What a Restaurant Needs to Receive Outside Investment: the Pretty-EBITDA Myth and the Reality of a Replicable Operations Manual

What does an investment committee look at first in a restaurant group?

It looks at the VARIANCE between locations, not the profit peak of the best one. A three-location group in Bogotá came to me with consolidated EBITDA of 17.4% and a due diligence folder built by their accountant:

clean financials, leases current, equipment appraised. The fund still asked for something else by name — the replicable operating manual explaining how a new server is trained until they perform the same in location 1 and location 3. That gap cost four months of negotiation and eight points of valuation. A committee doesn't buy the restaurant that already works; it buys unit number four, which doesn't exist yet, and the only thing letting it forecast that unit is the standard deviation across the three it can audit. If your strong location runs 22% and the weak one 9%, you don't have a group: you have one restaurant with two expensive branches.

79% turnover is the number that makes a profitable operation UNFINANCEABLE

According to the National Restaurant Association, foodservice turnover hit 79% in 2025, and an analyst translates that figure straight into cost of capital. The operating read is blunt: at 79% turnover you rebuild a full location team roughly every fourteen months, so every new opening starts with a learning curve paid at payroll prices instead of investment prices. Diego F. Parra keeps hammering at Masterestaurant that outside capital never rewards the founder's talent — it rewards the system's INDEPENDENCE from that founder. The concrete decision that comes out of the number: before opening the round, measure front-of-house turnover location by location for six straight months and push it under the sector's 79%, because one point of turnover above benchmark costs you more in multiple than two points of food cost. A replicable manual is NOT a franchise requirement you write after selling the first master; it is the asset the investor is buying.

The operating manual is the asset being valued, not paperwork for later

Sector data backs that ranking: the Small Business Administration reports roughly 20-25% of franchises close within five years versus roughly 50% of independents, and those twenty-five points of gap come from the documented system, not the logo. FRANdata counts more than 4,000 brands and over 200,000 franchisees operating on that logic. A group arriving without a manual is asking a fund to finance a hypothesis, and funds finance processes. Write the preshift first, then the service simulator and the suggestive-selling standard with its metric attached; capital arrives to pay for replicating that manual, never to draft it. Kitchens standardize with spec sheets and scales; the dining room is what refuses to copy itself, and that is where most due diligence collapses. By the second meeting a committee asks what happens if the owner is sick for three weeks, and the answer doesn't live in the spreadsheet: it lives in whether the preshift is written down, whether the service simulator exists, whether suggestive selling is measurable.

The bottleneck sits in the dining room, not the kitchen

I got this wrong for years, prepping owners for the financial conversation with a combed P&L and a five-year projection while the real risk sat two meters past the pass. Toast documents combined ongoing royalty plus marketing fees between 8.5% and 11.2% of sales in franchised QSR, and that percentage buys precisely the dining-room system you haven't written yet. Present the opening cost of your LAST location with real invoices, line by line, instead of a three-location average. The difference matters when the market moves: ACODRES reported Colombian foodservice sales falling −24% in the first half of 2024, with more than 1,600 restaurants closed in 2023 and over 2,700 cumulative closures per ACOGA, while Frisby led the category with revenue above 1.21 trillion pesos and 12% growth according to Valora Analitik. That contrast tells a committee your historical CapEx average predicts nothing.

Expansion CapEx is backed by invoices, not by averages

What does predict something is the refrigeration invoice you signed in March, with its lead time and payment terms. Build the folder around the most recent location's supporting documents plus a explained variance against the original budget. A single location is NOT chasing fund equity but bank debt or an operating partner, so the homework differs: document twelve months of stable contribution margin and cut front-of-house turnover, because one point of sale gives you no variance to show. The mid-size group of two to five locations is the one that enters the equity conversation, and the cut there is hard — two locations with auditable unit economics, EBITDA deviation between units under five points, manual written. Groups of six and up play a different game, the Technomic game, which counted thirty chains opening more than a hundred locations in 2024 led by Starbucks, Jersey Mike's and Wingstop; at that level nobody negotiates whether the system exists, only how much CapEx per unit it absorbs before payback stretches past thirty months.

Where these benchmarks come from and what they can't tell you?

The figures in this analysis come from three kinds of public source, and it helps to know what each one measures.

Turnover and closure data come from trade associations — the National Restaurant Association in the United States, ACODRES and ACOGA in Colombia — which survey members and therefore underrepresent the informal segment. Scale and growth data come from Technomic, FRANdata and Datassential, which track chains and by design never see the single-unit independent. Franchise cost data comes from Toast, a vendor with transactional access to its own client base. None of those sources measures the multiple someone will pay you: a committee sets that with your folder on the table. Use them to locate yourself against the sector, never as a substitute for your own audited numbers. Suppose the round collapses at the final stage, which happens often when the manual never shows up.

What I'd do if the fund walks and you still want to grow?

If at that point you open location four with your own cash, the predictable thing happens:

the location 1 team gets split to seed location 4, turnover climbs past the 79% the National Restaurant Association reports, consolidated EBITDA drops two or three points, and next year's round opens with worse variance than today's. The opposite path is uncomfortable and it works — freeze the opening for six months, write the manual alongside each location manager, track suggestive selling per server week by week, and go back to the committee with two locations whose margin gap fits inside five points. Start tomorrow by timing the preshift at your weakest location. The myth puts profit at the center; due diligence puts VARIANCE there. A fund evaluating what a restaurant needs to receive outside investment is not hunting the peak, it is measuring the spread across units, because that spread predicts how location number four, the one that does not exist yet, will perform.

Where myth and reality actually part ways?

Owners treat the replicable operations manual as a franchising requirement, something written after the first master agreement sells. In practice it works the other way around:

the manual is the asset being valued, and capital arrives to pay for it, not to write it. A group with no manual is asking an investor to fund a hypothesis. Conventional wisdom says the kitchen is the critical part of the system because food cost lives there. Yet kitchens standardize through spec sheets and gram weights, which are hard data; dining rooms standardize through judgment, behavior and timing, which is precisely what does not travel on its own. That is why an Interactive Training Kit with simulators and automated preshift weighs more in the folder than most owners imagine. Another myth holds that expansion CapEx gets negotiated at the end. A poorly supported CapEx dismantles the whole thesis instead: if the owner estimated USD 380,000 per unit and the previous invoice reads 447,000, the committee assumes every other figure carries the same optimistic bias and applies a blanket discount.

Where myth and reality actually part ways — in practice

Owners assume daily presence is a guarantee. Deloitte and several sector investment banks have spent years discounting exactly that: founder dependency is concentration risk, and in restaurant scaling it gets paid in multiple, not in compliments.