The Scaling Paradox: Why More Locations Don't Mean More Profit

1. Why doesn't opening more locations mean more profit?

Opening more locations multiplies revenue but usually destroys EBITDA per unit: if location #1 yields 18% and #5 barely 6%, the group grows in sales and shrinks in profitability.

I have audited groups of 2 to 15 units where consolidated revenue rose 140% in three years while aggregate net profit fell from 14% to 7%. Scaling is not a problem of ambition or location: it is a problem of unit economics. Every opening without a replicable model adds variability, and that dispersion is paid in margin. At Masterestaurant we repeat it in every board meeting: don't expand a restaurant you can't yet clone. The margin of your first location, if it was born from talent and not from a system, does not travel to the second unit; it stays in the first and leaves you with five registers that never reconcile the same way. Operational variability is the invisible tax of growth: every point of food-cost dispersion between locations is EBITDA that evaporates without appearing on any line of the income statement until it is too late.

2. What is operational variability and why is it an invisible tax?

A group with an average food cost of 30% sounds healthy, but if it swings between 26% and 38% across units, half your locations operate outside the profitable threshold.

I have measured differences of 9 to 12 food-cost points between the best and worst location in the same group, with the same menu and the same suppliers. That gap is 3 to 4 points of consolidated EBITDA that nobody sees in the monthly P&L. The problem is not the average; it is the standard deviation. When each location decides portions, waste and purchasing at the whim of the shift's head chef, entropy grows with every opening and margin dilutes. Improvised expansion treats each location as an artisanal project; systematic scaling treats each location as a replica of a proven model. The first accumulates entropy with every opening; the second reduces it. In the improvised model, the margin of location #1 is a happy accident that cannot be cloned: it depends on a star chef, an intuitive manager, an unrepeatable lease contract.

3. What is the difference between improvised expansion and systematic scaling?

In the systematized model, the margin is the result of a decision architecture that travels from one unit to the next without degrading. The proof is in the numbers:

groups that codify recipe sheets, target waste and purchasing schedules manage to keep food cost within a range of ±2 points across locations. Those that don't live with ranges of ±10 points. The same brand, two different economies. The gap is quantified by comparing per-unit margin against the group's weighted average, and it almost always reveals that most locations scale revenue and de-scale profitability at the same time. In the groups I audit, the pattern is constant: 20% of the units generate 60% of the EBITDA, and the bottom 30% operate near break-even or at an operating loss. When you open location #6 without fixing this, you don't dilute the problem: you replicate it. A group with an average ticket of 24 USD and 5 locations can bill 6 million a year and take home the same as it did with 3 locations.

4. How do you quantify the gap between revenue and profitability when scaling?

Revenue grew 66% and net profit stalled. That is the paradox of scaling: more volume, the same or less free cash, because operational variability ate the increase before it reached the bottom line.

The sequence that works is three phases in strict order: codify, standardize, then replicate. Phase 1, codify: document the model of your most profitable location in recipe sheets, target waste, cost per dish and purchasing protocols; without this, there is nothing to clone. Phase 2, standardize: bring all existing units to the same target food cost (±2 points) before signing a single new contract; fixing 5 locations costs less than dragging 5 errors into 10. Phase 3, replicate: only then open, using the proven model as a template, not as inspiration. At Masterestaurant we have seen groups recover 3 to 5 points of consolidated EBITDA in 12 months just by tightening phase 2, without opening anything. The error I see over and over is jumping straight to phase 3 because ambition demands growth; the register demands you first close the leak.

5. When should you NOT open your next location?

Don't open your next location if you still can't explain, with numbers, why your top-earning location wins. It is the acid test:

if the margin of your best unit is not codified into a system another manager can execute, every opening is a bet, not an investment. A hard rule I apply in board meetings: if the food-cost dispersion across your current locations exceeds 4 points, freeze expansion. First close that gap; every point you reduce is worth more than the incremental sales of a poorly calibrated new unit. A group of 4 locations that lowers its deviation from ±8 to ±3 points gains more EBITDA than by opening an average fifth location. Correct growth does not start with a lease contract; it starts with a model that yields 18% and knows why. Scale the system, not the hope. Improvised expansion treats each location as an artisanal project; systematized scaling treats each location as a replica of a proven model.

6. The structural difference the board overlooks

The first accumulates entropy with every opening; the second reduces it. In the improvised model, unit #1's margin is a happy accident that can't be cloned. In the systematized model, margin is the output of a decision architecture that travels from one unit to the next without degrading. Operational variability is the invisible tax on growth: every point of food-cost dispersion between units is EBITDA that evaporates without appearing on any line of the income statement until it's too late.