdf/plan/prevent/rms/426/fema426_ch1.pdf
  • nvlpubs.nist.gov https://nvlpubs.nist.gov/nistpubs/Legacy/SP/nistspecialpublication800-12.pdf
  • codes.findlaw.com https://codes.findlaw.com/ca/labor-code/lab-sect-6401-9/
  • popcenter.asu.edu https://popcenter.asu.edu/sites/g/files/litvpz3631/files/shoplifting_2nd_ed.pdf
  • osha.gov https://www.osha.gov/sites/default/files/publications/OSHA3153.pdf
  • cdc.gov https://www.cdc.gov/niosh/docs/wp-solutions/2023-121/2023-121.pdf
  • brc.org.uk https://brc.org.uk/news-and-events/news/operations/2025/ungated/brc-retail-crime-survey-2025/
  • customerexperiencedive.com https://www.customerexperiencedive.com/news/locked-up-merchandise-shrink-consumer-behavior-numerator/732333/