Franchise or Grow Company-Owned? The Capital Decision that Defines the Decade

The verdict: don't pick franchising or company-owned growth on taste, pick it on unit economics. Grow company-owned until you've proven the model replicates a unit-level EBITDA of ≥18% across three distinct locations; only then franchise to scale without draining cash. Franchising an unproven model hands your mistake to third parties and destroys the brand; hoarding all the CapEx once the model already replicates stalls your scaling and gifts territory to the competitor.
A restaurant group leader reaches this crossroads with two or three profitable locations and a billion-dollar question: grow on your own cash and keep all the EBITDA, or franchise and buy speed with other people's capital. The wrong answer costs a decade.
This brief is the written version of a boardroom keynote: decision architecture, unit-economics figures and a three-phase roadmap to scale a restaurant without betting the brand.
Side-by-side comparison
| Franchise | Grow company-owned | |
|---|---|---|
| Per-unit CapEx the group carries | ✕≈ $0 (funded by the franchisee) | ✓$180k–$650k per unit |
| Margin the group retains per unit | ✕5%–8% (royalty on sales) | ✓14%–22% (full unit-level EBITDA) |
| Opening speed (units/year) | ✕8–20 with a mature network | ✓2–4 capped by cash |
| Payback on the group's investment | ✕12–18 months (fee + royalty) | ✓28–42 months per location |
| Real operational control | ✕60%–75% (via manual + audit) | ✓95%–100% (own staff) |
| Brand dilution risk | ✕High if the model isn't proven | ✓Low (controlled variability) |
| Non-negotiable prerequisite | ✕Validated replicable operations manual | ✓Cash or debt for the CapEx |
1. Franchise or grow with your own capital? The verdict runs on unit economics
Don't choose between franchising or company-owned growth on taste: choose it on unit economics. Grow with your own cash until you validate that the model replicates a unit-level EBITDA ≥18% across three different locations, with different teams and different zones; only then do you franchise. I've seen it in dozens of groups: the operator who franchises with two locations and a 12% margin isn't scaling a business, he's scaling a problem. Franchising an unvalidated model multiplies the error, not the win. Growing company-owned keeps 100% of each unit's EBITDA and builds equity: a location generating 90,000 USD of annual EBITDA is worth, at a 4x multiple, about 360,000 USD that sits on your balance sheet, not the franchisee's. The speed of franchising only offsets that dilution once the model proves it replicates without you. Franchising isn't a growth strategy: it's a replication strategy, and it only works if you already have a system that produces results without the owner's charisma.
2. Franchising isn't growth: it's replicating a system that runs without the owner
The hard test is simple: pull the founder out of the location for 90 days. If food cost climbs from 30% to 37%, average ticket drops 15%, and the rating falls from 4.7 to 4.2, you don't have a franchisable model, you have a restaurant that bills because of you. At Masterestaurant we call this the replication test: an average manager, not a star, must hold food cost under 30%, the target ticket, and a rating above 4.5 for a full quarter. Only 3 out of every 10 groups reaching this crossroads pass it on the first try. The rest confuse an operation that depends on one person with a system that depends on a manual. Growing with your own capital maximizes margin per unit and equity value; franchising maximizes speed and territorial coverage. They're different objectives and the board must declare which one it pursues before signing anything.
3. Company-owned maximizes margin and equity; franchising maximizes speed
With your own cash you keep the full EBITDA —18% to 22% on sales in a healthy model— but you grow at the pace your cash flow allows: one or two locations a year. With franchising you collect an initial fee of 25,000 to 50,000 USD and a royalty of 5% to 7% of the franchisee's sales, so your margin per unit collapses to that 6% royalty, but you can open ten locations in the time it would take you to open two. The board's question isn't which is better: it's whether ten units at 6% are worth more than three units at 20%. The answer depends on your cost of capital and your horizon. The costliest capital mistake isn't choosing wrong between franchising or growing: it's franchising a model with unvalidated unit economics. When you sell a franchise of a model that yields 12% instead of 18%, you're not selling an opportunity, you're selling a liability.
4. The costliest capital mistake: franchising unvalidated unit economics
The franchisee invests 200,000 to 400,000 USD, doesn't recover it in the promised 36 months, and each dissatisfied one is a potential lawsuit and a one-star review that contaminates the entire brand. I've seen groups lose five years and their reputation by opening fifteen franchises on a model that never proved it could replicate. The rule is hard: validate unit-level EBITDA ≥18% across three locations run by different managers before selling the first contract. Multiplying a win makes you a category leader; multiplying an error turns you into the case study nobody wants to be. The due diligence that reveals whether a model is franchisable is the operational one, not the financial. The P&L numbers tell you whether the business made money last year; operational due diligence tells you whether it will make money when the owner isn't in the kitchen. The central question is one: does an average manager replicate food cost under 30%, the target ticket, and a rating above 4.5 without you in the location?
5. The due diligence that matters is operational, not financial
To answer it you need manuals a new employee can execute in 30 days, recipes standardized to the gram, a purchasing system that fixes food cost without the chef's judgment, and KPIs a franchisee can sustain alone. At Masterestaurant we measure 'founder dependency': if more than 40% of daily decisions pass through the owner, the model isn't ready to franchise, however profitable it looks on paper. The route to scale a restaurant without betting the brand has three phases and no shortcut. Phase one, validation: grow with your own capital to three locations and prove unit-level EBITDA ≥18% with different managers over 12 months; here you franchise nothing. Phase two, systematization: turn what works into manuals, recipes to the gram, and a KPI dashboard, and drop founder dependency below 40%; budget between 50,000 and 120,000 USD and six to twelve months to get the model ready.
6. The three-phase route to scale without betting the brand
Phase three, replication: only now do you franchise, with an initial fee of 25,000 to 50,000 USD, a 5% to 7% royalty, and a pilot of three franchisees before opening at scale. Diego F. Parra sums it up in boardrooms like this: whoever skips phase one isn't scaling, he's gambling; and the brand is the one thing you can't buy back when fifteen franchisees burn it at once. Franchising isn't a growth strategy: it's a replication strategy. It only works if you already own a system that replicates results, not a restaurant that sells well on the founder's charisma. Company-owned growth maximizes per-unit margin and equity value; franchising maximizes speed and territorial coverage. Different goals, and the board must declare which it's chasing before signing anything. The costliest capital mistake isn't choosing wrong: it's franchising a model with unvalidated unit economics.
7. The differences that decide the decade
You multiply the error, not the win, and every unhappy franchisee is a reputational liability. Operational due diligence —not the financial one— reveals whether the model is franchisable: does an average manager replicate the food cost, the ticket and the review without you in the room?
Franchise vs company-owned, criterion by criterion
FranchiseOther people's capital · speed
- Expansion capital funded by the franchisee: group CapEx near zero.
- Fast scaling: 8–20 openings a year with a mature network.
- Recurring income from royalty (5%–8% of sales) plus entry fee.
- Requires a proven, replicable operations manual before selling the first franchise.
- The group gives up 25%–40% of direct operational control.
Grow company-ownedMasterestaurant
- The group keeps the full unit-level EBITDA: 14%–22% per location.
- Near-total operational control: operational variability is governed up close.
- Speed capped by cash: 2–4 openings a year without aggressive debt.
- Full expansion CapEx on the group: $180k–$650k per unit.
- Slower payback (28–42 months) but higher equity value at exit.
Side-by-side comparison
| Franchise | Grow company-owned | |
|---|---|---|
| Per-unit CapEx the group carries | ✕≈ $0 (funded by the franchisee) | ✓$180k–$650k per unit |
| Margin the group retains per unit | ✕5%–8% (royalty on sales) | ✓14%–22% (full unit-level EBITDA) |
| Opening speed (units/year) | ✕8–20 with a mature network | ✓2–4 capped by cash |
| Payback on the group's investment | ✕12–18 months (fee + royalty) | ✓28–42 months per location |
| Real operational control | ✕60%–75% (via manual + audit) | ✓95%–100% (own staff) |
| Brand dilution risk | ✕High if the model isn't proven | ✓Low (controlled variability) |
| Non-negotiable prerequisite | ✕Validated replicable operations manual | ✓Cash or debt for the CapEx |
The capital scoreboard in figures
“They had three locations selling well and wanted to franchise 20 in two years. Operational due diligence showed food cost jumped from 29% to 41% whenever the founding chef was absent. We halted franchise sales, systematized the replicable operations manual over seven months and cut food-cost variability to ±2 points. Only then did we open the network: 11 franchises in 18 months, none below 17% EBITDA.”
Strategic roadmap: from profitable location to scalable network
Deliverable: per-location unit-economics dossier with food cost, labor, contribution and unit-level EBITDA. Success metric: unit-level EBITDA ≥18% replicated across 3 distinct locations with distinct managers. Without this threshold, there's no decision to make: the model isn't ready to franchise.
Deliverable: replicable operations manual + territorial prefeasibility with location intelligence for the next 10 sites. Success metric: food-cost variability ≤±2 points and average ticket replicated within ±5% across locations, measured with an average manager, not the founder.
Deliverable: expansion plan (franchise, company-owned or hybrid) with CapEx, payback and corporate governance defined. Success metric: 8–12 units opened in 12 months (franchise) or 3–4 (company-owned), none below the target unit-level EBITDA.
And with AI?
Standardize and replicate processes to scale and franchise with control. Diego F. Parra is an expert in AI applied to restaurants.
Free tools to apply this now
The ecosystem behind the decision
No capital decision of this magnitude gets made on intuition. The Masterestaurant ecosystem turns founder instinct into measurable decision architecture: unit economics, territorial prefeasibility and operational-variability control before committing a single dollar of CapEx.
Questions every board asks
When is it right to franchise instead of growing company-owned?
When is it right to franchise instead of growing company-owned?
When your model has already replicated a unit-level EBITDA of ≥18% across at least three distinct locations, with distinct managers and no dependence on the founder. Before that threshold, franchising hands your mistake to third parties and erodes the brand.
How much CapEx does company-owned growth demand?
How much CapEx does company-owned growth demand?
Between $180k and $650k per unit depending on format and market, entirely carried by the group. Payback runs 28–42 months, but you keep the full unit-level EBITDA (14%–22%) and gain equity value at exit.
What must operational due diligence validate before franchising?
What must operational due diligence validate before franchising?
That an average manager replicates food cost (≤32%), ticket and experience without the founder present. If results depend on the owner's charisma, you don't have a franchisable system: you have a restaurant that sells well.
Why is franchising an unproven model the costliest capital mistake?
Why is franchising an unproven model the costliest capital mistake?
Because you multiply the error, not the win: every unhappy franchisee is a reputational and financial liability. A badly launched network destroys in 18 months the brand that took a decade to build.
Sector data 2026 (official sources)
Verifiable industry benchmarks from official, non-commercial sources (government, industry associations, market research) - not competitors.
| Metric | Benchmark 2026 | Source |
|---|---|---|
| Prime cost a escala (multi-unidad) | 55–65% de las ventas | National Restaurant Association |
| Margen neto del sector | 3–9% | Statista |
| Operación fuera del local | ~75% del tráfico | Nation's Restaurant News |
| Hostelería en Europa | estadística oficial de restauración | Eurostat |
| Top 500 de cadenas | las 500 mayores cadenas concentran la apertura neta de unidades en EE.UU. | Nation's Restaurant News — Top 500 |
| Expansión internacional QSR | la expansión fuera de EE.UU. la lideran marcas de servicio limitado (QSR 50) | QSR Magazine |
Download this document as PDF
The full text is free to read on this page. To take the corporate PDF with you, leave your details — we'll also email you the direct link.
Related content
Grow your restaurant with the Masterestaurant method
Applied in +8.400 restaurants across 43 countries.
