Cosmen & Keiless recovers 4 margin points with real-time visibility

The challenge

Cosmen & Keiless operates 12 restaurant establishments in Madrid and Barcelona. Growth over the past few years was rapid and organic: each new location was opened in a hurry, connected to the central system however possible, and the particularities of each site generated exceptions that nobody documented.

The result was predictable: 12 locations, 12 versions of reality. Each location’s POS generated data, but consolidating it required a weekly manual process. Shrinkage was detected at the end of the month, when it was already too late to act. The real cost of each dish was an estimate, not a data point. And the P&L by location was an accounting exercise that arrived with a 3-week delay.

The CEO knew he was losing margin, but couldn’t pinpoint exactly where or why. The managers of each location reported that “everything was fine,” but the overall numbers told a different story. The gap between what the locations reported and what the accounting said was a black hole of information.

The shrinkage problem was especially painful. In the restaurant industry, shrinkage is a silent enemy: products that expire, miscalculated portions, internal theft, inventory errors. Without daily visibility, shrinkage accumulates for weeks before anyone notices.

The solution

The 2-week Blueprint revealed that the fundamental problem was the lack of a single operational truth. Each system told a different story and nobody could reconcile them in time to make decisions.

We implemented three Engine layers:

Visibility connected the 12 POS systems, inventory systems, key suppliers, and accounting into an operational dashboard the CEO can check from his phone. Real-time sales by location, by product, by time slot. Inventory updated after each service. Procurement costs consolidated daily.

Data unified the definitions that diverged across locations. What constitutes shrinkage. How the cost of a dish is calculated. What the average ticket includes. What counts as a service. Each location now operates with the same definitions, the same categories, the same criteria. When the CEO looks at the dashboard, he knows the numbers are comparable.

Control was the layer that generated the most immediate impact. We designed a KPI tree with automatic alert thresholds. When a location’s shrinkage exceeds 3% in a single day, an alert fires. When a dish’s food cost deviates more than 5% from its technical spec, an alert fires. When a location’s sales drop 15% compared to the same day the previous week without a known explanation, an alert fires.

The alerts aren’t emails that pile up in the inbox. They are actionable notifications that reach the location manager and the director of operations with the specific data point, the measured deviation, and recommended actions.

The results

Shrinkage was reduced by 30% in the first 60 days. The key was daily detection. When you know that today you lost 4% at the Malasana location, you can investigate tomorrow. When you find out a month later, the damage is done and the cause is unrecoverable.

Operating margin improved by 4 percentage points. In addition to shrinkage reduction, visibility into the real cost per dish allowed adjusting prices and recipes for products with negative margin that nobody had detected. Three signature dishes on the menu were losing money because ingredient costs had risen without the technical spec being updated.

The P&L for each location is now available in real time. The monthly close went from being a 3-week accounting archaeology exercise to a confirmation of data already in hand. The finance team, freed from manual reconciliation, now dedicates its time to profitability analysis and planning.

Deviation detection went from weeks to a single day. The CEO makes decisions with yesterday’s data, not last month’s. And location managers have clear, measurable targets reviewed weekly with automated data.

Engine layers used

• Visibility: Multi-location real-time dashboard, connectivity with 12 POS systems, inventory systems, suppliers, and accounting
• Data: Unified data dictionary for the restaurant industry, standard definitions for shrinkage, food cost, average ticket, and operational KPIs
• Control: KPI tree with automatic alerts, thresholds by location and by metric, actionable notifications to managers and leadership