How to Run a Loss Prevention Audit Across Multiple Restaurant Locations
Savi
A loss prevention audit is how multi-unit operators find out where money is actually leaking, whether that's cash handling, inventory shrink, or process gaps at the register. For operators running ten locations or two hundred, the challenge isn't wanting to run a loss prevention audit. It's doing it consistently, across every site, without a full-time investigator on staff.
This post walks through what a loss prevention audit actually covers, why it's harder at scale than it sounds, and how video data changes the math.
What a Loss Prevention Audit Actually Covers
A loss prevention audit is a structured review of the places money and inventory can disappear between the supplier and the register. For a restaurant brand, that typically means:
Cash handling: till counts, void and refund patterns, safe drops, and manager overrides
Inventory shrink: waste logs versus actual pulls, portioning consistency, and unexplained variance between what's ordered and what's sold
Compliance events: unauthorized discounts, comped meals, and off-the-clock activity near registers or safes
Internal theft: patterns that don't show up in a single transaction but emerge over weeks, like a team member who consistently voids orders during their shift
Traditionally, this audit runs on spreadsheets and gut instinct. A district manager pulls sales reports, cross-references them against known problem sites, and maybe reviews a few hours of DVR footage if something looks off. That process works, slowly, for one location. It falls apart at scale.
Why the Loss Prevention Audit Gets Harder as You Grow
Every new location is a new place for shrink to hide. A ten-unit brand can eyeball its numbers weekly. A fifty-unit brand can't, and a two-hundred-unit brand often doesn't even know which sites need a closer look until the quarterly numbers come in, months after the loss happened.
The core problem is visibility. Most operators have cameras at every site, but the footage lives on a local DVR that nobody watches unless there's already a reason to. A loss prevention audit under that setup means someone has to know where to look before they can look. By the time a variance report flags a site, the footage may already be overwritten, and the pattern behind it, whether it's a specific shift, a specific register, or a specific employee, is invisible.
This is the gap Savi's customers have run into directly. FiiZ Drinks used video paired with Event Search to discover $3,250 in internal loss in the first 90 days, loss that existing reporting hadn't surfaced on its own. A Scooter's Coffee franchisee caught $3,500 in internal theft in the same window and added 1.41% back to gross sales, prompting franchisee Craig Schroeder to say, "This system pays for itself." Those aren't numbers you get from a spreadsheet audit. They come from being able to see the moment behind the variance.
Building a Loss Prevention Audit Process That Scales
A loss prevention audit that works across many locations needs three things a manual process struggles to deliver:
1. A consistent data source at every site. If cameras, POS systems, and inventory tools aren't standardized across locations, your audit is only as strong as your weakest-instrumented site. Cloud video management puts every location's footage in one place instead of scattered across local DVRs that vary site to site.
2. A way to search for the moment, not scrub for it. The old way of reviewing footage means hours of fast-forwarding. Event Search lets an investigator jump to the transaction in question, a suspicious void, a manager override, a register drawer left open, and see exactly what happened around it in seconds instead of hours.
3. Reporting that flags outliers before they become a pattern. Enterprise reporting that compares sites against each other surfaces the location with an unusual void rate or an inventory variance that's drifting, so a district team can investigate proactively instead of waiting for the quarterly numbers.
This is also where IT teams get pulled into loss prevention audits, often reluctantly. A Burger King franchisee dealt with exactly this: a fragmented camera setup meant every loss prevention question turned into an IT ticket. Consolidating onto a single cloud platform gave GMs and DMs org-wide video access directly, described as "essentially a Google Search for our operations."
The Platform Behind the Audit
The same video dataset that powers a loss prevention audit doesn't have to stop there. Once footage and events are centralized in the cloud, that foundation supports operations questions about speed of service, compliance checks for brand standards, and training use cases, all from the same cameras already installed at every site. That matters because a loss prevention audit shouldn't require a separate system from the one operations uses to track drive-thru times or the one IT uses to manage camera infrastructure. As computer vision and AI capabilities advance, a cloud-architected dataset means a brand can adopt new tools without re-wiring a single site. It's a foundation decision for the whole business, not a point solution bought to solve one problem in one department.
Key Takeaways
A loss prevention audit covers cash handling, inventory shrink, compliance events, and internal theft, and gets exponentially harder to run consistently as location count grows.
Manual, spreadsheet-driven audits catch problems months after they happen, once footage is already gone.
Centralized cloud video plus searchable events let operators find the specific moment behind a variance instead of scrubbing hours of footage.
Real customer results back this up: FiiZ Drinks found $3,250 in internal loss and a Scooter's Coffee franchisee caught $3,500 in theft, both within the first 90 days.
The video dataset that supports a loss prevention audit is the same foundation that supports operations, IT, and training, so it's worth treating as an infrastructure decision, not a one-off tool.
Ready to see what a loss prevention audit looks like with the right visibility in place? See how Savi works, request a demo.
