The End-of-Day Close: Z-Reports, Cash Drawer Counts, and Three-Way Daily Sales Reconciliation

Your register says the day brought in $1,842 in cash. You count the drawer twice and get $1,797. Forty-five dollars is missing — and you have no idea whether it walked out as a miscounted twenty, an unrecorded refund, or something worse. If that scenario makes your stomach drop, you already understand why the end-of-day close matters: it is the one daily habit that turns "the money should be there" into "the money is there, and here is the paper trail proving it."

This guide walks through a complete end-of-day close for retailers and restaurants: X-reads and Z-reports, blind drawer counts, card batch settlement, and the three-way reconciliation that ties your POS, your cash, and your bank deposits together. Do it every day and discrepancies shrink from mysteries to footnotes.

What the End-of-Day Close Actually Is​

The close is a short reconciliation ritual you perform after the last sale (and, in multi-shift operations, after every shift). It answers three questions:

1. What should we have? The POS expected totals by tender type — cash, card, gift card, check, store credit.
2. What do we actually have? The physical cash count, the card batch totals, and the supporting slips.
3. What hit the bank? The deposit and processor settlement that should match the first two within a day or two.

When all three agree, the day is clean. When they disagree, the variance tells you where to look — today, while the trail is warm, instead of at month-end when it is stone cold.

Why daily beats weekly or monthly​

Small variances compound. A cashier who shorts customers on change by a dollar or two per transaction can generate hundreds in monthly overages that look like "found money" until a customer complains publicly. A void-abuse pattern is invisible in a monthly P&L but obvious in a daily void log. Reconciling daily also keeps your books honest: revenue is recorded when earned, deposits match sales, and your accountant never has to untangle a month of commingled cash.

X-Reads vs. Z-Reports: Know the Difference​

These legacy terms come from old standalone registers, but every modern POS still honors the distinction:

• X-read (mid-day report): Prints or displays accumulated totals without resetting anything. Run it mid-shift to check progress, spot-check a drawer, or preview the day. Nothing is finalized.
• Z-report (end-of-day report): Prints the final totals for the period and resets the running totals to zero (except the grand total). Running it closes the session. In many systems the day's sales cannot be split across dates afterward, so run it once, at the true end of the trading day.

A typical Z-report shows gross sales, net sales, tax collected, discounts, voids, refunds, paid-ins and paid-outs, and totals broken down by tender type and often by cashier or department. Treat it as the day's source of truth — everything else gets checked against it.

Restaurant wrinkle: close the tabs first​

In restaurants and bars, nothing reconciles until every open tab, table, and delivery order is closed to a tender. Many POS systems block the end-of-day routine while orders are open; if yours does not, make "zero open orders" an explicit checklist item. A tab closed to a saved card after midnight can split sales and settlement across two business days, which is the single most common source of "the batch doesn't match the Z" confusion.

The Close, Step by Step​

Adapt this sequence to your operation, but keep the order: finalize sales first, then cards, then cash, then records.

Step 1: Run an X-read and review the day​

Before anything is final, pull the mid-day read. Scan for anomalies while you can still fix them:

• Voids and refunds far above the usual rate
• Discounts concentrated on one employee
• A tender mix that looks wrong (all cash on a normally card-heavy day, or vice versa)
• Open discounts, manager comps, or gift-card redemptions missing notes

Investigate now. After the Z-report runs and the batch settles, corrections become next-day adjusting entries instead of clean fixes.

Step 2: Adjust tips and settle the card batch​

If you take cards — and especially if you take tipped card payments — this step is time-sensitive:

1. Enter all tip adjustments from signed slips before settling. Once the batch settles, tips generally cannot be added; missed tips come out of the house or the server's pocket, and neither option is pleasant.
2. Reconcile the batch totals to the POS card totals. The terminal or gateway batch report should match the X-read's card figure to the penny.
3. Settle the batch (manually, or confirm the auto-batch ran). Note the settlement date: funds typically land one to two business days later, and that lag is what you reconcile against in Step 6.

Keep signed card slips and tip records with the day's paperwork. In a chargeback dispute months later, that slip is your evidence.

Step 3: Run the Z-report​

Close the session and generate the Z-report for each till or terminal. Save or print a copy — many operators staple the Z-tape to the daily envelope as the cover sheet. If you run multiple dayparts (lunch and dinner sessions, for example), generate one Z per session and attach each to the same business day's paperwork deliberately; do not let sales from several days accumulate on one open session.

Step 4: Count the drawer — blind​

This is the heart of the close, and the technique matters:

1. Count blind. The person counting should not see the POS expected total first. Write down the counted cash by denomination, plus checks, gift certificates, and any paid-out receipts residing in the drawer. Blind counts prevent both innocent anchoring ("I must have miscounted, let me nudge it") and deliberate skimming to match the expected figure.
2. Remove the float. Subtract the standard opening float (say, $200 in small bills) — it was not today's revenue. Prepare tomorrow's float in a sealed, labeled envelope so opening requires no trip to the bank or safe.
3. Compare to the Z. Expected cash = Z-report cash sales + opening float − paid-outs + paid-ins. The difference is your over/short for the drawer.
4. Have a second pair of eyes on large variances. Set a threshold (many small operators use $5–$10 for a same-day recount and manager sign-off; larger stores often use $25) above which the drawer is recounted immediately and the variance is logged with a note.

Step 5: Record overs and shorts properly​

How you handle the variance line is where many small businesses go quietly wrong. A few rules that corporate retail learned the hard way:

• Overs are as bad as shorts. A drawer that is over every day often means a cashier is shortchanging customers — a reputation risk, not a bonus. Track both directions per cashier.
• Log every variance, however small. A running over/short log by cashier and date reveals patterns: one employee's repeated shorts, or shorts that always happen on the same weekday, point to training gaps or theft. Without the log, each day looks like an isolated accident.
• Post variances to a cash-over-short account in your books (an expense account for shorts, netted against overs), not by silently adjusting sales. Sales are what the Z says; the variance is a separate fact.
• Do not reflexively dock pay. In many states, deducting register shortages from wages is restricted or outright illegal, and federal law bars deductions that cut into minimum wage or overtime. Coach, retrain, reassign — and check your state's wage-deduction rules before touching a paycheck.

Step 6: Safe-drop, deposit, and match the bank​

Move the day's proceeds out of the drawer the same night:

1. Prepare the deposit — cash (minus the sealed next-day float), checks, and a deposit slip or bag log showing the breakdown.
2. Safe-drop immediately if the bank run happens in the morning. Never leave the day's cash in an unlocked drawer or office.
3. Match deposits to sales the next day. When the bank deposit posts and the card batch lands, confirm each matches the Z-report within the expected settlement lag. This third leg — bank vs. books — is the one that catches deposit errors, missing bag drops, and processor fees booked against the wrong day.

That is the full three-way match: POS expected = physical counted = bank settled. Any leg that disagrees names the problem: POS-vs-count variances are register issues; count-vs-bank variances are deposit or settlement issues.

Retail vs. Restaurant: Where the Close Differs​

The skeleton is the same, but the friction points differ:

Restaurants should reconcile each server's checkout (their declared cash owed vs. their card tips and sales) as part of the close, not as an afterthought. Retailers with gift-card programs should track redemptions against outstanding liability so breakage estimates stay honest.

Common Discrepancies and Their Usual Causes​

When a leg of the triangle disagrees, work through these before assuming theft:

• Drawer short by a round amount ($10, $20, $50): Usually a miscounted bill during a rush or change made from the wrong denomination slot. Recount by denomination.
• Drawer over by small amounts: Often a mis-keyed tender — a cash sale rung as card, or vice versa. Check whether the card batch is off by the mirror amount; offsetting variances ("up and down") net out at the daily total.
• Card batch short vs. POS: Unbatched or declined-then-forced transactions, tips entered after settlement, or a terminal whose batch closed before the last sale. Compare transaction counts, not just dollars.
• Bank deposit short vs. Z: Processor fees netted from the settlement (book gross sales and fees separately), a missing check, or a deposit bag logged but never dropped.
• Tax collected looks wrong: Discounts applied before vs. after tax, tax-exempt sales without certificates on file, or a new menu item or SKU mapped to the wrong tax rate in the POS.

Document the resolution on the daily report even when it is trivial ("$20 short — bill stuck together, found on recount"). The log is the evidence that variances get investigated, which is exactly what an auditor, insurer, or buyer wants to see.

Making the Close Stick: Roles, Tools, and Habits​

A close procedure only works if it survives busy nights and staff turnover:

• One person per till. Accountability requires it. If two cashiers share a drawer, no variance can be assigned, and the careful employee ends up subsidizing the careless one. Do a proper handover count at shift changes.
• Separate the drawer from the safe. The till is for sales and change only — no paying vendors, no petty-cash loans, no IOUs. Everything else goes through the safe with a receipt, so paid-outs reconcile cleanly.
• Write it down. A one-page closing checklist (close orders → X-read → tips → batch → Z → blind count → variance log → safe drop → deposit prep) beats tribal knowledge. Laminate it and keep it at the register.
• Spot-check mid-shift. An occasional unannounced till count keeps everyone honest and catches problems while the shift's transactions are still fresh.
• Let the POS do the math. Modern systems auto-compare counts to expected, flag variances, and export the daily journal. If yours integrates with your accounting software, the daily sales entry can post automatically — but still review the variance log yourself. Automation records the facts; judgment reads them.

Book the Day Right: From Z-Tape to Ledger​

The close produces the numbers your books need. A clean daily sales entry debits cash and card receivables, credits sales and sales-tax payable, and routes the variance to cash-over-short:

• Record gross card sales, then book processor fees as a separate expense when the settlement lands — netting fees against revenue understates both and muddies margins.
• Keep gift-card sales out of revenue until redemption; they are a liability the day they are sold.
• Reconcile sales-tax payable to the Z's tax-collected figure monthly at minimum, so filing season holds no surprises.

Do this daily and your month-end close becomes a review instead of an excavation. Tracking these figures in a transparent, version-controlled ledger means every day's Z-tape ties to an entry you can audit years later.

Keep Your Daily Numbers Audit-Ready​

A disciplined end-of-day close — Z-report pulled, drawer counted blind, batch settled, variances logged, deposit matched — is what separates businesses that hope the money is right from businesses that know. It takes fifteen focused minutes, and it pays for itself the first time it catches a missing deposit or a drawer pattern before it becomes a loss.

As you tighten your daily routine, maintaining clear financial records is what makes the whole system trustworthy. Beancount.io provides plain-text accounting that gives you complete transparency and control over your financial data — no black boxes, no vendor lock-in. Get started for free and see why developers and finance professionals are switching to plain-text accounting.