How to Reissue a Receipt When Your Printer Dies Mid-Service

The sale is done. The payment went through, the journal entry exists, and the only thing that failed is a printer. Take the customer name and an email address or phone number, tell them the receipt is on its way within the hour, and keep the queue moving. Everything below can be done after service.

Reprint from the point of sale first

A reprint from the system is always better than a reconstruction, because the system has the sale exactly as it was rung and you do not. Four paths exist in almost every point of sale, in rough order of how quickly you can reach them.

1. Last-transaction reprint. A button on the sale screen that re-sends the most recent receipt to the printer. It is the fastest path and the narrowest: it usually only holds the last sale, sometimes the last few, and it is gone the moment the next order is rung. If the printer failed mid-service, use this before you ring anything else.
2. Transaction lookup. Search by transaction number, by time window, by amount, or by the last four digits of the card, and reprint from the result. This is the path that survives the rest of the shift, and it is the one worth practising before you need it.
3. Electronic journal. Every point of sale keeps a journal of the raw receipt text as it was sent to the printer. Reprinting from the journal reproduces the sale byte for byte, including the EMV block on a card slip, which no manual reconstruction will. On card transactions this matters: processor documentation treats the application identifier, the authorisation code and the copy legend as required content on a cardholder receipt, and the journal is the only place those values still exist.
4. Back-office report. A transaction detail report from the management application, printed to a normal office printer or exported to PDF. It will not look like the tape, but it is complete and it is authoritative, and it is the right fallback when the terminal itself is the thing that died.

One thing to check before you start: many terminals only allow a reprint by a user with a manager role, and some allow exactly one reprint per transaction and then lock it. Both restrictions exist for the same reason a duplicate has to be marked — an unlimited supply of identical-looking receipts is a refund fraud tool. Find out which restriction applies to your equipment on a quiet afternoon, not during a Saturday rush.

When the point of sale cannot: what a duplicate must say

Sometimes the terminal is the casualty — a dead tablet, a failed upgrade, a card reader that took the sale and lost the session. Then you are issuing a replacement rather than a reprint, and the replacement has to declare itself.

Mark it DUPLICATE or REPRINT, in the header where it cannot be missed. The reason is not bureaucratic. An unmarked copy of a receipt is a second instrument for the same transaction: it can be presented at a returns desk after the original already was, submitted to an employer alongside the original on a different expense line, or filed twice in a bookkeeping system that de-duplicates on document number. The marking is what turns a second piece of paper into a copy of the first.

A duplicate should carry the original date and time, the reprint date and time, and the original transaction number, so anyone reading it can see both the sale and the copy as separate events.

      HARBOUR LANE COFFEE
         (555) 010-4472
--------------------------------
       *** DUPLICATE ***
    REPRINT OF ORIGINAL SALE
  NOT A NEW SALE - DO NOT PAY
--------------------------------
ORIGINAL   08/05/2026 14:22
TXN        0142-03-004417
REPRINTED  08/05/2026 15:07
--------------------------------
2 FLAT WHITE               9.00
  - OAT MILK               1.40
1 ALMOND CROISSANT         4.25
--------------------------------
SUBTOTAL                  14.65
TAX                        0.92
TOTAL                     15.57
--------------------------------
VISA CREDIT               15.57
CARD #           **** **** 1234
AUTH CODE: 05123D
--------------------------------
REISSUE REF  RS-2026-0044
AUTHORISED BY  M. OKONKWO (MGR)
--------------------------------
           THANK YOU

Note what is not on it. There is no expiration date and there are four card digits, not six or eight. That is not a style choice — a cardholder receipt may not show more than the last five digits and may not show the expiry at all, and a hand-built replacement is still a cardholder receipt. If you are unsure what the original showed, print four digits. Fewer is always compliant; more never is.

Never issue a new transaction number

This is the single most expensive mistake in the whole process, and it is the intuitive one, because a new document feels like it should get a new number.

Reuse the original. A new number creates a second transaction in the day. The register total then exceeds the settled cash and card takings by the amount of the sale, and the shift will not reconcile — you will spend the end of the night hunting a discrepancy you manufactured yourself. If the number came from a sequential book or a numbered receipt series, a skipped or doubled number is exactly the pattern an auditor looks for, because it is also what deliberately suppressed sales look like.

The customer-facing risk is just as real. A receipt with a fresh number and today’s timestamp reads as a second purchase. If it reaches a card issuer during a dispute, you are explaining why your own records show two sales and one settlement.

The correct structure is one transaction number, one sale, and any number of copies — each marked, each carrying its own reissue reference if you number those separately. A reissue reference like RS-2026-0044 lives in its own sequence and never touches the sales sequence.

Match the format it is replacing

A replacement that looks nothing like the original creates work for everyone downstream. The customer’s bookkeeper is matching it against a card line; your own day-end is matching it against the tape; an expense reviewer is checking it against a policy that expects certain fields. Every field you drop is a query someone has to raise.

Reproduce the same paper width, the same field order, and the same tax presentation. If the original was an 80 mm thermal slip with a taxable and non-taxable subtotal split, the replacement should have both lines, not a single tax figure. If it was a card slip, it needs the tender block, the authorisation code and the copy legend. If the printed lines do not sum to the printed total, the document is worse than useless — it is the first thing a reviewer notices and the reason the whole submission gets rejected.

This is where a format library earns its keep. Pick the format that matches what your printer normally produces, fill it from the journal or the back-office report, and export it at the same width. The receipt maker is free for exactly this case, with no account and no watermark, because a business that needs a replacement receipt at half past three on a Saturday should not be creating a login first.

Emailing or texting the PDF

No federal rule requires a receipt to be on paper. An emailed or texted PDF is the same document in a different medium, and for most customers it is the more useful one, because it will not fade and it can be forwarded straight into an expense tool.

Two practical points. First, ask before you collect: an email address taken for a receipt is not consent to a marketing list, and treating it as one is how a helpful gesture becomes a complaint. Second, send the whole receipt, not a summary. What the customer actually needs it for determines whether a totals-only document is acceptable, and in every common case it is not:

• Warranty or a return. Needs the item description, the date of purchase and the transaction number. A total proves nothing about which item was bought.
• An expense report. Needs itemised lines and the tax stated separately, because the reviewer is checking policy compliance per line and reclaiming pre-tax cost.
• An insurance claim. Needs the item, the date, the price paid and your business details, and it will be read by someone looking for a reason to ask a follow-up question.
• A landlord, an employer or an agency. Needs the period or service the payment covers, which is a field a bare card receipt does not have at all.

Send a PDF rather than a photo of a screen, and name the file with the transaction number so it is findable in six months by both of you.

Keep a reissue log

A reissue log is four columns and it takes ten seconds to fill. Its value shows up months later, when somebody asks why two documents exist for one sale.

• Date and time of the reissue, separate from the original sale time.
• Original transaction number, which is the key that ties the copy to the sale.
• Reason — printer failure, customer lost the original, expense-report request, warranty claim. Patterns in this column tell you when a printer is dying before it dies.
• Who authorised it, by name. If reprints are a manager function, this is the record that the control was actually applied.

Keep the log with the rest of the supporting records for the year. The IRS expects a business to keep records that support the income and deductions on its return, and to be able to explain what a document is; a reissue log is what turns two receipts with one number from an anomaly into a documented, ordinary event.

Preventing the next one

1. Carry more roll stock than feels sensible. A roll costs less than a minute of a Saturday queue. Keep spares where the person on the till can reach them without leaving the counter, and replace the roll at the coloured warning stripe rather than at the end.
2. Keep a spare printer, configured. An identical unit, already paired and already set to the right print width, sitting in a cupboard. An unconfigured spare is not a spare; it is a project for the middle of service.
3. Have a mobile fallback. A phone that can pull the transaction and email a PDF gets you through a dead printer with no reconstruction at all. Test that the person on shift can actually log in to it.
4. Test the thermal side of a new roll. Before it goes in, scratch the paper hard with a fingernail: friction heat develops a grey line on the coated face, and that face has to point at the head. Thirty seconds here prevents the most common blank-receipt call.
5. Practise the lookup path once a quarter. Everyone can press the reprint button. Far fewer people can find a transaction from two hours ago in the journal under pressure, and that is the path that matters when the reprint button has nothing left in it.

Common questions

How do I reprint a receipt from my POS?

There are four common paths and you should try them in this order: the last-transaction reprint button on the sale screen, a transaction lookup by number or by card, the electronic journal viewer, and a back-office transaction detail report. The first two live on the terminal; the last two are usually in the manager or back-office application and will still work when the sale screen has moved on.

What has to be on a duplicate receipt for a customer?

It must be visibly marked DUPLICATE or REPRINT, carry the original transaction number and the original date and time, state the date it was reprinted, and otherwise reproduce the sale exactly as it was rung. A duplicate that looks identical to the original is a document that can be presented twice, which is the problem the marking exists to solve.

My receipt printer is not working. What do I do right now?

Complete the sale first — the payment and the journal entry do not depend on the printer. Then check the obvious three: the roll is in the right way up, the lid is latched, and the cable or pairing is live. If the sale is already recorded, take the customer name and email or phone, tell them the receipt is coming, and reissue from the journal within the hour.

Can I email a receipt instead of printing one?

Yes. No federal rule requires a receipt to be on paper, and an emailed or texted PDF is the same document in a different medium. The two things that matter are that the customer agreed to receive it that way and that the file is the whole receipt rather than a summary — an expense reviewer, a warranty desk and an insurer all need the itemised lines, not a total.

How do I write a replacement receipt by hand?

Reproduce the original sale field for field, mark it DUPLICATE, and keep the original transaction number. A handwritten replacement needs the business name and contact detail, the original date, the original transaction number, the itemised lines with unit prices, the tax stated separately, the total, the tender used with at most the last four card digits, and the signature and title of whoever wrote it.

Should a reissued receipt have a new transaction number?

No. A new number creates a second transaction in the day, so the register total will exceed the cash and card settlement by the amount of the sale and the day will not reconcile. It also gives the customer a document that looks like a second purchase, which is a chargeback waiting to happen. Reuse the original number and mark the copy.