Appendix C — The Business Plan Workbook

This appendix is the blank version of the document the book builds. Every chapter ends with a 🍽️ The Business Plan checkpoint that contributes one section of a plan for a constructed example restaurant; here you assemble the same forty sections for your restaurant, with your rent, your menu, your market, and your money.

It is deliberately empty. There is no completed plan in these pages to copy, and that is not an oversight — a finished plan is a thing to admire, and admiring one is the opposite of the work. Where a fragment of an example is genuinely useful to show a format, it appears in two or three lines and is labeled as an illustration. Everything else is yours to fill.

Work it with a spreadsheet open beside you. Most of these sections are arithmetic wearing a paragraph.

C.1 How to use this workbook

Start with the thing nobody tells you.

A business plan is not a document you write to raise money. It is the instrument by which you find out whether the business you are imagining can exist. The version that goes to a lender is a by-product — a cleaned-up export of a much messier private process. Most of a plan's value is produced before anyone else reads a word of it, in the hour when you cost your menu, apply your own rent, and find that the number at the bottom is not the number you assumed.

That hour is the point. Write the plan to persuade and every assumption comes out flatter than it should be, and the plan succeeds at persuading you. Write it to find out and it will occasionally tell you something you did not want to hear, which is the only reason to write it.

Work it in order, and expect to go backwards

The forty checkpoints are sequenced the way the decisions actually depend on each other. Site before floor plan, floor plan before capacity, capacity before revenue, menu before cost cards, cost cards before the P&L. Out of order, you get a plan whose sections each look reasonable and which does not reconcile with itself.

But working in order does not mean working once. Section 27 will invalidate something you wrote in Section 11. You build a cost card in Section 11 at a \$21.00 price with a healthy contribution margin; then in Section 27 you design a marketing plan with a weeknight prix fixe and a happy-hour price, and the item's real average selling price is no longer \$21.00. The margin you banked was never real.

Go back and fix Section 11. That is the workbook functioning correctly, not failing. A plan that never sent you backwards is a plan in which nothing tested anything. Expect three or four passes, and expect the first to be miserable.

Three working rules

1. Write the number, then write where it came from. Every figure here is either derived from another figure in the plan, quoted from a document you can produce, or assumed. Assumed numbers go in the register in C.2. There is no fourth category, and "industry average" without a source is an assumption, not a fact.
2. Do the arithmetic in a spreadsheet, not in prose. Prose hides errors. "Labor should run about thirty-two percent" survives any amount of rereading; a schedule that sums to \$9,140 a week against \$27,000 of sales does not.
3. Date everything. Costs move. A plan built on a February produce sheet and a summer wage market is two plans stapled together. Put a "prepared as of" date on the cover and on the register, and refresh both when the plan sits for a quarter.

This is not a legal or accounting product. Entity choice, licensing, wage and hour law, lease terms, and tax treatment vary by state, county, and city, and they change. The checkpoints tell you what to produce and what to ask; none of them tells you what the law is where you are. Use an attorney and an accountant for anything that binds you.

C.2 The assumptions register

This is the single artifact that separates a legible plan from an assertive one, and it belongs at the front of your working file even though it goes near the back of the finished document.

Every plan rests on numbers nobody can prove. What your average check will be. How many covers you will do on a Tuesday in February. Whether the build-out comes in at the bid. An assertive plan sets those numbers in the same typeface as the rent, which is verifiable, and hopes the reader does not notice the difference. A legible plan marks them.

Add a row to this register every single time you write a number you cannot derive. Not the ones you feel uncertain about — all of them. The dangerous assumptions are rarely the ones that felt shaky when you typed them.

Illustrative fragment, to show the register's register — two rows only, not a model to copy:

Naming your own weakest assumption

This is the move most first-time plan writers cannot make themselves do, and it is the strongest one available to them.

Name your weakest assumption before a reader finds it. Write, in the body: the most sensitive input in this plan is weeknight cover count in months four through nine; it is an estimate, not an observation; here is the range, here is the business at the low end, and here is what we would do.

A reader who finds an unmarked soft number on their own concludes that either you did not know it was soft — in which case you cannot be trusted with the rest of the arithmetic — or you knew and chose not to say. A plan that marks its most sensitive input as low-confidence and shows a range is more persuasive than one that asserts a point estimate, because it demonstrates that somebody in the building can tell a fact from a hope. That capacity is what is actually being evaluated.

⚠️ Where the Money Leaks

The assumption that never gets written down is the ramp.

If your Year 1 revenue is a single annual figure divided by twelve, you have made an assumption so large it is invisible: that month two and month nine are the same business. The opening spike is real, it is short, and the trough after it is where undercapitalized restaurants die. Put the ramp in the register explicitly — the percentage of stabilized volume you expect in each of months one through twelve — mark it low confidence, and check it against Section 33.

Sort the finished register by confidence, low first. The top of that list is your plan's actual risk profile, and it is usually more honest than the risk section you will write in Section 38.

C.3 The arithmetic ties

A plan reconciles or it does not. These are the eleven places yours must tie to itself. They are stated here, at the front, so you build toward them — operators who meet them for the first time at assembly spend a week rebuilding sections in the wrong order.

1. Cost cards foot. Every component on every cost card sums to the stated plate cost, waste allowance included. If a card says \$8.52, the lines above it add to \$8.52.
2. Menu mix sums to 100%. Your projected mix percentages across the menu categories you are engineering total 100.0%, not 98 and not 104.
3. Weighted food cost ties to the COGS food line. Menu mix × each item's food cost, weighted and summed, produces the food cost percentage you used on the P&L. If your cards imply 31.4% and your P&L says 30.0%, one of them is fiction.
4. Pour cost ties to the COGS beverage line. Same test, run separately for beverage. Beer, wine, spirits, and non-alcoholic have materially different costs and a materially different mix.
5. Food COGS + beverage COGS = COGS. In dollars. The blended percentage is a result of that addition and the sales mix, never an input to it.
6. The labor schedule produces the labor line, burden included. Your week-at-a-glance schedule, annualized and grossed up for payroll taxes, workers' compensation, and any benefits, equals the labor dollars on the P&L. Scheduling wages alone and calling it labor understates the line by a substantial margin.
7. Prime cost = COGS + labor, computed from dollars. Add the dollars, then divide. Adding two rounded percentages is how a 60.4% prime cost gets reported as 60.0%.
8. The P&L foots to the dollar. Every subtotal, every year, all three years.
9. Revenue ties to capacity, and covers never exceed seats × turns. Your revenue projection decomposes into covers × average check, and those covers are physically producible in the room you have drawn, in the hours you are open, at a turn rate you can defend.
10. Break-even uses the plan's own fixed/variable split. Not a textbook split. The fixed costs in your break-even are the fixed costs on your P&L, and the contribution margin ratio comes from your own COGS and variable labor.
11. Twelve months of cash forecast reconcile to annual profit, with the differences named. Profit and cash differ for identifiable reasons — timing of receipts and payments, sales tax collected and remitted, principal amortization, capital expenditure, prepaid items, inventory build. Name each difference. An unexplained gap between the two is an error, not a nuance.

Dollar figures are canonical; percentages are rounded displays. Never reconstruct a line item from a percentage. If you need food cost dollars, take the dollars. Multiplying \$1,550,000 by a displayed 30% and calling the product the food COGS line drifts the P&L by hundreds of dollars a category and thousands in total, and the statement will not foot — which a reader finds in ninety seconds with a calculator.

Here is the dependency spine, which is also the order in which a break in the chain propagates:

THE PLAN'S DEPENDENCY SPINE                                  [structure, not values]

  site + lease  ──►  floor plan  ──►  seats × turns ──►  covers ──┐
                          │                                       │
                     hours open ─────────────────────────────►  RevPASH
                                                                  │
  menu ──► cost cards ──► menu mix ──► weighted food cost ──┐     │
                    │                                        │     ▼
  beverage program ─┴──► pour cost ─────────────────────────►├─► REVENUE
                                                             │     │
  org chart ──► staffing guide ──► schedule ──► labor $ ─────┤     │
                                          (+ burden)         ▼     ▼
                                                          COGS + LABOR
                                                                │
                                                          PRIME COST
                                                                │
  occupancy + other operating + G&A ────────────────────────►  P&L
                                                                │
                            ┌───────────────────────────────────┤
                            ▼                                   ▼
                      BREAK-EVEN                          CASH FORECAST
                            │                                   │
                            └──────────► SENSITIVITY ◄──────────┘

Read it left to right and notice that almost everything downstream of the lease is a consequence of it. This is why the site and lease section is Checkpoint 6 and not Checkpoint 26: rent is the one number in the plan you cannot revise later by working harder.

C.4 Part I — The Business Behind the Food (Checkpoints 1–5)

The first five sections establish what the business is, who it is for, what it is called, how the document is organized, and where the money comes from. Readers skim them and writers over-invest in them. Keep them short and specific; the plan is won in Parts III and VII.

Checkpoint 1 · The concept statement and the viability target

WHAT YOU PRODUCE — one paragraph describing the restaurant, plus a five-row table of the financial conditions the plan must satisfy to be worth writing.

IT MUST CONTAIN

• Service style, seat count, cuisine or culinary point of view, days and dayparts, beverage posture.
• The neighborhood and the type of market, in one clause.
• The ownership team's actual experience, stated honestly, including the gap.
• The viability target: a revenue order of magnitude, a prime-cost ceiling, an occupancy ceiling, an operating profit that covers debt service with something left, and a working-capital reserve.

IT MUST TIE TO — Section 31. The revenue order of magnitude here and the Year 1 revenue on the three-year P&L must be the same business. If the P&L lands 30% below this target, you did not discover a new fact — you failed the test you set yourself in Section 1.

HOW IT GOES WRONG — the concept paragraph describes a feeling ("warm, ingredient-driven, neighborhood") instead of a business. Every clause should be falsifiable. "56 dining seats plus 12 at the bar, dinner five nights and weekend brunch, \$40–50 dinner check" is a claim. "Approachable yet elevated" is not.

Checkpoint 2 · Market analysis and the competitive set

WHAT YOU PRODUCE — a market section with daytime and residential population, income and household data for the trade area, and a competitive set table of six to twelve operators.

IT MUST CONTAIN

• A defined trade area — a radius, a drive time, or a walkable district boundary. Defined, then used consistently.
• Public demographic data with the source and vintage named.
• Direct competitors (same check, same occasion) separated from indirect ones.
• What each competitor does better than you will, in writing. This is the credibility test.
• The gap you are entering and why it is a gap rather than a market that has already answered.

IT MUST TIE TO — Section 24. Your average check assumption must sit inside the range this table documents, or you must explain what makes you the exception. A check \$14 above every competitor in the district is a claim requiring evidence.

HOW IT GOES WRONG — the competitive set is a list of restaurants you dislike. Include the busy ones, especially the ones whose food you do not respect, because they are taking the occasions you want. A competitive set with no strong competitors describes a market that does not exist.

Checkpoint 3 · Brand and positioning

WHAT YOU PRODUCE — a one-page positioning statement: name, the promise, the guest you are for, and the operational consequences of both.

IT MUST CONTAIN

• The name, with a note that you have checked availability of the entity name, the domain, and the social handles, and screened for conflicting marks.
• A positioning sentence naming the guest, the occasion, and the alternative you win against.
• Three to five brand attributes, each with the operating decision it forces — a promise of generosity means portion sizes and pour standards, which means a cost card.
• What you are explicitly not.

IT MUST TIE TO — Sections 10 and 22. A brand attribute that produces no menu decision and no service standard is decoration. Trace at least three attributes to a specific line in the menu or the steps of service.

HOW IT GOES WRONG — the brand section is written as copy rather than as constraint. Its job in a business plan is not to sell the reader; it is to commit you to standards that cost money, so that the cost shows up in Parts III, IV, and V rather than surprising you in month five.

Checkpoint 4 · Plan structure and the section map

WHAT YOU PRODUCE — the table of contents for your finished plan, with an owner and a due date against each section, and a one-line note on what evidence each section will rest on.

IT MUST CONTAIN

• All sections in final order (see C.12), with page allocations.
• Which sections go in the body and which become appendices.
• The evidence each section needs: a signed LOI, three contractor bids, a payroll quote, a POS proposal, a health-department checklist.
• The dates by which perishable evidence expires.

IT MUST TIE TO — Section 40. This map is the checklist you audit the assembled package against. If a section on this map has no page in the final document, you either dropped it or absorbed it, and you should know which.

HOW IT GOES WRONG — no evidence column. A plan structured only by topic lets you write all forty sections without ever producing a document that supports one of them, and the missing bids and quotes surface at the worst moment, when a reader asks for them.

Checkpoint 5 · The capital stack and sources & uses

WHAT YOU PRODUCE — a sources-and-uses statement: every dollar the project needs and every dollar that funds it, with the two columns equal.

IT MUST CONTAIN

• Uses: construction, equipment, smallwares and FF&E, pre-opening (labor, training, licensing, initial inventory), professional fees, deposits, and a working-capital reserve.
• A construction contingency, stated as a percentage of the construction number and held separately from the reserve. These are two different pots and confusing them is the classic first-time error.
• Sources: owner injection, any tenant-improvement allowance, equipment lease or finance, debt, investor equity — each with terms, rate, and amortization where applicable.
• The security and personal obligations attached to each source, written plainly.

IT MUST TIE TO — Sections 9, 33, and 31. Uses must equal the pre-opening budget in Section 9; the reserve must equal the reserve you draw down in the cash forecast; the debt terms must produce the debt-service line beneath the P&L.

HOW IT GOES WRONG — the working-capital reserve is set to whatever is left after the build-out is funded. It is a requirement, sized from the cash forecast in Section 33, and if the stack cannot fund it, the project as drawn is not funded. Lenders and investors will each have their own expectations about how much of the owner's own money is at risk; those expectations vary by program, by institution, and by deal, so ask rather than assume.

C.5 Part II — Space (Checkpoints 6–9)

Four sections that convert a location into permanent constraints. Everything here is expensive to change and most of it cannot be changed at all, which is why Part VII is so sensitive to it.

Checkpoint 6 · Site and lease abstract

WHAT YOU PRODUCE — a site analysis and a one-page lease abstract summarizing every economic term of the deal.

IT MUST CONTAIN

• Address, square footage split FOH / BOH / storage, generation of the space (first-generation shell versus second-generation restaurant), and what exists that you can reuse.
• Base rent per square foot, NNN or CAM charges, escalations by year, and total annual occupancy in dollars.
• Term, options, tenant-improvement allowance, free-rent period, who controls the build, and the scope of any personal guarantee.
• Use clause, exclusivity, hours restrictions, assignment and sublet rights, holdover and restoration obligations.
• Landlord-delivered condition, and specifically what the existing infrastructure will not support — hood capacity, grease interceptor, electrical service, gas supply, HVAC make-up air.

IT MUST TIE TO — Section 31. Total annual occupancy from this abstract, escalated correctly, is the occupancy line for all three years. Occupancy as a percentage of sales is an output, never an input.

HOW IT GOES WRONG — the abstract quotes base rent and omits NNN, so occupancy is understated by 15–25% for the life of the plan. Quote the all-in annual dollar figure and show the components beneath it.

Checkpoint 7 · Floor plan and capacity

WHAT YOU PRODUCE — a dimensioned floor plan and a capacity table: seats by type, turns by daypart, covers by day.

IT MUST CONTAIN

• Seat count by zone — dining room, bar, private space, patio and its season — with table mix (two-tops, four-tops, communal, banquette) and the resulting party-size flexibility.
• Aisle widths, accessible routes, restroom counts, and the note that ADA and local code requirements must be verified with your architect and building department.
• Kitchen and dish square footage with the stations it must hold.
• Turns by daypart with the service duration that produces them, and hours of operation.
• Covers per day, per week, and per year, with a seasonality factor.

IT MUST TIE TO — Sections 24 and 31. Covers × average check = revenue, and covers can never exceed seats × turns. If the P&L needs volume this room cannot produce, the plan is broken here, not there.

HOW IT GOES WRONG — turns are assumed rather than derived. A 1.8 turn on a four-hour dinner service means an average table duration under 85 minutes including turnover time. Write the duration, then the turn, and check that your menu and service style can actually run at that pace.

Checkpoint 8 · Entity, licensing, and permits

WHAT YOU PRODUCE — a licensing matrix: every permit, license, and registration required before you can serve, with issuer, cost, lead time, and status.

IT MUST CONTAIN

• Entity type and ownership percentages, with a note that entity choice is an attorney-and- accountant decision with tax consequences that vary by state.
• Federal, state, and local registrations — employer identification, sales tax, withholding — plus music licensing, signage, sidewalk or patio permits, and the resale certificate.
• Health department plan review, food establishment permit, manager certification, food-handler requirements.
• Building permit, occupancy, fire, hood suppression, grease interceptor.
• Alcohol license: type, issuing authority, whether the jurisdiction is quota-limited, transfer cost if applicable, and realistic lead time.

IT MUST TIE TO — Section 9. Every lead time in this matrix is a bar on the pre-opening timeline; the longest one sets your opening date, and it is very often the alcohol license.

HOW IT GOES WRONG — the plan budgets license fees and ignores license time. A six-month alcohol process against a four-month build means paying rent on a finished restaurant that cannot sell a glass of wine. Verify every timeline with the issuing authority directly, and put the date in Section 9.

Checkpoint 9 · Pre-opening timeline and budget

WHAT YOU PRODUCE — a week-by-week timeline from lease execution to opening day, and a pre-opening budget in dollars.

IT MUST CONTAIN

• Critical path with dependencies: permits, demolition, rough-in, inspections, equipment delivery, installation, final inspections, hiring, training, friends-and-family service.
• Rent commencement date versus opening date, and the rent you pay in between.
• Pre-opening payroll — management hired at week minus eight to twelve, hourly staff at minus two to three, all of it paid before a dollar of revenue.
• Initial inventory (food, beverage, smallwares, paper, chemicals), training, uniforms, POS configuration, menu printing, and opening marketing.
• A stated contingency and the decision rule for using it.

IT MUST TIE TO — Sections 5 and 33. The pre-opening budget total is a line in Uses; its weekly cash outflows are the opening weeks of the cash forecast, which is where a delayed opening becomes visible as a number rather than a worry.

HOW IT GOES WRONG — the timeline has no slack and the budget has no delay scenario. Model opening four and eight weeks late and read the cash line. If eight weeks late is fatal, you have learned something about the reserve in Section 5 that no amount of optimism about contractors will resolve.

C.6 Part III — The Product (Checkpoints 10–16)

Seven sections, and the arithmetic heart of the plan. A P&L built on uncosted recipes is a wish with subtotals.

Checkpoint 10 · The menu

WHAT YOU PRODUCE — the opening menu as guests will see it, plus standardized recipes for every item on it.

IT MUST CONTAIN

• Every item by category, with prices, in menu order.
• A standardized recipe per item: ingredients, quantities by weight or count, yield, portion size, and procedure — written so two different cooks produce the same plate.
• Cross-utilization mapped: which ingredients appear on how many items.
• Station assignment per item, so the menu can be checked against the line you drew in Section 14.
• Seasonal change cadence and what stays fixed.

IT MUST TIE TO — Sections 11 and 13. Every item here needs a cost card there; every ingredient here needs a spec and a par there. Menu items with no cost card are the most common single break in the whole plan.

HOW IT GOES WRONG — the menu is too long for the kitchen and too long for the plan. Every added item is another cost card, another spec, another par, another item that ages in the walk-in. If you cannot cost it, you cannot serve it.

Checkpoint 11 · Cost cards and pricing

WHAT YOU PRODUCE — a costed recipe card for every menu item, and a pricing rationale for each.

IT MUST CONTAIN

• Every component priced at as-purchased cost, converted to portion cost through a stated yield.
• A waste and spillage allowance, stated as a percentage, applied consistently.
• Plate cost, menu price, food cost percentage, and contribution margin in dollars.
• The pricing method for each item, and where you deliberately departed from it — anchors, value items, and items you price above the target because they can carry it.

IT MUST TIE TO — Section 31, through Section 12. The mix-weighted food cost from these cards is the food COGS percentage on the P&L. Not a target you chose — the number your own cards produce.

HOW IT GOES WRONG — costing at as-purchased price without a yield factor. A whole fish at \$9.00 a pound that yields 45% is \$20.00 a pound on the plate. Yield is where cost cards lie, and they lie in the direction that makes your food cost look achievable.

Checkpoint 12 · The menu-engineering matrix

WHAT YOU PRODUCE — a projected menu-mix analysis placing every entrée in one of four quadrants, with an action per item.

IT MUST CONTAIN

• Projected units and mix percentage per item, summing to 100%.
• Contribution margin per item in dollars, from Section 11.
• Average CM and average mix, computed, as the axis lines.
• Quadrant assignment: star, plowhorse, puzzle, dog.
• The action per quadrant — protect, re-cost or re-portion, reposition or rename, cut.

IT MUST TIE TO — Sections 11 and 31, on the weighted food cost. Mix × item food cost, summed, must equal the food COGS percentage in the P&L. This is tie number 3 and it is the one most plans fail.

HOW IT GOES WRONG — the projected mix is uniform, or it is flattering. Assigning every item the same mix percentage produces a weighted food cost that means nothing. Project the mix the way the menu is actually designed to sell, and mark it low confidence in C.2, because pre-opening mix is a genuine guess.

Checkpoint 13 · Purchasing, specs, and par levels

WHAT YOU PRODUCE — a purchasing plan: vendor list, product specifications, par levels, order and delivery calendar, and receiving procedure.

IT MUST CONTAIN

• Primary and backup vendor per category, with delivery days and minimums.
• A written spec for every significant item — grade, size, pack, brand or equivalent.
• Par levels by item, tied to delivery frequency and usage.
• Order days, delivery days, and who receives — with the rule that the person who orders is not the only person who checks the invoice against the truck.
• Inventory count cadence and the count sheet's organization (by storage location, in walk-order).

IT MUST TIE TO — Sections 11 and 33. Spec changes change cost cards. Par levels and delivery terms set your inventory investment and your payables timing, both of which land in the cash forecast.

HOW IT GOES WRONG — pars are set to "never run out," which converts cash into inventory that spoils. Par is a function of usage between deliveries plus a safety buffer, and on a two-delivery week it is a much smaller number than instinct suggests.

Checkpoint 14 · The kitchen plan

WHAT YOU PRODUCE — a station-by-station kitchen layout with an equipment schedule and a production plan.

IT MUST CONTAIN

• Station map with each menu item assigned to a station, and the peak-hour item count per station.
• Equipment schedule: item, capacity, utility requirement, cost, new or used, lead time.
• Refrigeration and dry storage capacity checked against par levels from Section 13; ventilation, make-up air, fire suppression, and grease interceptor checked against Section 6's existing condition.
• Prep plan: what is produced daily, what is batched, and the prep hours it takes.

IT MUST TIE TO — Sections 5 and 19. The equipment schedule total is the equipment line in Uses; the prep plan's hours are BOH hours in the staffing guide. Prep hours are the labor cost that scratch cooking buys, and if they are not in the schedule, the labor line is wrong.

HOW IT GOES WRONG — the station map is checked for lunch and not for the peak fifteen minutes of Saturday. Count items per station at the busiest quarter hour the plan projects. A station that must fire nineteen plates in fifteen minutes on equipment that holds eight is a plan for tickets that die in the window.

Checkpoint 15 · The beverage program

WHAT YOU PRODUCE — the beverage program with costed pours, projected beverage mix, and pour cost by category.

IT MUST CONTAIN

• Program by category: beer, cocktails, spirits, non-alcoholic, with the wine list handled in Section 16.
• Costed pours: bottle cost, pour size, pours per bottle, cost per pour, price, cost percentage.
• Projected beverage mix by category and the blended pour cost that results.
• Beverage attachment rate: projected beverage sales as a percentage of total sales.
• Standards for measurement — jiggers or pour spouts, portion control, and the comp and spill policy.

IT MUST TIE TO — Sections 24 and 31, on two numbers: the beverage percentage of the sales mix, and the beverage COGS line. Beverage attachment is one of the highest-leverage assumptions in the plan, because pour cost is materially better than food cost and every point of mix shift moves blended COGS.

HOW IT GOES WRONG — a single blended pour cost is asserted rather than built. Beer, cocktails, and wine have very different cost structures; a blended figure that is not built up from category mix is a guess dressed as a calculation.

Checkpoint 16 · The wine list

WHAT YOU PRODUCE — the opening wine list with cost, price, and margin per bottle, and the by-the-glass program costed separately.

IT MUST CONTAIN

• Bottle count and the list's shape by category, price band, and region.
• Cost, price, and contribution margin in dollars per listing, with the pricing approach stated — and the note that straight multiple pricing produces absurd numbers at the top of a list.
• By-the-glass: pours per bottle, cost per pour, price, and the shelf-life and waste assumption.
• Opening inventory investment in dollars and expected turns per year.
• Storage requirement, checked against Section 7's square footage.

IT MUST TIE TO — Sections 15, 31, and 33. Wine pour cost folds into blended beverage cost; the opening list is a cash purchase in the pre-opening budget and a slow-turning asset thereafter.

HOW IT GOES WRONG — the list is built to the owner's taste and priced by multiple. A 40-bottle list that turns twice a year is capital sitting in a rack. Show the turns, and set the depth of the list from the turns rather than from enthusiasm.

C.7 Part IV — People (Checkpoints 17–21)

Five sections producing the fastest-moving controllable line in the business. Labor is not a percentage you choose; it is a schedule you write.

Checkpoint 17 · Organizational chart and hiring plan

WHAT YOU PRODUCE — an org chart with every position, and a hiring plan with dates, wage rates, and sourcing.

IT MUST CONTAIN

• Every position, salaried and hourly, FOH and BOH, with reporting lines.
• Headcount by position at open and at stabilization.
• Wage or salary per position, benchmarked against your local market and dated.
• Hire dates relative to opening, keyed to the timeline in Section 9.
• Turnover assumption by position and the cost of a single replacement — recruiting, training hours, and reduced output — computed in dollars for at least one position.

IT MUST TIE TO — Sections 19 and 31. Positions and rates here are the inputs to the schedule there; the schedule produces the labor line. If a position exists on this chart and never appears on the schedule, the labor line is understated.

HOW IT GOES WRONG — the org chart omits the owner's salary. If you intend to draw a wage, it is a labor cost and belongs in prime cost. A plan that shows profit only because the operator works free has not demonstrated a business.

Checkpoint 18 · The training plan

WHAT YOU PRODUCE — a training program: content, hours, sequence, and cost, for opening and for ongoing hires.

IT MUST CONTAIN

• Opening training schedule by position with total paid hours, costed at the rates in Section 17.
• Certification requirements — food handler, manager certification, alcohol server training where required — with a note to verify requirements locally.
• Menu and beverage education, including tasting the full menu, with the food and beverage cost of doing so.
• Service standards training keyed to Section 22, and the practice services before opening.
• Ongoing onboarding for a new hire in month seven: hours, trainer pay, and the productivity ramp.

IT MUST TIE TO — Sections 9 and 31. Opening training hours are a pre-opening budget line; ongoing training hours are a recurring labor cost that most plans forget entirely.

HOW IT GOES WRONG — training is budgeted as an event rather than a rate. In an industry with high turnover, training is a continuous production cost. Express it as hours per new hire × hires per year and put the product in the labor line.

Checkpoint 19 · The labor model and the schedule

WHAT YOU PRODUCE — a staffing guide keyed to forecast volume, and a full week-at-a-glance schedule that produces the labor dollars in the P&L.

IT MUST CONTAIN

• The staffing guide: for each volume band, how many of each position work each shift.
• A representative week's schedule at plan volume, and a second at a slow week's volume.
• Weekly wages by position, then the burden — payroll taxes, workers' compensation, benefits — applied as a stated percentage.
• Fixed labor (salaried management, the opening prep cook, the closing dishwasher) separated from variable labor. This split feeds break-even.
• Overtime policy and the assumption about how much overtime the plan carries.

IT MUST TIE TO — Section 31, on the labor dollars. Annualize this schedule with seasonality and it must equal the labor line. This is tie number 6, and burden is where it usually breaks.

HOW IT GOES WRONG — the labor percentage is chosen and the schedule is reverse-engineered to match it. Build the schedule the room actually needs, total it, then read the percentage. If the percentage is unacceptable, change the schedule, the hours, or the menu — not the arithmetic.

Checkpoint 20 · The wage-and-hour compliance review

WHAT YOU PRODUCE — a compliance section stating how you will pay people lawfully in your jurisdiction, and what it costs.

IT MUST CONTAIN

• Applicable minimum wage — federal, state, and local, whichever governs — with the effective date and any scheduled increases.
• Whether your jurisdiction permits a tip credit, and if so, the notice, recordkeeping, and tip-pool rules you will follow. Several states have no tip credit at all.
• Overtime rules, meal and rest break requirements, and any predictive-scheduling or paid-leave ordinance that applies to you.
• Employment eligibility verification, recordkeeping, and classification of every worker — with the note that misclassification is expensive and common.
• Scheduled wage increases modeled into Years 2 and 3.

IT MUST TIE TO — Sections 19 and 31. A scheduled minimum-wage increase in Year 2 changes the Year 2 labor line, and a plan that holds labor flat across three years while wages legislate upward is wrong on its face.

HOW IT GOES WRONG — the section is a paragraph saying "we will comply with all applicable laws." That sentence transmits nothing. Name the actual rules that apply to your address and their cost. All of this varies by state, county, and city and all of it changes — verify locally, with an employment attorney, before you build wages on it.

Checkpoint 21 · The culture statement and retention plan

WHAT YOU PRODUCE — a short statement of how this restaurant treats the people who work in it, and the retention measures that follow, with their cost.

IT MUST CONTAIN

• Two or three commitments that cost money — scheduling practice, break discipline, a wage floor above minimum, a path to advancement, benefits.
• The harassment and complaint policy, and how a complaint reaches someone with authority.
• Your target turnover by position against the industry pattern, and what you are doing to beat it.
• The retention math: replacement cost per position × reduced turnover = dollars saved, netted against what the commitments cost.

IT MUST TIE TO — Sections 17 and 31. If retention measures cost \$1,100 a month, that money is in the labor or benefits line. If they save more than they cost, show the arithmetic; if you cannot show it, fund them anyway and say why.

HOW IT GOES WRONG — the culture statement is aspirational and free. Culture that costs nothing changes nothing. Every credible commitment here has a dollar attached, and a reader looking for seriousness will look for exactly that.

C.8 Part V — Service (Checkpoints 22–26)

Five sections covering the thing you actually sell — where a plan stops being a manufacturing document and becomes a revenue model.

Checkpoint 22 · Service standards

WHAT YOU PRODUCE — written steps of service by daypart, with timing standards and the staffing they require.

IT MUST CONTAIN

• Steps of service from greet to farewell, with a time standard on each (greet within 60 seconds, first drink within a stated window, and so on).
• Section sizes by position and the resulting guest-to-server ratio.
• Table-turn timing: the courses, the pacing, and the total duration that produces the turns in Section 7.
• Reservation and waitlist policy, pacing rules, and how large parties are handled.
• Who runs the pass and how tickets are fired.

IT MUST TIE TO — Sections 7 and 19. The service duration here produces the turn assumption there; the guest-to-server ratio here produces the FOH headcount on the schedule.

HOW IT GOES WRONG — the standards describe fine-dining service on a casual-dining labor model. Count the touches your standards require, multiply by covers, and check that the schedule contains the hours. Standards the schedule cannot fund are not standards.

Checkpoint 23 · Guest experience and recovery

WHAT YOU PRODUCE — the feedback and recovery system, with a comp policy and a stated comp budget.

IT MUST CONTAIN

• How complaints are surfaced in the room, and who is empowered to fix one without asking.
• The comp and void policy: who may authorize, at what dollar level, and how it is recorded.
• Comps as a budgeted percentage of sales, with a review cadence.
• Review and feedback monitoring: which platforms, who responds, how fast, in what voice.
• The repeat-visit assumption — what share of covers you expect to be returning guests, and why.

IT MUST TIE TO — Sections 24 and 31. The comp percentage is a real line that reduces net sales; the repeat-visit rate is a load-bearing input to the revenue model, because a first visit barely covers the cost of acquiring it and a second visit contributes nearly all of its margin.

HOW IT GOES WRONG — comps are unbudgeted and unmonitored. Comps that run 2% instead of 0.8% on a \$1.4M restaurant are roughly \$17,000 a year, which arrives entirely out of profit and shows up on no report anyone reads.

Checkpoint 24 · The revenue model and RevPASH

WHAT YOU PRODUCE — a revenue build by daypart and day of week, and a RevPASH analysis of the room's seat-hours.

IT MUST CONTAIN

• Covers by daypart by day, from Section 7, with seasonality.
• Average check by daypart, built from the menu and beverage attachment rather than assumed.
• Revenue by daypart, week, and year, split food and beverage.
• RevPASH — revenue ÷ (available seats × hours open) — by daypart, which identifies the hours the room is not earning.
• Which dayparts carry themselves and which are subsidized, stated plainly.

IT MUST TIE TO — Sections 7, 12, 15, and 31. Covers ≤ seats × turns; average check must be consistent with the menu prices and the beverage attachment; food/beverage split must match the mix used in COGS. This is tie number 9.

HOW IT GOES WRONG — a single annual revenue number divided by twelve. Revenue is a build, not a figure. Build it by daypart, and the weak daypart will announce itself before you have signed a lease that requires it.

Checkpoint 25 · The food safety plan

WHAT YOU PRODUCE — a written food-safety program: the hazard controls, the logs, the training, and the accountability.

IT MUST CONTAIN

• HACCP-style identification of hazards by process, with the critical control points named.
• Temperature standards consistent with the FDA Food Code framing your jurisdiction has adopted — cold holding at or below 41°F, hot holding at or above 135°F, poultry to 165°F, the temperature danger zone — with the note that local adoption varies and yours governs.
• Logs — receiving, cooling, holding, cleaning — with who signs and who audits; sanitizer type and concentration (quaternary ammonium commonly 200–400 ppm) with test strips and verification.
• Allergen procedure, consumer advisory if you serve raw or undercooked items, and the person certified in food protection management.
• Self-inspection cadence, using the health department's own checklist.

IT MUST TIE TO — Sections 8, 18, and 34. Certification requirements from the licensing matrix, training hours from the training plan, and log review from the controls calendar. A safety program that nobody is scheduled to verify is a binder.

HOW IT GOES WRONG — the plan describes food safety as a value rather than a system. Name the control, the measurement, the record, and the person. A health inspection is a systems audit, and a consequential failure — no hot water, a sewage backup, a serious temperature violation — can close your doors the same day and blow a hole in the cash forecast.

Checkpoint 26 · The technology stack

WHAT YOU PRODUCE — a system inventory: what you will run, what each costs, what it connects to, and what it produces.

IT MUST CONTAIN

• POS: hardware, licenses, monthly cost, payment processing rate and per-transaction fee.
• Reservations and waitlist, scheduling and time-clock, inventory and recipe costing, accounting and payroll.
• Integration map: which system feeds which, and where data has to be re-entered by hand.
• Total technology cost — up-front and monthly — and the credit-card processing cost as a percentage of sales, which is one of the largest lines in "other operating."
• The reports you will actually read weekly, named, and who produces them.

IT MUST TIE TO — Sections 31 and 34. Monthly software and processing costs are other-operating lines; the reports named here are the mechanism by which the controls in Section 34 happen at all.

HOW IT GOES WRONG — processing fees are omitted or guessed. At 2.5–3.5% of card sales on a business with a mid-single-digit margin, card processing is not a footnote. Get a real rate quote, including the per-transaction fee, and put the annual dollar figure in the P&L.

C.9 Part VI — Channels (Checkpoints 27–30)

Four sections about revenue that does not walk through your front door. Each has a different cost structure from dine-in, and treating them as "extra sales" is the fastest way to add volume and subtract profit.

Checkpoint 27 · The marketing plan

WHAT YOU PRODUCE — a twelve-month marketing plan with a budget, a calendar, and a measurement method.

IT MUST CONTAIN

• Opening marketing separated from ongoing marketing, with separate budgets.
• Channels with cost per channel and what each is meant to produce — trial, frequency, or check.
• The calendar: what runs when, tied to the seasonality in Section 24.
• Any promotional pricing — prix fixe, happy hour, weeknight offers — with its effect on average check and contribution margin computed, not assumed.
• How you will measure: the metric, the baseline, and the review date.

IT MUST TIE TO — Sections 11, 12, and 24. This is the section that sends you backwards. A promotional price changes an item's effective selling price, which changes its contribution margin, which changes the menu-engineering matrix and the weighted food cost. Go back and fix the cost cards.

HOW IT GOES WRONG — marketing is a percentage of sales with no plan attached, and discounts are treated as marketing rather than as margin. A 20% discount on a 30% food cost item does not cost you 20% — it costs you a much larger share of that item's contribution, and the arithmetic is worth doing before the promotion, not after.

Checkpoint 28 · The off-premise strategy

WHAT YOU PRODUCE — a decision on takeout and delivery, with a channel-level contribution analysis for each channel you enter.

IT MUST CONTAIN

• Channels: direct takeout, direct delivery, third-party marketplace, and the commission rate for each.
• Packaging cost per order, itemized, and its effect on food cost percentage for off-premise items.
• Which menu items travel and which do not, and the resulting off-premise menu.
• Contribution per off-premise order after commission and packaging, compared with the same order dine-in.
• Operational impact: who makes it, where it stages, and how it affects ticket times for the dining room.

IT MUST TIE TO — Sections 24 and 31. Off-premise revenue is part of the revenue build; its commissions and packaging are costs that a dine-in-derived COGS and other-operating line do not contain. Model the channel separately or the blended margin is wrong.

HOW IT GOES WRONG — third-party commission is ignored in the margin math. A commission in the range commonly charged by marketplace platforms, plus packaging, can consume most or all of an item's contribution margin. Several cities have passed commission-cap ordinances; check whether yours has, and do the arithmetic per order before you sign.

Checkpoint 29 · Events and catering

WHAT YOU PRODUCE — a private-events and catering program with pricing, minimums, contracts, and capacity limits.

IT MUST CONTAIN

• What you sell: buyouts, semi-private, off-site catering, and which of those the room can support.
• Pricing structure: per-person, food and beverage minimums, service charge, deposit, and cancellation terms.
• The displacement calculation — what a buyout earns versus what the same seats would have earned on that night at normal volume.
• Incremental labor and any rental or transport cost.
• Contract terms and who signs them.

IT MUST TIE TO — Sections 24 and 31. Event revenue is a distinct line in the revenue build with its own margin. Displacement is the tie people miss: a Friday buyout that grosses less than a normal Friday is a loss wearing a large invoice.

HOW IT GOES WRONG — a service charge is treated as revenue without resolving how it is distributed and disclosed. Service-charge treatment and disclosure requirements vary by jurisdiction and have real wage-and-hour consequences; settle it with your attorney and your accountant before you print a contract.

Checkpoint 30 · Alternate formats

WHAT YOU PRODUCE — a short, honest assessment of any additional format you are considering, or an explicit statement that you are not.

IT MUST CONTAIN

• The format under consideration — retail, packaged product, a cart or truck, a bar-only concept, a ghost or shared-kitchen channel, a wholesale line.
• Incremental capital required and where it comes from.
• Incremental licensing, which is often the deciding constraint.
• Management attention required, stated in hours, and whose hours they are.
• The go/no-go criterion and the date it will be evaluated.

IT MUST TIE TO — Sections 5 and 35. Any format requiring capital appears in Uses or is explicitly deferred to the growth criteria in Section 35. Do not fund a second business inside a plan for the first one.

HOW IT GOES WRONG — an alternate format is included to make the plan look bigger. A reader reads an unfunded second concept as a founder who is not yet committed to the first. "We have considered X and will not pursue it before Year 3" is a stronger sentence than a half-built plan.

C.10 Part VII — The Money (Checkpoints 31–34)

Four sections in which every earlier section either reconciles or does not. Built honestly, this part is mostly assembly. Built otherwise, this is where you find out.

Checkpoint 31 · The three-year P&L

WHAT YOU PRODUCE — a three-year projected profit-and-loss statement, monthly for Year 1 and annual for Years 2 and 3, in dollars with percentages displayed alongside.

IT MUST CONTAIN

• Revenue split food and beverage, from Section 24, with the Year 1 ramp by month.
• COGS split food and beverage (Sections 12, 15, 16), and labor with burden (Section 19) including salaried management and the owner's wage. Prime cost gets its own line.
• Occupancy from Section 6, escalated per the lease.
• Other operating, built line by line — utilities, supplies, repairs, insurance, marketing, technology, card processing — not as a plug.
• G&A, then operating profit, then debt service and its split between interest and principal.

| Line | Y1 \$ | Y1 % | Y2 \$ | Y2 % | Y3 \$ | Y3 % | |---|---|---|---|---|---|---| | Food sales | | | | | | | | Beverage sales | | | | | | | | Total revenue | | 100.0% | | 100.0% | | 100.0% | | Food cost | | | | | | | | Beverage cost | | | | | | | | Total COGS | | | | | | | | Hourly wages | | | | | | | | Salaried wages | | | | | | | | Payroll burden | | | | | | | | Total labor | | | | | | | | PRIME COST | | | | | | | | Occupancy | | | | | | | | Other operating | | | | | | | | General & administrative | | | | | | | | Operating profit | | | | | | | | Debt service | | | | | | | | Net | | | | | | |

IT MUST TIE TO — everything upstream, and to Section 33 downstream. Every ratio in C.3 lands here.

HOW IT GOES WRONG — the P&L is built top-down from target percentages. It reads beautifully and it is worth nothing, because it contains no information that was not assumed. Build every line from its own section and let the percentages be whatever they are. If prime cost comes out at 66%, that is the plan telling you something in the only language it has.

Checkpoint 32 · Break-even and sensitivity

WHAT YOU PRODUCE — a break-even analysis in sales dollars and in covers per night, plus a sensitivity table on the plan's three or four most load-bearing drivers.

IT MUST CONTAIN

• Your own fixed/variable split, taken from Section 31 and Section 19's fixed-labor line.
• Contribution margin ratio, computed from your COGS and variable labor.
• Break-even sales, break-even covers per year, and break-even covers per night by daypart.
• The margin of safety: plan covers minus break-even covers, in covers and as a percentage.
• Sensitivity on the drivers that actually move the outcome — covers, average check, food cost, labor rate.

| Driver | Plan value | Low case | Operating profit, low \$ | High case | Operating profit, high \$ | |---|---|---|---|---|---| | Covers per night | | | \$ | | \$ | | Average check | | \$ | \$ | \$ | \$ | | Food cost % | | | \$ | | \$ | | Labor rate / hours | | | \$ | | \$ |

Vary one driver at a time; a table in which everything moves at once teaches nothing about which input matters.

IT MUST TIE TO — Section 31 for the split, Section 24 for the check, Section 7 for the ceiling. This is tie number 10, and the ceiling matters: if break-even covers per night exceed seats × turns, the room cannot break even at any level of effort, and no other section of the plan can rescue that.

HOW IT GOES WRONG — break-even is computed with all labor treated as variable, which understates fixed costs badly and produces a comfortable number. A meaningful share of restaurant labor — management, the opening prep cook, the closing dishwasher — is paid whether you do 40 covers or 140. Put it on the fixed side where it belongs.

Checkpoint 33 · The cash flow forecast

WHAT YOU PRODUCE — a thirteen-week rolling cash forecast for the opening quarter and a monthly cash forecast for the first twenty-four months, with the low-water mark identified.

IT MUST CONTAIN

• Beginning cash, receipts, and every category of disbursement, weekly.
• The timing differences that make cash unlike profit: card settlement lag, vendor terms, biweekly payroll, monthly rent, quarterly or monthly sales-tax remittance, insurance and license lumps.
• Sales tax shown as collected and then remitted — it was never your money, and it inflates the balance until the day it leaves. Debt service split into interest and principal, since principal reduces cash but not profit.
• The reserve drawn down explicitly, with the week it reaches its lowest point stated in dollars.
• A delayed-opening scenario and a slow-ramp scenario.

IT MUST TIE TO — Sections 5 and 31. Twelve months of this forecast must reconcile to Year 1 profit with every difference named — tie number 11 — and the low-water mark sizes the working-capital reserve in the capital stack. If the reserve in Section 5 is smaller than the trough here, go back and change Section 5.

HOW IT GOES WRONG — the forecast assumes revenue and expense land in the same month they are earned and incurred. They do not, and the gap between them is exactly where profitable restaurants run out of money. Model the timing, not the accrual.

Checkpoint 34 · Financial controls

WHAT YOU PRODUCE — a written controls program: who counts what, who reconciles what, who reviews what, and on what day of the week.

IT MUST CONTAIN

• Weekly: inventory counts (food and beverage), purchases, payroll, and a one-page flash report carrying prime cost.
• Daily: cash handling, deposit, void and comp review, and the shift-report reconciliation.
• Monthly: bank reconciliation, P&L review against plan with variance explanations, and vendor statement review.
• Separation of duties — who orders, who receives, who approves invoices, who signs checks — and what you do when the answer to all four is "the owner," because in a small operation it often is.
• The variance thresholds that trigger action, and the action.

IT MUST TIE TO — Sections 26 and 31. The reports come from the systems in Section 26; the plan figures are the benchmarks the variance is measured against. Ideal versus actual food cost is the mechanism, and it only exists if someone counts.

HOW IT GOES WRONG — controls are described as intentions rather than as scheduled tasks with names against them. "We will monitor food cost closely" is not a control. "The chef counts the walk-in every Sunday night, the flash report is on the owner's desk Monday at 10 a.m., and a variance over 1.5 points triggers a line-item usage review" is one.

C.11 Part VIII — What Comes Next (Checkpoints 35–40)

Six sections on what happens after the plan works, what happens if it does not, and how the document is assembled and defended.

Checkpoint 35 · Growth criteria

WHAT YOU PRODUCE — the written conditions that must be true before you consider a second anything, decided now while you are calm.

IT MUST CONTAIN

• Financial thresholds: sustained prime cost, sustained operating margin, months of consecutive performance, and reserve restored to a stated level.
• Operational thresholds: a general manager and a chef who can run the first unit without you, with documented systems.
• The date, no earlier than which, the question may be asked.
• Who decides, and what evidence they will look at.

IT MUST TIE TO — Sections 31 and 34. Every threshold must be a number the controls program actually produces, or it is unmeasurable and will be waived on a good quarter.

HOW IT GOES WRONG — the criteria are written after the opportunity appears. Write them now, when no attractive second site is in front of you, because a plan written in advance is the only protection against the version of you who has just walked through a beautiful vacant room.

Checkpoint 36 · Expansion analysis

WHAT YOU PRODUCE — a pro-forma for the next unit or the next format, built to the same standard as this plan and stated as a scenario, not a commitment.

IT MUST CONTAIN

• Capital required and its source, with the effect on the first unit's balance sheet and guarantees.
• What transfers (systems, recipes, brand, purchasing power) and what does not (the specific location, the specific staff, the specific market).
• The management structure required, including who runs unit one on the day you are at unit two.
• Second-unit economics: a separate P&L, not a copy of the first with the address changed.
• The downside case: what happens to unit one if unit two underperforms.

IT MUST TIE TO — Sections 5 and 35. Expansion capital either exists as a source in a future stack or is a future financing, and either way it is not in the current plan's Uses.

HOW IT GOES WRONG — the second unit is modeled at the first unit's stabilized volume from day one, and its ramp is skipped. Every unit ramps. Model unit two's Year 1 the way you modeled unit one's Year 1, including the trough.

Checkpoint 37 · Sustainability

WHAT YOU PRODUCE — a practical sustainability section: waste, energy, sourcing, and what each one costs or saves in dollars.

IT MUST CONTAIN

• Food waste: where it occurs, how it is measured, and the cost of the current level.
• Energy and water: equipment efficiency, hood and HVAC scheduling, and any available utility rebate.
• Sourcing commitments and their cost premium, stated honestly, with the marketing value stated separately and not double-counted.
• Waste stream handling: recycling, composting, oil, and disposal costs.
• Any claim you intend to make publicly, and the evidence that supports it.

IT MUST TIE TO — Sections 11 and 31. A sourcing premium changes cost cards; efficiency measures change the utilities line. A sustainability section with no effect anywhere in the financials is decorative.

HOW IT GOES WRONG — the section makes claims the operation cannot substantiate. Claim only what you can document, because a sourcing claim you cannot support is a credibility problem in a plan whose entire value is credibility.

Checkpoint 38 · The risk register and contingency plan

WHAT YOU PRODUCE — a risk register with likelihood, impact in dollars, mitigation, and trigger, plus a written contingency plan for the two or three scenarios that could actually end the business.

IT MUST CONTAIN

• Risks named specifically: construction delay, permit delay, key equipment failure, loss of the chef or the GM, a bad first inspection, a slow ramp, a rent escalation, a major road or transit disruption, a demand shock.
• Dollar impact and duration for each, not a severity word.
• Mitigation already funded (insurance, reserve, cross-training, backup vendors) versus mitigation that would need to be funded.
• The trigger for each: the observable condition at which you act, decided in advance.
• The contingency plan: what gets cut, in what order, and what the reduced operating model looks like.

IT MUST TIE TO — C.2 and Section 33. Your lowest-confidence assumptions should reappear here as your highest-ranked risks; if they do not, one of the two documents is not honest. Each downside scenario should be runnable through the cash forecast.

HOW IT GOES WRONG — the register lists risks and stops. A risk with no trigger and no dollar figure is a worry. The trigger is the whole value of the exercise, because it converts a decision made in a panic into a decision made in advance.

Checkpoint 39 · Exit and succession

WHAT YOU PRODUCE — a short section on what happens to this business when an owner leaves, voluntarily or otherwise.

IT MUST CONTAIN

• Ownership structure, vesting, and the buy-sell provisions among partners.
• What happens on death, disability, or a partner wanting out — and whether it is funded.
• Key-person dependency: what only one person knows, and the plan to document it.
• Transferability: whether the lease is assignable, whether licenses transfer, what a buyer would actually be buying.
• The realistic exit paths — sale as a going concern, sale to a partner or a manager, orderly wind-down — and what each requires.

IT MUST TIE TO — Sections 5, 6, and 8. Personal guarantees, lease assignment rights, and license transferability determine whether an exit is possible at all. A guarantee that survives a sale is a fact you want to know now.

HOW IT GOES WRONG — the section is skipped because the business has not opened. Partnerships dissolve most painfully when nothing was written down while everyone was friendly. Two pages now prevents a catastrophe later, and any serious reader notices whether they are there.

Checkpoint 40 · The assembled package

WHAT YOU PRODUCE — the finished document: the full plan in reader order, with the executive summary written last and a reconciliation summary that proves the numbers agree.

IT MUST CONTAIN

• Every section from C.4 through C.11, in the order set out in C.12.
• The executive summary — two pages, written after everything else — carrying the concept, the ask, the use of funds, the headline financials, and the team.
• Supporting evidence in appendices: cost cards, the full schedule, the lease abstract, resumes, bids and quotes, market data, and the assumptions register.
• The one-page reconciliation summary described in C.12.
• A full pass against the eleven ties in C.3, documented.

IT MUST TIE TO — all thirty-nine preceding checkpoints. This is the audit, and it is where the plan either holds together or is revealed to be thirty-nine documents in a binder.

HOW IT GOES WRONG — the package is assembled without a final arithmetic pass, because by this point you are tired of it. Do the pass. Every number in the executive summary must be traceable to a page in the body, and a reader will spot-check exactly the ones you did not re-derive.

C.12 Assembling the package

You have forty sections in working order. The finished document goes in reader order, which is different, because it is organized for someone who does not yet know anything about your business and is deciding in fifteen minutes whether to keep reading.

READER ORDER — the finished document

  1  Cover page + table of contents            date it; number every page
  2  Executive summary                         2 pages, written LAST
  3  Concept and business description          from Checkpoints 1, 3
  4  Market analysis and competitive set       from Checkpoint 2
  5  Site, lease, and facility                 from Checkpoints 6, 7
  6  Menu and product                          from Checkpoints 10, 15, 16 (cards to appendix)
  7  Operations                                from Checkpoints 13, 14, 22, 25, 26, 34
  8  Management, staffing, and culture         from Checkpoints 17-21
  9  Marketing and channels                    from Checkpoints 27-30
 10  Financial plan                            from Checkpoints 31, 32, 33
       - assumptions summary
       - three-year P&L
       - break-even and sensitivity
       - cash flow forecast
       - sources and uses (Checkpoint 5)
 11  Risk, growth, and exit                    from Checkpoints 35, 36, 38, 39
 12  Reconciliation summary                    one page
 13  Appendices                                evidence, in the order the body cites it

Write the executive summary last, and write it from the finished numbers. Every figure in it must already exist in the body. A summary drafted early always carries an aspiration the financial section later contradicted, and a reader who finds a headline number that does not match the P&L has learned that the document was not checked — which they will then assume about everything else.

Two pages. The concept in a paragraph. The market in a paragraph. The team, with the gap named and the answer to it. The ask: how much, in what form, for what. The headline financials: Year 1 revenue, prime cost, operating profit, break-even covers per night, and the low-water cash point. Then one sentence stating what makes this business work and one stating the largest risk to it. If you cannot write those two sentences, the plan is not finished.

Body versus appendices

The body is the argument. The appendices are the evidence. The test is whether a reader needs the document to follow the argument or to verify it.

Body: concept, market, site summary, the menu as guests see it, the operating approach, the management structure, the marketing plan, the three-year P&L, break-even, the cash forecast, the sources and uses, and the risk section.

Appendices: every cost card, the full week's schedule, the lease abstract, resumes and personal financial statements, contractor bids and equipment quotes, the licensing matrix, the market data with sources, the assumptions register, and the monthly Year 1 P&L.

If the body runs past about thirty pages, something in it belongs behind the tab.

The reconciliation summary — and why producing it is itself a test

One page, at the end of the body. It lists every headline number in the plan and the page on which that number is derived.

Here is the part worth understanding: you will not be able to complete this page if the plan does not reconcile. You reach the food cost row, go to fetch the derivation, find the cost cards imply 31.4% while the P&L says 30.0%, and catch a break you would otherwise have shipped.

That is what the page is for. It reads as a courtesy to the reader and it functions as an audit of the author. Fill it in by hand, from the pages, without copying from your spreadsheet. A number that cannot be traced to a page is not derived — it is asserted — and it belongs in the register instead.

C.13 The submission

Three kinds of reader will look at this document, and they read for different things. Knowing which one you are in front of tells you what to have ready.

A lender reads for repayment. Their question is not whether your restaurant is a good idea; it is whether the business can service the debt and what happens if it cannot. They will go to the financial section first, and inside it to the cash flow and the debt service. The general concept they are testing is debt service coverage:

$$\text{DSCR} = \frac{\text{net operating income}}{\text{total debt service}}$$

A ratio of 1.0 means every dollar of operating income goes to the loan payment and nothing is left for a slow February, an equipment failure, or the owner. Lenders look for a cushion above that; how much, and how they define the numerator, varies by institution, program, and deal. Ask yours rather than guessing. They will also look at your injection, your collateral, your personal credit and financial statement, your industry experience, and your reserve. Lenders commonly attach conditions to a loan — ordinary and expected — and what those are depends entirely on the lender, the program, and the specific credit. Do not plan around an imagined one.

An investor reads for return and for the people. They spend more time on the concept, the market, and the team than a lender does, and they go to growth and exit before break-even. Their questions are about upside and about what their money buys — ownership percentage, distribution priority, governance, and how they eventually get out. An investor is buying a partnership; a lender is buying a payment stream.

A landlord reads for the ability to pay rent for the length of the term: the capital stack (is this project funded?), the operating experience, the concept's fit with the rest of their property, and the guarantee behind the lease. They are also, quietly, assessing whether your restaurant improves the building. A landlord who believes in the concept is often the most flexible party in the deal on tenant improvement and free rent, and the least flexible on term.

What gets asked

Prepare answers to these, because some version of each is nearly always asked:

• Where did your revenue number come from, and what happens at 80% of it?
• What is your weakest assumption? (You should already have answered this in the plan itself.)
• Walk me through your food cost — how did you get to that number?
• Who runs the restaurant when you are not there?
• What is your worst week of the year, and how do you get through it?
• How much of your own money is in this, and what happens if the project runs over?
• What have you signed, and what have you not signed?

⚖️ Code and Compliance

Before anything is submitted, have the documents that bind you reviewed by professionals. A lease, an operating agreement, a personal guarantee, and a loan commitment are long-term obligations whose terms vary enormously and whose consequences are not intuitive. Entity choice, wage-and-hour compliance, tip handling, and license transferability all vary by state, county, and city. An attorney and an accountant cost a small fraction of what a term you did not understand will cost you over ten years.

The last thing to do before you send it

Give the plan to someone who does not know your business and ask them to find the number they do not believe. Not to be encouraging — to find it. Whatever they name is what the reader who matters will name, and you would rather hear it now.

C.14 When the plan does not work

Sometimes you will finish this workbook and the answer will be no.

The rent is \$118,000 a year, the room seats 62, the check the neighborhood will bear is \$38, and when you multiply covers by check and subtract a prime cost you have built honestly from your own cost cards and your own schedule, there is not enough left to cover occupancy, service the debt, and pay you. You will check it three times. It keeps coming out the same way, because arithmetic does.

That is not a failure of the workbook. That is the workbook doing the only genuinely valuable thing it can do.

Here is the honest frame. Roughly one in four restaurants does not reach its first anniversary, and something close to six in ten are gone within three years. Those are not the folklore numbers — ninety percent in year one has never been demonstrated and is not true — and the real pattern is more instructive, because it means most casualties are businesses that worked, for a while, on a cost structure that was two or three points wrong the entire time. A plan that does not clear its obligations on paper is describing that cost structure before you have signed for it.

Finding that out for the cost of a spreadsheet is the single best outcome this appendix can produce. Not a consolation. The best outcome. Sixty hours and no money bought you what the alternative costs: a ten-year lease, a personal guarantee, a build-out you funded, your savings, and two or three years of the hardest work of your life to reach the same answer — at which point the answer takes everything you put in, and it is still no.

Some restaurants should not be opened. That is not defeatism; it is the reason to do the arithmetic.

What to do with a no

A no is usually a no to a configuration, not to a career. Before you accept it as final, check which variable is carrying it, because the plan will tell you:

• If occupancy is the problem, it is the site, not the concept. The same restaurant at \$26 a foot instead of \$34 may clear easily. Go back to Checkpoint 6.
• If the check average is the problem, it is the market or the menu. A concept that needs \$18 more per guest than the district has ever paid is a concept in the wrong district.
• If prime cost is the problem, look at which half. High food cost with low labor is a purchasing and portioning issue. Low food cost with high labor is a production model that may not fit the volume — a real question about whether the concept can be executed at this scale.
• If capacity is the problem — break-even covers exceeding seats × turns — no amount of effort fixes it. The room is too small for the cost base, or the service is too slow for the room.
• If it is only the ramp, the business may be sound and merely underfunded. That is a capital problem with a different set of answers than a viability problem. Do not confuse the two, in either direction.

Change one thing at a time and re-run the ties in C.3. You may find the fourth configuration works. You may find that none of them do, and that the version of this restaurant that clears its obligations is not a restaurant you want to run — which is also a real answer, and a better one to reach in a spreadsheet than in year two.

The one thing you must not do

Do not adjust the assumptions until the plan says yes.

It is the easiest thing in the world. Move covers up eight a night. Move food cost down a point and a half. Push the labor number by trimming two shifts you know the room will need. Each individual change is defensible. Together they produce a document that clears every hurdle and describes a restaurant that does not exist, and you will then go and sign a lease on the strength of it.

That is the mechanism. Not bad luck, not a hard industry, not a cruel market — a spreadsheet that was edited until it agreed. Every operator who has handed back a set of keys can tell you which cell they changed.

The plan is not the goal. Knowing is the goal. Work it honestly, name what you assumed, mark what you are least sure of, and let it tell you what it tells you.

Then decide.