What a restaurant needs to receive outside investment: the file an investor actually reads

Separate personal money and close twenty-four reconciled months of P&L

The first deliverable is not a deck, it is a bank account dedicated to the business plus twenty-four monthly P&L files reconciled against statements, with owner draws on their own line and never buried inside operating expense. That rebuild, usually five to seven weeks of backward bookkeeping, is what produces a defensible adjusted EBITDA in a sector whose average pre-tax net margin sits near 5% at full service (National Restaurant Association 2026). The numeric checkpoint is blunt: the gap between declared sales and bank deposits must land under 1.5%, and if your history drags three mixed accounts behind it, start there before booking any meeting. A committee forgives one bad year that comes with an explanation; it will not forgive a fourteen-month documentary hole with no valued inventory. Once the history closes, costing begins, and the deliverable here is a spec sheet per dish carrying weights, yield loss, current purchase prices, individual food cost and contribution margin in cash, alongside a menu engineering matrix with its four quadrants.

Cost the menu plate by plate and push prime cost under 64%

The house rule governs: 32% is the food cost CEILING per plate, never the target, and labour, rent and utilities never load onto the plate, they belong to break-even. Verify three consecutive months of prime cost ≤64% with theoretical and actual food cost inside two points of each other. This matters because a third of operators, 33% according to Deloitte Restaurant of the Future 2025, name labour cost as their main margin pressure, and no investor funds a margin the owner cannot explain line by line. A fund is not buying your cooking, it is buying your ability to repeat it, and the document proving that is the MTIE —Transfer and Establishment Engineering Model— with eight critical processes written down: purchasing, receiving, production, service, cash close, maintenance, hiring and escalation. Add signed spec sheets and a management matrix naming a trained backup for every position, versioned and dated. The proof of life is deliberately uncomfortable: you disappear for fourteen straight days, sales drop no more than 5% and food cost rises no more than 1.5 points.

The MTIE turns craft into an asset somebody else can carry

Diego F. Parra keeps hammering this point inside Masterestaurant, because the step explains the entire valuation band: identical EBITDA fetches 2x when 85% of critical calls run through the owner, and reaches 4.5x once the system absorbs them. An investor is not funding your current unit, they are funding the second and the third, so they need to watch the method behind your choice. Cross eleven variables per candidate: residential and office density, footfall by daypart, area average ticket, category saturation within 800 metres, access, visibility, parking, rent per square metre, fit-out cost, delivery coverage and public works risk. The deliverable is a scored dossier of three candidates, and two thresholds close the argument: projected rent ≤8% of conservative estimated sales (CBRE Restaurant Real Estate Outlook 2025) and capex payback ≤30 months. With 60% of independent restaurants gone before year three (U.S. Bureau of Labor Statistics 2025), that selection discipline carries more weight at the table than any five-year projection ever will.

A nine-folder data room and a pitch that reaches margin on slide two

Build the data room before the first call, not after the first awkward question: nine folders —financials, tax, labour, leases, permits and licences, suppliers, brand and intellectual property, insurance and the MTIE— with access control and a query log. On top sits a twelve-slide pitch whose second slide already shows single-unit economics: average ticket, covers, capex, EBITDA and payback in months. The checkpoint measures response time: any committee question gets answered from a data room document within 24 hours, with nothing fabricated. This sequence compresses due diligence from the usual 14-20 weeks to 5-7, and that compression is not administrative comfort, it is price, since every extra round of clarifications cools the committee and brings a downward revision of valuation. Five failures account for most rejections, and none of them involve the food. One, rebuilding only the last six months because the earlier ones were messy: the hole weighs more than the bad result.

The five mistakes that sink the process, and how to dodge them

Two, cutting food cost by shrinking portions, which fixes the month and kills repeat visits. Three, writing the manual and training nobody on it, which leaves the MTIE as dead paper. Four, falling for the cheap unit on a street with no evening traffic and signing five years of rent on a feeling. Five, opening the meeting with the grandmother story and reaching margin on slide nine. A sixth catches almost everyone off guard: partly informal payroll, which stops being a saving during due diligence and becomes a contingent liability the buyer's lawyer prices in full. You are ready when seven boxes tick without caveats. Sales-to-deposits gap under 1.5% across the twenty-four months. Food cost per plate ≤32% and prime cost ≤64% held for three months. A proven fourteen-day absence with sales dropping under 5%. A lease permitting assignment, rent ≤8% of sales, minimum five years remaining.

Closing checklist: how to know the file is genuinely ready

A dossier of three territorial candidates with payback ≤30 months. A nine-folder data room, closed, with its query log. A rehearsed pitch that answers unit economics on slide two. With that file the completion rate moves from one conversation in ten to one in three, and the multiple band shifts from 1.5x-2.5x to 3.5x-5x. Start this week with the bank reconciliation; without it, the other six boxes cannot be audited. History outranks projection. A fund discounts your future numbers almost entirely and prices off the past, so twenty-four reconciled months beat fifty slides; when the history is missing, the investor stops negotiating multiple and starts negotiating control, and that is where you lose your seat. Founder dependence is paid for with a discount. If 85% of critical calls —purchasing, hiring, pricing, costing— pass through you, the buyer is acquiring a job rather than an asset, which is exactly why identical EBITDA fetches 2x in one deal and 4.5x in another.

The four differences that decide the round

Territory stops being a hunch. Territorial prefeasibility, location intelligence applied to the trade, crosses residential and office density, footfall by daypart, area average ticket and category saturation; with eleven variables the opening error drops measurably, and without them you sign five years of rent on a feeling. The file compresses time, and time is price. Every extra round of clarifications cools the investment committee and usually brings a downward revision; a data room closed on day one is the cheapest lever you have to defend your multiple.