Restaurant cash close process — end-of-day reconciliation playbook (India)

About this piece. Most Indian single-outlet restaurants close cash by feel. Cashier counts the drawer, manager glances at the POS total, somebody mutters "thoda kam hai" and everyone goes home. This is the missing playbook — the 7-step nightly close that turns the cash drawer into a number you can defend on Tuesday morning when the variance question lands.

What "cash close" actually means

Cash close is not the same as cashing up. Cashing up is counting the drawer. Cash close is the process that ends with one number — net cash for deposit — that ties to the POS, the petty-cash float, the UPI/card settlement, and the day's discounts and refunds, with every variance explained on paper.

If your nightly process produces a number but no explanation when the number is off, you don't have a close — you have a count.

The operators in NCR I've spoken to who run a clean close usually take 8–12 minutes per night. The ones who don't end up spending 40 minutes on Saturday morning unwinding the week. The math on time-spent always favours the disciplined nightly close.

The 7-step end-of-day cash close

Run these in order. Skipping a step is fine — skipping the order is not.

Total: 11 minutes if the float is fixed and the PCVs were filled at the time of payment, not at 11:30pm.

The reconciliation formula

The whole close sits on one identity:

Expected drawer cash = Opening float
                     + Cash sales (POS Z-report)
                     − Cash refunds
                     − Cash paid out via PCVs

If actual drawer cash equals expected, variance is zero. If not, variance = actual − expected. Threshold for "investigate now": ±₹100 for outlets doing under ₹40,000/day, ±0.25% of cash sales for larger outlets. Anything inside threshold gets logged. Anything outside gets a one-line reason on the close sheet before anyone leaves.

A typical 30-cover dhaba in a tier-2 city will run three to five "₹20 short" days a month and one or two "₹500 over" days a year. Both patterns deserve a note. The ₹20 shorts compound to ₹600/month if ignored. The ₹500 over usually means a refund was given without a POS entry — which is worse than the short, because it's an unrecorded sale.

What the close sheet looks like (one half-sheet)

This is the form that goes in the manager's clipboard. Print it on A5, two-up on A4, fifty per pad.

┌────────────────────────────────────────────────────┐
│ DAILY CASH CLOSE — [outlet name]                   │
│ Date: __/__/____   Shift: ____   Cashier: _______  │
│                                                    │
│ A. Opening float                       ₹ ________  │
│ B. POS Z-report — cash sales           ₹ ________  │
│ C. POS Z-report — cash refunds         ₹ ________  │
│ D. Petty cash vouchers (sum + count)   ₹ ____ /__  │
│                                                    │
│ Expected drawer cash = A + B − C − D   ₹ ________  │
│ Actual drawer cash counted             ₹ ________  │
│ Variance (actual − expected)           ₹ ________  │
│                                                    │
│ Reason for variance:                               │
│ ___________________________________________        │
│                                                    │
│ Deposit envelope (actual − float)      ₹ ________  │
│ Float retained for next shift          ₹ ________  │
│                                                    │
│ Cashier sign: _______   Manager sign: _______      │
└────────────────────────────────────────────────────┘

That sheet is the audit artefact. Store it in a clipboard for 90 days, then bundle by month. If GST or Income Tax ever comes knocking, this is the trail that ties the bank deposit to the POS to the books.

The four numbers that must tie

By the time the manager signs the close sheet, four numbers from four different systems should reconcile:

1. POS Z-report cash sales (what the till says you took)
2. Drawer count − float + vouchers (what the cashier physically holds)
3. Deposit envelope (what hits the bank tomorrow morning)
4. DSR cash row (what gets entered in the day book)

If any two disagree by more than the variance threshold, the close is not done. Most operators stop at step 1 and 2 matching and assume the rest will follow. They don't, and that's where the month-end reconciliation pain begins.

Six common close mistakes (and the fix)

1. Counting before printing the Z-report. The cashier counts ₹14,300, then sees the Z-report says cash sales were ₹15,100. The number gets adjusted to fit the report. Order matters: Z-report first, count second, never the other way round.
2. Treating UPI failures as cash. Customer paid by UPI, transaction failed silently, customer left assuming success. That's a missing ₹420 the cashier will eat unless reconciliation catches it. Always cross-check UPI count against the payments app dashboard, not memory.
3. Skipping the float reset. The cashier leaves ₹3,260 in the drawer "because the morning shift will need change". Now tomorrow's open isn't a fixed float and reconciliation has no reference. Always pull the deposit envelope first; refill the float to its label number.
4. No PCV at all for sub-₹50 spends. "Bas ₹30 ka dhaniya tha." Multiplied across 60 spends a month, that's ₹1,800 of unrecorded outflow that breaks reconciliation. Every outflow gets a voucher. No exceptions.
5. Refunds entered as discounts. A refund reduces both cash and sales. A discount only reduces sales. Mixing them up makes the GSTR-3B filing wrong. Train the cashier to use the refund button, not the discount button, when cash leaves the drawer for a customer.
6. Single signature. Manager signs the close sheet; cashier doesn't. Or vice versa. Two-signature rule is the only thing that survives a payroll dispute three weeks later. Both sign. Always.

Where this fits in the daily-ops loop

The cash close is the load-bearing ritual of the day. Three pieces feed into it and one piece flows out:

• Imprest cash float — the fixed reference number (read the spoke)
• Petty cash vouchers — the audit trail for each outflow (read the format)
• POS Z-report discipline — the till's version of the day
• Bank deposit slip — the morning hand-off to the bank

When the close runs cleanly, the deposit slip is a 90-second job. When it doesn't, the deposit becomes a ₹1,000-too-much or ₹1,000-too-little event that the owner discovers from the bank statement next week.

What to do this week

If you don't currently run a 7-step close, pick three nights this week and run it manually. Print the close sheet (or copy the format above into a notebook). Keep the timing — Z-report first, count second, reconciliation third. By night three, the variance pattern will start to surface, and that pattern is what the close exists to make visible.

The discipline is the product. The form is the scaffold. Everything else — including Restaurant Daily — is just an automation of a loop you can run on paper tonight.

Related on Restaurant Daily

• Daily cash reconciliation in Indian restaurants — POS, drawer, deposit
• What causes cash variance in restaurants — 11 root causes and the fix for each
• Restaurant till close checklist — printable 14-step end-of-day form