Food Hall Operator Bookkeeping: Percentage Rent by Vendor, CAM True-Ups, and One POS Across a Dozen Kitchens

Last Friday your hall rung up $58,000 — the best night since opening. Tacos, ramen, a couple hundred cocktails, all ordered from one app, paid at one kiosk, deposited into one bank account. Here is the question that decides whether your business survives: how much of that $58,000 is actually yours?

Some of it belongs to the noodle vendor whose sales triggered her percentage rent. Some of it is sales tax you collected on behalf of a dozen separate businesses. Some of it is yours outright — the bar program, the service fees, tonight's private event deposit. A food hall operator is not a restaurant with extra registers. You are running a landlord, a bar, an events business, and a payments platform simultaneously, and all four share one till. The bookkeeping has to be built for that reality from day one, because the segment is growing fast — a 2026 industry census counted 458 operating U.S. food halls, up nearly 25% in two years, with over 100 more projects in the pipeline — and the halls that fail rarely fail on concept. They fail on cash they couldn't account for.

You're Running Four Revenue Models in One Ledger​

Before touching journal entries, get clear on what the business actually is. Industry benchmarking breaks operator revenue into five streams, and each one has a different margin profile and a different bookkeeping treatment:

The operational insight hiding in that table: a hall without a bar is a landlord, and a landlord's economics are thin. A hall with a strong bar is a hospitality business. In a typical mid-size, ten-vendor hall, vendor food revenue of $6.5 million at a 12% rent rate produces about $780,000 of operator rent income — but adding a $2.5 million bar program can take total operator revenue past $3 million and swing EBITDA from low single digits to the twenties. The bookkeeping lesson is structural: these streams must never be blended. Percentage rent, bar sales, service fees, and event income belong in separate income accounts, because each has a different cost attached, a different tax conversation, and a different story to tell when a number moves.

How Percentage-Rent Deals Are Actually Structured​

Vendors in food halls typically pay for space one of three ways:

1. Straight percentage of gross sales. The purest food hall model. Rates run roughly 8–15%, varying by concept: high-volume concepts (burgers, pizza, high-turnover bowls) pay 8–10%, standard concepts pay 10–13%, and low-volume or specialty stalls — coffee, dessert, niche cuisines — pay 12–15%. The logic: the operator takes more of a margin-light concept's top line and less of a margin-heavy one's.
2. Base rent plus percentage over a breakpoint. A hybrid drawn from shopping-center leasing: base rent of $2,000–$6,000 per month per stall, plus 5–10% of gross sales above an agreed threshold. The breakpoint is often "natural" — annual base rent divided by the percentage factor — so a $36,000 base at 8% sets a $450,000 breakpoint, and percentage rent applies only to sales above it.
3. Flat stall fee. Rarer, and generally a warning sign for the operator's negotiating position, because it caps your upside and disconnects your income from the hall's performance.

Worked example for a ten-vendor hall: ten vendors averaging $650,000 in annual in-hall sales is $6.5 million of vendor food revenue. At an average 12% rent rate, that's $780,000 of operator rent income — before the bar, fees, or events contribute a dollar. Which makes the obvious point: the accuracy of your percentage-rent collection is worth more than most cost-cutting you'll ever do. A systematic 2% under-collection — the wrong sales base, missed categories, unreported delivery orders — is a $13,000-a-year leak in that hall, and much more in a larger one.

The Gross-Sales Definition Is Where the Money Leaks​

Every percentage-rent lease hinges on one phrase: "gross sales." The exclusions negotiated into that definition are where operators quietly lose money or accidentally overcharge — either of which becomes a dispute at renewal. The common ones, each needing a matching rule in your books:

• Sales tax collected. Never part of gross sales. If your POS blends tax into the sales figure it reports per vendor, you're overcharging every vendor every month.
• Tips and gratuities. Excluded almost universally — they belong to staff, not to the sales base.
• Refunds and voided orders. Deduct in the period they occur, matched against the original sale.
• Gift cards. Sales of gift cards typically count only when redeemed, not when purchased — and a card sold by one vendor and redeemed at another needs a rule before it happens, not after.
• Employee meals and discounts. Usually excluded; usually miscoded at the register.
• Delivery and off-site catering. Here's the big one: many vendor agreements cover only in-hall sales, while third-party delivery commissions (15–30%) eat the margin on off-site orders. Whether delivery sales count toward gross sales must match between the lease, the POS configuration, and your rent calculation — the three disagreeing is the single most common percentage-rent dispute.

The practical fix is to encode the lease's gross-sales definition into your POS at the vendor level, so the "reportable sales" number the system produces is contractually correct by construction, and to send each vendor a monthly statement showing reported sales, each exclusion applied, and the resulting rent. That statement is your audit trail. Operators who skip it end up reconstructing a year of rent from raw register data when a vendor disputes — exactly the manual-reconciliation trap industry surveys price at 10–20 hours per week and $15,000–$50,000 per year of avoidable cost.

Booking Percentage Rent Correctly​

The accounting question is when to recognize the income. Percentage rent is a sales-based variable payment: under ASC 842, variable lease payments that depend on the tenant's sales are excluded from the initial measurement and classification of the lease, and recognized as variable lease income in the period the underlying sales occur. You do not estimate it, defer it, or smooth it.

In practice that means:

1. Accrue monthly from POS data. At each month end, compute each vendor's reportable gross sales from the settlement reports, apply the contracted rate, and book the rent as income with a corresponding receivable (if you invoice) or a contra against what you owe them (if you deduct at settlement — more below).
2. True up when reported sales differ. If a vendor's self-reported sales differ from the POS record, book the difference in the month you resolve it, with a note.
3. Never net it against your costs. Percentage rent is income at ~100% margin; CAM recoveries are expense offsets. Blending them makes your landlord economics and your operating costs both unreadable.

A clean account structure keeps this legible: one rent-income subaccount per vendor (income:rent:percentage:taco-stall, income:rent:percentage:ramen-stall, …). When a stall goes dark or a vendor is replaced, you can still see exactly what that stall earned, which is precisely the number you need for the turnover decisions coming later in this guide.

One Checkout, Twelve Merchants: The POS Settlement Problem​

The defining operational feature of a modern food hall is also its defining accounting problem: a guest orders from four kitchens and the bar on one device, pays once, and walks away. That single $41.50 transaction is actually several:

• the taco stall's $14.50 of food sales,
• the ramen stall's $16 of food sales,
• your bar's $11 cocktail,
• and the sales tax on all of it, collected on behalf of three different tax-paying entities (and possibly a local food-and-beverage tax besides).

That means your nightly settlement file is not a report — it's your primary source document, and the clearing account is your best friend. The pattern that works:

1. Post the day's card settlements to a clearing account as money that is real but not yet allocated.
2. Split by merchant. The settlement report allocates every order to the vendor (or bar) that fulfilled it. Each vendor's gross sales, net of their contractual deductions — percentage rent, technology fees, any agreed marketing contributions — becomes the amount you owe them. Sales tax never enters anyone's revenue; it parks in a liability account attributable to the specific vendor.
3. Reconcile three documents: the bank deposit, the POS settlement report, and the per-vendor statements you issued. All three must tie. A residual that doesn't clear is a bug — in a rate, a fee, a tax setting, or a refund timing difference — and in a well-built ledger every residual has a named account to live in, which makes it findable in minutes instead of quarters.
4. Get the sales-tax attribution in writing. When the operator owns the POS and the payment processing, states start asking marketplace-facilitator-style questions: who collected the tax, on whose behalf, and who remits it? Some halls remit each vendor's tax on their behalf and deduct it at settlement; others push liability back to vendors with a clean split of the reports. Either works; what doesn't work is discovering the ambiguity during an audit. And remember your percentage-rent calculation and your tax attribution must agree on the definition of each vendor's sales — one number, used consistently everywhere.

CAM: Cost Recovery That Behaves Like Revenue​

Common-area maintenance — cleaning, security, landscaping, shared utilities, the bathrooms everyone complains about — runs $14–$28 per square foot per year in typical halls. CAM charges recover it from vendors, and the discipline is: CAM is a cost pass-through, not a profit center. Under-pricing it is a documented margin killer; over-collecting it is a vendor-relations killer, because vendors audit CAM reconciliations the way landlords audit percentage-rent sales reports.

The mechanics:

• Allocate by each vendor's pro-rata share, almost always stall square footage as a share of leasable (or occupied — check each lease) area.
• Estimate monthly, reconcile annually. Collect estimated CAM monthly, then true up at year end with a reconciliation statement per vendor showing actual pool costs, their share, what they paid, and the resulting invoice or credit.
• Respect each lease's pool definition. In a hall with ten vendor agreements you may have ten slightly different definitions of recoverable expenses — one lease caps controllable CAM growth at 5% a year, another excludes capital repairs, a third allows a 10–15% administrative fee on the pool. Your CAM pool ledger needs the same per-vendor treatment your rent math has.
• Track the pool gross, and book recoveries as offsets to the expense accounts they reimburse, so your P&L shows what the hall actually costs to operate. When recoveries run short of actuals every year, the gap is visible immediately — that's the signal to re-estimate.

The annual true-up statements are also your best defense in vendor disputes: a vendor who can see the pool, the math, and their share rarely escalates; a vendor who receives an unexplained number on an invoice always does.

Vendor Turnover Is a Bookkeeping Event, Not Just a Bummer​

Every stall that goes dark costs $50,000–$150,000 between lost percentage rent during the downtime, build-out contributions for the replacement, leasing and legal costs, and the marketing push to introduce the new concept. Industry benchmarks treat occupancy as a first-class metric: below 80% is weak, 85–95% is average, 95%+ is strong. New halls typically open at 60–75% occupancy and fill over the first two months — which is why operators are advised to budget six to twelve months of operating reserves before day one.

What that means for the books:

• Book turnover costs when incurred, against the stall. Downtime, write-offs, and replacement build-outs attached to a stall subaccount let you answer the only question that matters at renewal: is this stall, net of everything, producing?
• Watch revenue per vendor as a leading indicator. Below $400,000 per vendor per year is weak; $550,000–$850,000 is average; $1 million+ is strong. A vendor sliding toward the weak band will default on rent before they tell you — percentage-rent income trending down per vendor is your early-warning system.
• Treat the vendor pipeline like inventory. A replacement concept signed and waiting is the asset that keeps occupancy — and your ~100%-margin rent income — from blinking out during a turnover.

The Numbers That Tell You the Hall Is Working​

The operator benchmarks worth taping to the office wall:

Two of these deserve special attention. Revenue per square foot is the industry's favorite cross-check, and food halls justify their build-outs with it: standalone restaurants average $150–$350 per square foot, while successful halls run $300–$700 — that spread is why developers keep building them. And bar share of revenue is the margin lever most operators under-pull, per the stream table above.

Note that none of these metrics can be computed from a blended "hall sales" number. Every one of them requires the per-vendor, per-stream separation this whole guide is arguing for.

A Monthly Close Built for a Food Hall​

Pulling it together, the close that keeps a multi-vendor hall honest:

1. Daily: reconcile card settlements to the clearing account; verify every vendor's day nets to gross sales less contractual deductions.
2. Weekly: review sales-tax liability by vendor against remittance requirements; investigate any residual over $50 in the clearing account same-week.
3. Monthly: accrue percentage rent per vendor from settlement data; issue vendor statements (reportable sales, exclusions applied, rent, fees); post bar COGS and beverage-cost review; update the CAM pool with actuals; book event and sponsorship revenue by event; compute the benchmark table.
4. Quarterly: re-verify POS gross-sales configurations against each lease's definition; review occupancy and revenue-per-vendor trends; flag stalls approaching the weak band.
5. Annually: CAM reconciliation with true-up invoices or credits to every vendor; stall-level profitability review feeding turnover decisions.

Run a Dozen Kitchens From One Plain-Text Ledger​

This is a structure plain-text accounting was born for. A beancount ledger gives every entity in the hall its own place to live — income:rent:percentage:ramen-stall, income:bar:cocktails, income:fees:technology, expenses:cam:cleaning, assets:settlement:clearing, liabilities:sales-tax:taco-stall — so the monthly entry is the explanation, and every vendor statement ties to a settlement file you can re-derive years later. Because the ledger is text, the lease abstracts, POS exports, and CAM worksheets can sit next to it in version control, and a vendor's dispute becomes a diff instead of spreadsheet archaeology. The documentation covers the import pipeline for getting bank and POS data in automatically, and Fava gives you live dashboards over the whole hall — revenue per vendor, bar share, and CAM pool included — without building a single spreadsheet.

Simplify Your Financial Management​

Operating a food hall means holding money that belongs to a dozen other businesses and proving, every month, that each of them got exactly what their agreement promises. Books that mirror that structure — per-vendor rent accounts, a disciplined clearing account, a CAM pool with per-lease rules — are what turn the Friday-night rush into numbers you can act on instead of reconcile in your head. Beancount.io provides plain-text accounting that's transparent, version-controlled, and AI-ready, so every settlement, statement, and true-up stays traceable. Get started for free and run your hall on books as clean as your inspection scores.