Across 5 comparable QSR locations operating under a multi-unit franchise portfolio, COGS as a percentage of sales dropped from 15.44% to 15.01% year over year. Sales grew 2.45% while COGS declined 0.39%, corresponding with margin expansion during a period of rising supplier costs and inflation.
This is not a case study about one store. This is the aggregate picture of what happens when five locations all count inventory every week and let data inform ordering decisions instead of habit, memory, or fear.
The data below comes from a real multi-unit QSR operator with 20 years of experience managing 15 stores across multiple brands. The five locations included in this analysis are the stores that were operating in both years under the same ownership, with no acquisitions or divestitures during the comparison period. All numbers are real. All results happened despite rising supplier costs and inflationary pressure.
| Metric | Prior Year | Current Year | Change |
|---|---|---|---|
| Total Sales | $2,919,857 | $2,991,297 | +2.45% |
| Total COGS | $450,838 | $449,092 | -0.39% |
| COGS as % of Sales | 15.44% | 15.01% | -0.43 percentage points |
Sales went up. COGS went down. The margin expanded by 0.43 percentage points across the portfolio. In a year where supplier costs were rising, that is not supposed to happen. COGS should increase with inflation. It did not, because the ordering got better.
| Location Type | 2025 COGS% | 2026 COGS% | Change | Sales Trend |
|---|---|---|---|---|
| QSR multi-brand (3 brands) | 13.38% | 13.55% | +0.17pp | +21.6% growth |
| QSR snack brand | 14.94% | 14.97% | +0.03pp | -5.7% decline |
| QSR snack brand | 15.67% | 15.34% | -0.33pp | -7.7% decline |
| QSR bakery | 20.61% | 17.66% | -2.94pp | -3.1% decline |
| QSR bakery | 16.26% | 15.99% | -0.27pp | +1.1% growth |
Every location either improved its COGS margin or held it within 0.17 percentage points. No location got significantly worse. Three locations showed measurable margin improvement. Two held steady in very different conditions: one during 21.6% sales growth, one during a 5.7% sales decline.
Two of the five locations saw sales decline year over year, yet both improved or held their COGS margin. This is the counterintuitive result that weekly counting can produce. When sales decline, most operators see COGS percentage rise because they keep ordering at the same level. Weekly counts catch the decline early and adjust PAR levels downward before excess ordering compounds.
One location grew sales by 21.6% and saw COGS margin increase by only 0.17 percentage points. Without weekly counting, growth typically causes over-ordering because the manager bumps up orders across the board to avoid stockouts. The rolling 3-week average adjusts PAR levels gradually for only the items that are actually selling faster, preventing blanket over-ordering.
Combining raw COGS reduction ($25,080 across four locations) and efficiency gains during scaling ($6,911 at the growth location), the total COGS reduction observed was approximately $32,000 year-to-date across 5 locations. Annualized, that projects to roughly $48,000 per year across 5 stores. Scaled across a 15-store portfolio, the number compounds significantly.
Every location in this analysis was using the same approach: count every item every week, review usage data and PAR levels in the app, order based on the data, confirm the order was submitted, and repeat the next week. This is a 5-step loop called The TRACK Method.
| Step | Action | What It Produces |
|---|---|---|
| Track | Count every item every week | Real usage data instead of estimates |
| Review | Check Stock Dots, PAR levels, usage averages | Know what is overstocked and what is low |
| Act | Order what is below PAR, generate vendor email | Order sized to actual need, not habit |
| Confirm | Finish and notify, accountability trail | Proof the count happened and the order was placed |
| Keep | Repeat weekly | Compounding accuracy over time |
The method is simple. The discipline is what makes it work. Five locations, different sales trends, different starting COGS levels, all improving or holding steady. The common factor is the weekly count.
Yes. All five locations faced supplier cost increases and inflationary pressure during the comparison period. COGS should have risen simply from higher prices per unit. Instead, better ordering offset the price increases. The locations ordered less of what they did not need, which more than compensated for higher per-unit costs on what they did need.
Individual locations can show improvement within the first month. Portfolio-level results become meaningful after 8 to 12 weeks of consistent weekly counting across all locations. The rolling averages need time to stabilize, and PAR corrections compound over consecutive weeks. Full-year results, as shown here, represent the cumulative effect of 52 weeks of incremental improvement.
The mechanism is the same regardless of location: weekly counts generate usage data, usage data informs PAR levels, PAR levels inform order quantities. Locations with higher starting COGS ratios will see more dramatic improvements. Locations already operating efficiently will see smaller, incremental gains. Both are real and both compound over time.