How a 24-Location Restaurant Group Cut Cold-Chain Losses and Trimmed Demand Peaks With EnergyOS
The operator and their world
Picture a regional quick-service restaurant group with roughly 24 locations spread across a mix of suburban strip centers, standalone drive-thrus, and a few urban storefronts. The menu leans on fresh proteins and dairy, which means every site runs walk-in coolers, reach-in freezers, prep-line refrigeration, and a wall of cooking equipment that all draw power hard during the lunch and dinner rushes. Margins in this business are thin, and the two costs that move the needle most are labor and energy. Labor gets watched hour by hour. Energy, until recently, was a mystery that showed up on a utility bill weeks after the money was already spent.
The operations team was competent and stretched. District managers drove between four and six sites a week, checking temperature logs on clipboards, spot-checking equipment, and reacting to problems after they happened. There was no single place to see how all 24 restaurants were performing against each other, and no early warning when a compressor started to fail or a demand peak was about to reset the billing baseline for the month.
The core problem was visibility. Energy was the second-largest controllable cost after labor, yet the team had no way to compare sites or spot a peak until the bill arrived, and cold-chain failures hit without warning.
The challenge
Two pain points kept surfacing. The first was cold-chain risk. A walk-in cooler that drifts out of range overnight can spoil thousands of dollars of inventory before anyone opens the door in the morning. Over a year, a handful of these events across a portfolio this size added up to real money, and a few of them landed in the five-figure range for a single site. Manual temperature logs caught problems only when someone happened to be standing in front of the unit with a clipboard.
The second pain point was demand. Utility bills for commercial accounts often carry a demand charge tied to the single highest spike of usage in the billing period. When the fryers, ovens, HVAC, and refrigeration all ramped at once during a rush, a site could set a new peak that inflated its bill for the entire month. Nobody saw those peaks form in real time, so nobody could do anything about them. And because every location reported separately, leadership could not tell whether the store on Route 30 was simply busier than the one downtown or quietly wasting energy through failing equipment and bad scheduling.
The deployment
Emergent Metering approached this as a managed service rather than a hardware sale. The team installed wireless power submeters on the main feeds and key equipment groups at each restaurant, then added cold-chain temperature sensors inside every walk-in cooler, freezer, and critical prep unit. The sensing hardware is powered by Monnit and delivered inside the EnergyOS platform by Emergent Metering, so the operator dealt with one partner and one dashboard instead of stitching together devices, gateways, and software on their own.
Because the meters and sensors are wireless, installation did not require rewiring the buildings or shutting down service during business hours. Within a short rollout window, all 24 sites were reporting into EnergyOS. A metering tree organized the raw data by site and by equipment type, and virtual meters rolled everything up into a portfolio view. For the first time, the group could open one screen and see all 24 restaurants side by side.
What EnergyOS did day to day
Once the data started flowing, the platform went to work in a few practical ways.
Benchmarking came first. EnergyOS ranked all 24 sites against each other on energy use, normalized for size and traffic. The outliers stood out immediately. Two or three restaurants were consuming far more than peers of similar volume, which pointed the maintenance team straight at aging compressors, doors left propped open, and HVAC running against refrigeration in the same space.
Demand management came next. The team set demand limits for each site inside EnergyOS and turned on peak alerts. The platform forecasts when a location is trending toward a new peak and sends a warning before the spike locks in. Managers used that lead time to stagger prep schedules and shift HVAC cycles so the heaviest loads did not all land in the same fifteen-minute window.
Cold-chain protection ran around the clock. When a cooler or freezer drifted out of its safe range, EnergyOS fired a real-time alert routed to the on-site manager by text, with escalation to the district manager and a call if no one acknowledged it. An excursion that used to be discovered at 6 a.m. was now caught within minutes, often while it was still a minor compressor hiccup rather than a full failure.
Every morning, an AI daily site summary landed in the operations inbox, plain-language notes on which sites ran hot, where a peak formed, and which units flagged overnight. Cost allocation and budget-versus-variance tracking gave finance a clean way to hold each location accountable to its own number.
The results
The outcomes here are illustrative of what operators in this category typically see, not audited figures from a single named client. Directionally, the picture looks like this.
Energy waste surfaced fast at the outlier locations. Once the worst performers were identified and their equipment and schedules corrected, the group saw energy reductions in the range of roughly 8 to 15 percent at those sites, the kind of improvement that is common when hidden waste finally becomes visible.
Demand peaks came down as managers used forecasts and alerts to shift prep and HVAC cycles out of the rush window, softening the demand charges that had quietly inflated bills for months.
Cold-chain losses dropped the most in human terms. Catching excursions in minutes instead of hours meant fewer walk-in failures turned into full inventory write-offs, and the occasional five-figure loss became a near miss instead of a claim.
One more result was quieter but meaningful. The clipboards went away. A single dashboard replaced manual temperature logs, giving managers time back and giving leadership a portfolio view they never had before.
What it means for similar operators
Any multi-site food service group is sitting on the same two problems this operator faced, spend they cannot see and risk they cannot predict. Wireless submetering and cold-chain sensing, delivered as a managed service inside one platform, turn both into something you can act on before the money is gone. The value is not in the sensors themselves. It is in the consolidated view, the alerts that reach the right person in time, and the benchmarking that tells you which of your locations is quietly costing you the most.
If you run cold chain across multiple sites and you are still finding out about problems on the utility bill or the morning walk-in check, there is a better way to run it.
Schedule a Platform Demo. Call 215-645-7141.