How to balance a cash register: the end-of-day checklist that catches every peso, dollar, and cent
Every till that's ever been counted has come up short at some point. The difference between a well-run counter and a leaky one isn't zero variance — it's a closing routine that makes variance visible, explainable, and boring. Here's the 10-minute procedure.
Wemu Team
Product
If you've ever stood at the counter at 9pm staring at a drawer that's $14.50 short with no idea why, this guide is for you. Balancing a register isn't accounting — it's a physical routine. Done the same way every night, it takes ten minutes and turns "the till is off again" from a recurring mystery into a rare, explainable event.
Why drawers go out of balance
- Wrong change given during a rush — by far the most common cause, and it goes both ways (overs are as suspicious as shorts)
- Sales rung at the wrong amount, then "fixed" with the next customer instead of a correction
- Refunds or voids handed out in cash but never logged in the POS
- Discounts applied by hand at the drawer instead of through the system
- Cash drops to the safe mid-shift that nobody wrote down
- And yes — occasionally, theft. But assume process failure first; it's the cause nine times out of ten
The 10-minute closing procedure
- 1Close the register in your POS first — this locks the day's transactions and produces the expected-cash figure (your Z-report or end-of-day summary)
- 2Count the drawer twice, largest notes to smallest coins, and write down the total before you look at the expected figure — counting toward a known target invites bias
- 3Subtract your float (the fixed starting amount, e.g. $200) to get actual cash takings
- 4Compare actual takings against the POS expected-cash figure and record the variance — even when it's zero
- 5Investigate anything over your threshold while the shift is fresh: re-count once, then check refunds, voids, no-sales, and cash drops in the POS log
- 6Have the closer and one other person initial the count, bag the takings, and reset tomorrow's float to the exact same amount
Count blind, always
The single highest-leverage habit in this whole guide: count the drawer before looking at what the POS says should be there. A closer who knows the target will unconsciously "find" it.
Set a variance policy so nobody has to guess
| Variance | What to do |
|---|---|
| Under 1% of cash takings (or ~$5) | Log it and move on — chasing coins costs more than they're worth |
| 1–2% or $5–$20 | Re-count, check voids/refunds/no-sales for the shift, note the likely cause |
| Over 2% or $20, or any variance three shifts in a row | Owner reviews the shift log and transaction history the same night |
| A pattern tied to one person or one shift | Compare per-staff drawer history over the month — patterns end arguments |
The point of a written policy isn't punishment — it's the opposite. A closer who's $3 short shouldn't be sweating, and a drawer that's $40 over shouldn't be celebrated. Both just get logged, explained, and signed.
What your end-of-day report should tell you
- Sales by payment method — cash expected in the drawer vs card settlements coming from the processor
- Refunds, voids, and discounts for the shift, each tied to the staff member who rang them
- No-sale drawer opens (the drawer opened without a transaction — worth watching)
- Cash drops taken to the safe during the shift
- Who was clocked in on the register during the day
If your current setup can't tell you those five things, balancing will always feel like detective work. A modern POS produces this automatically: in Wemu, closing a shift generates the summary — sales by tender, refunds and voids by staff member, drawer events — and the daily report lands in your inbox, so the 9pm mystery becomes a 30-second read.
Make closing time boring
Wemu POS tracks every drawer event, ties refunds and voids to the staff member who made them, and emails you the end-of-day summary. 7-day free trial.
Start free trial →