Picture a home textiles seller shipping about a thousand parcels a month. On the first day of the 11.11 campaign, 260 parcels are packed and ready at 7:30 a.m. The courier is due at eight and doesn't come. At nine, a message from the carrier's customer service says pickup times will be "updated due to regional congestion." At eleven, there is still no pickup.
The seller has a second carrier agreement, signed last year so they would not depend on a single carrier. But not one shipment has ever gone through that account. The label template has never been tested, nobody knows how cash-on-delivery collection is set up with that carrier, and the warehouse printer is configured for a different label size. The first labels come out at two in the afternoon, while the campaign's orders have not slowed down for a minute.
Nobody made a mistake in this story. The carrier failed, and the seller looked prepared. The problem was this: a backup that has never been used isn't a backup; it's a contract.
This guide covers the three different ways an outage shows up, the pieces a backup needs in order to actually work, and how to test them before the November campaigns start.
The multi-carrier strategy guide explains why you should work with more than one carrier, and what a shipping orchestrator is explains how the choice between them gets made. This post covers the question that sits between the two: if you have a second carrier, does it actually work on the morning you need it?
An Outage Shows Up in Three Different Ways
"The carrier is down" is not one event. How it arrives changes what you need to do.
| Outage | What it looks like | First move | What you need from the backup |
|---|---|---|---|
| Labels can't be created | The integration errors out, labels don't arrive or arrive very late | Label new orders with the backup carrier | An open account and a label flow that has been tried |
| Pickup doesn't happen | Labeled parcels sit in the warehouse | Re-label parcels with the backup and request a backup pickup | Same-day pickup capability |
| Delivery stalls in one region | Parcels leave but wait in a specific region | Move only new orders to that region onto the backup | Known regional coverage and performance data |
The second row is the hardest, because a label bought from one carrier cannot be transferred to another. You have to cancel the old label and buy a new one. Cancellation and refund terms vary by carrier; learn them before outage day, because that is not a day for finding things out.
The third row is the sneakiest. Labels print, parcels move, no alarm goes off. You usually find out when a customer writes "where is my order?". To see it earlier you need to monitor delivery time and on-time rate per carrier; the method is in the carrier performance guide.
That carriers don't perform the same at the same time is not an assumption either. In its own analysis of next-day delivery for two major Austrian carriers, Seven Senders reports that during the observed period one stayed consistently in the green while the other saw delays. The same post says they rerouted some customers' parcels to carriers that were still working during the first wave of the pandemic. We know of no public comparison at that level of detail for Türkiye, but the mechanism is the same: a backup works because carriers rarely fail on the same day in the same way.
A Backup on Paper Is Not a Backup
When a seller has a backup account but has never used it, this is what usually happens on outage day:
- The account isn't active or is incomplete. An ID, contract or customer code is missing, and the first shipment returns an error.
- Cash-on-delivery setup is unknown. Whether collection is cash or card and when the money reaches you can differ at the backup carrier. Cash-on-delivery management already demands tight follow-up; outage day should not be the first time you learn the rule.
- The weight rules differ. Two carriers may not calculate dimensional weight or bracket it the same way. The same parcel can land in a different price bracket at the backup carrier; the logic in the dimensional weight guide applies here too.
- Prohibited and restricted items differ. A product one carrier takes may be refused by the other. Knowing which carrier takes which products in advance keeps you from turning back half your parcels on outage day.
- Label and pickup routines haven't settled. Label size, printer settings, pickup time and pickup point have never been tried once.
- No bargaining position. When you go to the backup carrier on outage day, you are a first-time customer; urgency is the only leverage you have on price and capacity.
Nearly all of this is solved by running real shipments through the account once. None of it is something you want to solve on outage day.
The Four Parts of a Backup Plan
1. A warmed-up second account
Send a small, steady flow of shipments through the backup account. The goal is not savings, it is keeping the account alive: valid credentials, a label flow that has been tried, and cash-on-delivery and returns that have processed at least once on a real parcel. To start, giving the backup carrier one region or one product group is enough; even a few dozen shipments a month keeps the account live. It also gives you a real alternative, backed by real shipments, in contract renewal talks.
2. A matching table
One page: for each region, product type and payment type, who is the primary carrier and who is the backup? Which backup for cash on delivery, which for bulky items, which for remote addresses? Writing this once removes the "who do we give it to now?" debate on outage day, because you won't have time to debate it then.
3. A written trigger
The answer to "when do we switch?" should not be a decision made on outage day. Set a threshold from your own history and write it down: for example, the number of parcels left uncollected the same day, or how far a region's average delivery time rises above normal. Once the threshold is clear, switching is a procedure rather than a panic. A tool that shows delivery time, on-time rate, cancellations and returns per carrier on one screen makes the threshold easier to set.
4. A communication plan
When the carrier changes, the tracking number and link change. You need a ready template and an automatic channel to tell the customer; WhatsApp shipping notifications are the fastest way. On marketplaces the handling deadline keeps running through an outage; to avoid missing it, prioritize marketplace orders, as covered in the Trendyol order management guide.
This Won't Happen By Itself
First, an honest limit: Shipink does not detect that a carrier is failing and move your orders to another one on its own. Nobody can know that on your behalf; where "slowing down" turns into "down" is a line your threshold has to draw.
What Shipink does is cut the switch from hours to minutes and keep the backup ready:
- Several sources in one account. You can use your own carrier accounts, Shipink-agreement carriers and integrated providers in the same account and see quotes side by side; see the carriers page for coverage. If one carrier's live quote request fails, the other carriers' quotes are unaffected; the quote list comes back with the remaining ones.
- Automation rules. Rules apply to new orders, and you can build conditions on fields such as province, district, postal code, payment method, sales channel, product SKU, weight and order total. For matching orders the rule creates the shipment automatically, and for the carrier you can name a specific carrier and account, or use presets like "cheapest," "fastest" or "AI selection." Rules are ordered, and a rule can stop the ones below it from running. So the outage-day job is: add a rule on top that points to the backup carrier, or deactivate the old one. Automation rules are available on the Pro and Enterprise plans.
- Bulk shipment creation. Because rules only act on new orders, orders already waiting in the warehouse go through bulk shipment creation. At the end you see how many of the selected orders were created, and labels that failed are flagged separately when you print.
- Per-carrier performance. The reports screen shows delivery time, on-time rate, and cancellation and return counts per carrier, for both outgoing and incoming shipments. That is how you spot the third kind of outage, a silent regional slowdown.
One catch: if a rule uses a Shipink-agreement carrier, it charges the label fee to a saved card, so add the card beforehand and try the rule with a real order. The first time a rule runs should not be on outage day.
Why Now?
Three reasons converge.
Warming up a backup takes time. Getting the account live, testing cash-on-delivery and returns, and settling the label template does not finish in a day. The time left before the November campaigns is enough to do this calmly; trying to do it in the middle of a campaign is not.
An outage costs the most on campaign days. The same delay creates a support ticket on a quiet Tuesday and hundreds of orders waiting at once during campaign week. We have already written about how failed delivery rates climb in peak periods and a peak season preparation plan; a backup plan is the item on that list most often skipped.
New-year rate cards land in the same window. Carriers' new rates tend to arrive in the last quarter. With a backup carried by real shipment data, you don't walk into that conversation from a weak position.
A Test You Can Run This Week
You can test the plan within a week.
- Verify the backup account. Is the second carrier's account active, with cash-on-delivery and returns set up?
- Ship three different parcels. One cash-on-delivery, one to a remote region, one bulky. Print the label, check the pickup routine and watch the delivery result.
- Write the matching table. Primary and backup carrier by region, product type and payment type.
- Write your trigger. From your history: at what level of delay does the backup kick in?
- Rehearse outage day. Add a rule and switch it off; try bulk shipment creation once for ready orders; prepare the customer notification template.
- Learn the label cancellation terms. How do cancellation and refunds work for a label already bought from your primary carrier?
You can finish these six in one afternoon. When you do, you hold a backup you have seen work, not a contract.
See the Backup Work Once
Outages are usually temporary and forgotten when they end. But when one lands on a campaign day, the gap between a seller caught unprepared and one who is ready opens in those few hours. A backup plan is not used often, which is exactly why you can't know it works until it has been tested.
Before November, run the backup once with real shipments. Being ready costs an afternoon; being unready costs a campaign day. To set up carriers and rules on one screen, you can try Shipink for free, look at automation rules, or write to us to talk through your setup.
