Friday morning, forty-one orders to get out. Thirty-eight print a label. Three come back with the same message: could not create label.
All three are going abroad. The addresses are fine. What they are missing is a phone number, which nobody at your store failed to collect, because the field at checkout says optional and every one of those customers read it correctly.
So those three orders sit in the queue looking exactly like the thirty-eight that shipped, because Shopify has no status that means this one cannot be labeled. Someone finds them on Monday, or the customer finds them on Wednesday.
The short answer
Shopify's checkout phone field has three states, and none of them fit a store that ships internationally. Do not include, optional, or required, applied to every country and market you sell to. There is no per country rule without a checkout app on Shopify Plus.
So you either tax every domestic checkout with a field it does not need, or you accept that a share of your international orders will arrive without the one thing their carrier insists on. Most stores pick the second, correctly, and then build nothing to cover it.
The fix is a check that runs between the order landing and the label printing: does this order carry what its destination actually needs, and if not, can the number be recovered from somewhere else on the order before anyone emails the customer. It usually can.
The setting is all or nothing, which is why nobody turns it on
Shopify's checkout form options give the shipping address phone field three states. Do not include. Optional, which is the default on new stores. Required, which the documentation describes as the setting for when your payment provider or shipping carrier needs a phone number.
That is the whole control surface, and the same page says plainly that these customer information settings apply to every country and market you sell to. To vary a checkout field by country you need an app that customizes checkout, and those are a Plus capability.
Merchants have been asking for the per country version for years. In one thread a reply laid out steps for a country setting inside form options, and three separate merchants came back to say they could not find it in their own admin, the last of those in October 2024. A reply in June 2026 concluded that a paid app is the only route. A second thread asks for the same thing in the title.
One workaround in those threads is honest about what it is: leave the field optional and rewrite its label in your theme content to explain that international orders need a number. That asks politely. It enforces nothing.
Two ways this fails, and the quiet one costs more
The visible failure is the one from Friday morning. Error
030009 is a carrier validation error, usually returned by
FedEx, and
the
most common trigger is a missing or malformed recipient phone.
Malformed matters as much as missing. The number needs its country code,
and the digit count has to fit the destination, so a ten digit US number
on a parcel to Germany fails exactly like a blank field.
Its neighbors are worth knowing while you are in there. A blank first or last name gets rejected outright. An address line over the carrier's limit fails, which is 35 characters for DHL and UPS and 32 for Evri. And some carriers want a valid phone on your own origin address under Settings and then Locations, where one blank field fails every label you try to buy rather than one.
At least that failure announces itself. The quiet one does not. The label buys, the label prints, and the number simply is not on it.
A merchant shipping about twenty international orders a month described the result: parcels reach the destination country and stop there, held at the local post office, with nobody able to reach the recipient to arrange the handover. They compared their Shopify label against one from another provider that carried the number, reshipped orders at their own cost, counted the loss in thousands, and said support dismissed it twice. That is one store's account of its own carrier mix rather than a universal rule. The shape of it is the part to take seriously: you pay for the goods, you pay for two labels, and the customer tells people you lost their order.
The pickup point hole that no setting will close
This is the version that should bother you most, because it is a Shopify integration failing against a Shopify setting.
Merchants shipping to pickup points with Mondial Relay report that the pickup flow skips the standard shipping address form, which is the only form the required phone setting governs. So the field stays optional no matter what the admin says. Shopify Shipping then refuses the label, with the message that a recipient phone number is required for pickup point labels.
And pickup point orders cannot be edited, so even when the customer emails the number back, there is nowhere to put it.
In a parallel thread a merchant puts it in one line: for pickup point delivery it has become optional, so most customers simply do not enter it. Their cart attribute workaround failed because customers could delete the prefilled number. Support spent more than an hour on it and attributed the problem to the carrier, which the merchant disputed, since the integration is Shopify's own.
The workarounds cost real money. One merchant priced manual labels on the carrier portal at about 5.99 euro against about 3.30 euro on Shopify's negotiated rate, and another migrated their carrier account by hand to get to 4.51. As of the end of August 2026 the thread records only that the report was forwarded to Shopify's product team. No fix, no timeline.
It is worth understanding why the carrier will not bend on this. A pickup point network needs the number to tell the buyer the parcel has arrived, not merely to attempt a delivery, which is why one pickup app's documentation simply tells merchants to set the field to required and accept the cost elsewhere.
The number is often already on the order
Before any of the expensive remedies, the cheap one. A Shopify order can carry a phone number in three places: on the shipping address, on the billing address, and on the customer record, because the checkout contact method can be a phone number instead of an email.
The label reads the shipping address. Only the shipping address. Which means an order can show a phone number in the admin, to a human looking at the order page, and still fail at the label.
This is common enough that order tagging apps document it as a standard recipe: tag the order with the shipping phone, and fall back to the billing phone when the shipping one is empty. Read that as a finding rather than a feature. The number you are about to email a customer for is frequently sitting one field away.
What this costs on an ordinary month
Take a store doing fourteen hundred orders a month with fifteen percent going abroad. That is two hundred and ten international parcels. With the field optional, say half of them arrive with no phone on the shipping address, so a hundred and five.
Most of those deliver without incident, because most routes never need the number. Say five percent end badly: five parcels. Each one costs the goods, the outbound label, the replacement label, and twenty minutes of somebody's afternoon. On a forty-six dollar item and an eighteen dollar label that is around four hundred dollars a month in hard cost, before the refund you give to keep the review clean.
Four hundred dollars is not the interesting part. The interesting part is that none of it lands in any report. There is no line called orders we could not label. The only trace is three rows in a fulfillment queue that look identical to the rows that shipped.
What Shopify Flow gets you, and where it stops
Flow can do the detection, and you should go and build this today. Its conditions read nested fields, with the shipping address given as the documented example, and it has an empty or does not exist operator that returns true for a blank or null value. An order created trigger, that condition, and an add tag action is a ten minute build that puts a visible marker on every order missing a number.
Then it stops. Flow cannot ask the customer for the number, because its one email action only reaches customers subscribed to marketing, which is its own long story. It cannot write a number back onto the shipping address when one arrives. It cannot tell whether this particular destination and shipping method even need one, so it flags the Denver order next to the pickup point order in Lyon. And it can do precisely nothing about an order Shopify will not let anyone edit.
Flow gives you the flag. Everything after the flag is still a person with a browser.
What the automation actually has to do
As a Dugong playbook, in plain prose:
# trigger
Order created or updated,
plus a sweep of the fulfillment queue
# steps
1. Check every order for the
fields its label needs
2. Look in the other two phone
fields first
3. Ask the customer once, where
they read
4. Write the number onto the
shipping address
5. Hold the order, not the whole
queue
6. Require the number where it
earns its keep
7. Report the misses every week
Step one is a rule per destination and shipping method, not a blanket check for a blank field. A domestic ground parcel usually does not care. An express international parcel, a pickup point delivery and a handful of countries care very much. Flagging everything trains your team to ignore the flag.
Step two is the fallback, and it resolves more of these than anything else in the list. Shipping address, then billing address, then the customer record, then any earlier order from the same customer. Only a genuine blank across all four is worth a message.
Step three is one message, sent on the channel the customer actually reads, saying what you need and why, with a link that takes them one tap to reply. Not a support ticket. Not an email that sounds like a newsletter. The reply rate on this is the difference between the playbook working and the playbook being a report nobody acts on.
Step four is the part teams skip. The number has to land on the shipping address, because that is where the label looks. A phone number recorded in a help desk thread, or typed into the order notes, has not solved anything.
Step five keeps the warehouse moving. Hold the affected order and let the other thirty-eight print. The common failure mode is a batch that errors out in the middle and a picker who starts the whole run again.
Step six is where the checkout setting finally earns a decision. If your weekly report shows the misses concentrated in one market or one shipping method, scope a checkout rule to that case, or make the field required and accept the conversion cost knowingly. For pickup point orders, upstream is the only place this is solvable at all.
Step seven is the number that keeps it from drifting back. Misses per week, split by country and shipping method, with the recovery rate next to it. If that number is rising, something changed in your mix, and you will see it in a week rather than in a quarter of chargebacks.
Do this before you close the laptop
Three things, about fifteen minutes, no tooling required.
Open Settings, then Checkout, and read what your shipping address phone field is actually set to. A surprising number of merchants have never looked, and are running whatever the store shipped with.
Then open Settings, then Locations, and check that every location you ship from has a phone number on it. That one takes thirty seconds and fails every label when it is wrong.
Last, filter last month's orders to the ones going abroad and count how many have no phone on the shipping address. That count is the size of your exposure, and until you have it this whole problem stays theoretical.
If you ship internationally from Shopify and have a version of this
story, the parcel held in a sorting office for a month, the pickup point
order you could not label, the number that was on the billing address
the whole time, the inbox is open:
field-notes@dugong.live. We are collecting case studies for
the next issue.