Switching gift card platforms without breaking balances
I run migrations for merchants moving their gift card programs onto Wrapped, and the part people worry about is almost never the part that goes wrong. They worry about the new platform. What actually determines whether a switch goes smoothly is what happens to the cards that already exist.
Every outstanding gift card is a liability. Someone paid you real money and holds a promise you have not yet honoured. That promise does not care which vendor you use, and in most jurisdictions it survives the contract you are cancelling. So the bar for a migration is not "the new system works" — it is "no customer notices anything happened."
What actually has to move?
More than most people expect, and the omissions are the expensive part. Here is what I ask for on every migration and why each field matters.
| Data | Why it matters | What happens without it |
|---|---|---|
| Card code | The physical or digital card in the customer's hand carries this number | Every existing card is dead. Non-negotiable. |
| Remaining balance | What the customer can actually spend | You either overpay or short-change people, and you will find out at the till |
| Expiry date | Consumer law sets minimum validity in most markets | You risk expiring cards early, which is a compliance problem, not an admin one |
| Original issue date | Drives liability aging and escheat obligations | Finance loses the aging view; unclaimed property tracking breaks |
| Transaction history | Disputes, fraud investigation, audit trail | You cannot answer "where did the rest of my balance go?" |
| Recipient details | Resending a lost card, and marketing consent | You cannot help a customer who has lost their email |
Codes, balances and expiry are the minimum viable set. Issue date and history are what separate a migration your finance team is comfortable with from one they quietly distrust for a year.
What should you ask your current provider before you commit?
Ask before you give notice, not after. Once you have cancelled, your leverage is gone and you are working to their timetable.
- Can I export the full ledger, and in what format? "You can see it in the dashboard" is not an export. Ask for CSV with the fields above.
- Is there a window after cancellation? Some providers cut access immediately. Wrapped gives 30 days, for context on what reasonable looks like.
- What is the notice period? This sets your whole timeline. Find it in the contract rather than asking support.
- What happens to outstanding cards on your side? You want them deactivated on the old platform at cutover, so nothing can be redeemed twice.
- Are there physical cards in circulation? If codes were printed, they cannot be reissued. The new platform has to accept the existing number format.
If you are still choosing where to move to, the alternatives shortlists and the head-to-head comparisons are the place to start — both say plainly when staying put is the right answer.
How do you avoid a gap where cards do not work?
Sequence it so the two systems overlap rather than hand over. The order that works:
- Import into the new platform first, while the old one is still live. Nothing is switched on; you are just loading data.
- Reconcile before you cut over. Total issued, total redeemed and outstanding balance should match to the cent. If they do not, stop and find out why — this is the single most important checkpoint in the whole process.
- Spot-check real cards. Take a dozen at different balances, including a zero-balance and a partially-redeemed one, and check each individually.
- Cut over at a quiet trading hour. Not Saturday afternoon.
- Deactivate on the old platform immediately after. This closes the double-redemption window.
- Watch the first week closely. Most issues surface as a single confused staff member, not an outage.
The reconciliation step is where I spend most of my time, and it is worth the patience. A discrepancy found before cutover is a spreadsheet problem. The same discrepancy found after is a customer standing at your counter.
What do you tell customers?
Ideally, nothing — and that is the point. A migration done properly is invisible: the same card, the same code, the same balance. Emailing everyone to announce a platform change mostly generates questions and reminds people they are holding a card they had forgotten about, which is not the moment you want to pick.
The exception is if anything visible genuinely changes — a new balance-check URL, a different storefront address, a card design refresh. Then tell them plainly and briefly, and make sure your support team has the answer before the email goes out rather than after.
What does staff training need to cover?
Less than you think, and it is worth being specific rather than thorough. Front-line staff need three things: how to redeem, how to check a balance, and what to do when a card does not scan. That last one matters most, because it is the only one they will hit under pressure with a queue behind them.
The failure I see is a team that was trained on the new platform but never told what happens to old cards. Someone presents a card issued eighteen months ago, the staff member is not confident it will work, and they hedge in front of the customer. Tell the team explicitly that every existing card has been carried across and will work exactly as before.
How long does it take?
For a single site with a clean export, a working day. For a multi-site group with physical cards in circulation and a finance team that wants the history reconciled, plan for two to three weeks end to end — most of which is waiting on the export from the incumbent rather than anything on the new platform.
Self-managed migration import is included on every Wrapped plan, and premium onboarding at $300 one-time adds managed migration if you would rather someone else did the reconciliation. Both are on the pricing page. Either way, the integration side is usually the quick part — see the list of POS and eCommerce integrations for what connects natively.
One last thing. Do not schedule a migration in your peak gifting season. I have done December cutovers and they work, but there is no upside to the timing and every problem is louder when volume is high.
