The ticket asked for a settings field for each store, where someone would type in the public holidays. We built a calendar that already knows when Easter falls, and kept the field for the days only the business can decide.
Shopify estimated delivery dates live at checkout. But “when will it arrive?” starts on the product page. A date there is a promise, so it should come from the warehouse’s stock and calendar, not from a flat “3–5 days”.
At dtails we built this for a retailer with physical stores and a separate Shopify store per country. The product page reads the stock state from the ERP, picks the matching promise, and counts working days like the warehouse’s wall calendar: weekends crossed out, holidays marked, the stocktake circled.
This is for you if you sell long-lead items next to stock lines, or run more than one country store. If everything ships from one shelf within days, Shopify’s checkout dates probably cover you.
Shopify’s own delivery dates start at checkout
Shopify’s native dates are for items in stock, and they show at checkout. Here is where they stood when we checked the Help Center on 17 November 2026:
- Manual delivery dates work on every plan, once a product’s inventory is above zero. With flat rates, only stores shipping from Canada, the US, the UK, France, Italy, Spain, Germany or Australia get them.
- Automated delivery dates predict the date from your fulfilment history, on Basic or higher, in the US, Canada, New Zealand and 32 European countries.
- Both show up at checkout, in confirmation emails and on the order pages. Neither shows on the product page.
- Shop Promise puts a badge on the product page that can show an arrival date. The badge needs a US location, Shop Pay and tracking on at least 90% of orders; the date also needs automated dates turned on.
That leaves out the two cases a retailer with long lead times has to answer: an item on its way to the warehouse, and an item made to order.
Each stock state makes a different promise
The delivery line comes from a small decision table over the ERP’s stock states. Each state gets its own sentence, and only an empty warehouse needs a computed date.
Warehouse stock repeats a promise each store already makes: one number of working days for parcel delivery, another for everything else. Dates are for what a shopper can’t guess: a booked arrival, or a made-to-order lead time.
Shop-floor stock doesn’t ship from the warehouse, so it counts like an empty warehouse. The same function decides whether the item can be bought online at all, so the delivery line and the buy button can’t disagree. Try each stock state in the demo at the top.
Liquid ships the data; the browser reads the clock
Liquid knows the product and the store. Only the browser knows what day it is and what each shop holds right now:
- Today. Liquid’s
'now'reflects “the time that the Liquid was last rendered”, and Shopify caches pages, so a server-side date can carry yesterday’s date. Same trap as the member prices I wrote about earlier. - Stock per location. Liquid’s
store_availabilitysays whether a location has the item. The Storefront API’sStoreAvailabilitysays how many. - Fields the API can’t see. One ERP metafield in our build has no Storefront API access, so Liquid prints it into the page instead.
A small custom element takes today’s date in the store’s time zone, fetches live stock once, runs the table and fills in the date. Pick another variant and it runs again.
Compute the holidays; keep a list only for the days you choose
The project manager’s ticket asked for settings to type in each store’s public holidays. We built the calendar to compute them instead: the fixed days, plus Good Friday, Easter Monday, Ascension Day and the rest, counted from Easter with a well-known algorithm. The setting became “extra closure days”, for the days the warehouse shuts by choice, like Christmas Eve or a stocktake.
The delivery note said why: Easter moves every year, and a half-filled list gives wrong dates without anyone noticing. We listed the year’s computed dates for the client to check and offered to switch back to a plain list. The PM confirmed they looked right in testing and would ask the client for their closure days.
The calendar holds this year and the next two, because an ERP arrival date has no upper limit. Without next year’s holidays, a late-December arrival would quietly count a holiday as a working day.
The calendar belongs to the store that ships
Whose holidays count depends on where the parcel leaves from. Our two countries run as separate Shopify stores on one theme, each shipping from its own warehouse. A shopper from the second country, buying on the first store, still waits for the first country’s holidays. So the store picks the calendar, never the visitor’s country or language.
One catch: theme settings are copied between our stores, so a per-store value in one setting would be overwritten. Those settings get a column per store, picked by the store’s currency, which the copy can’t change.
The same thinking caught a wording change. The second store promises “from N working days” instead of a range. But translations belong to a language, and the first store also offers the second store’s language, so editing that sentence would have changed the first store’s promise too. The second store got a sentence of its own. A language isn’t a store.
Decide what blank means before the feed does
Sooner or later a feed sends an empty field, a past date or a number as text. Decide what each means before launch, because the page turns it into a promise.
- A missing lead time isn’t zero. The browser converts text to numbers, because the ERP has sent numbers as text before. But a blank never becomes 0 (in JavaScript,
Number('')is 0), or an item nobody has ordered would “ship tomorrow”. - A past date is a missing date, not “already arrived”. The line falls through to the lead time.
- Silence beats a guess. If nothing resolves, the line stays hidden. The merchant can enter a fallback date, shown only when Shopify says the item is out of stock.
- Print the year once lead times run long. In October, “ships from 30 March” leaves the shopper guessing.
Check three products before every holiday season
Before Easter and before Christmas, pick three products: one in the warehouse, one with an arrival booked, one special order. Count each date yourself on the warehouse’s calendar and compare it with the page. If they disagree, the data or the calendar is wrong. Add next season’s closure days while you’re there.
Then check the other places that promise a date. If your checkout shows one, make it read the same stock and the same rule. Shopify now also sends sales channels a delivery range for each variant, in calendar days. My reading: a page that counts working days and a feed that counts calendar days can promise one shopper two dates.
Three things I’d like Shopify to build in
As of November 2026, Shopify doesn’t offer these, and each would retire a piece of this build:
- Product-page dates in every country Shopify already predicts dates for, from the checkout engine. That would retire our date engine.
- A working calendar per shipping location that themes can read. Shopify already keeps a fulfilment time in business days in the Admin API. Holidays and closures next to it would retire ours.
- ERP arrival dates in Shopify’s incoming inventory. Liquid already has
variant.next_incoming_date, filled by transfers, purchase orders and apps. If the ERP wrote there, a metafield and a rule would go. Ask your integration partner whether it can.
If I’ve got any of this wrong, especially Shopify’s delivery-date rules, I’d like to hear it.
Frequently asked questions
Can Shopify show an estimated delivery date on the product page without an app?
As of November 2026, only through Shop Promise: its product-page badge can show an arrival date when automated delivery dates are on, and it needs a US location, Shop Pay and tracking on at least 90% of orders. Shopify’s manual and automated delivery dates show at checkout, in confirmation emails and on the order pages, not on the product page. Everywhere else, the theme or an app has to work the date out.
Why not calculate the delivery date in Liquid?
Liquid’s current date reflects when the page was last rendered, and Shopify caches rendered pages, so a date worked out on the server can still carry yesterday’s date. Liquid also can’t tell you how many units each location holds. Let Liquid print the data and the sentences, and compute the date in the browser, in the store’s own time zone.
How do you handle public holidays in a Shopify delivery estimate?
Compute them per country: the fixed dates plus the ones that move with Easter, for this year and the next two. Keep a short list, maintained by the merchant, for the store’s own closure days. Holiday laws do change, so check the computed list once a year.
What should the product page show when an item is out of stock?
If an arrival is booked, a ships-from date: the arrival plus one working day. If not, a special-order date: today plus the lead time in calendar days, then one working day. If neither is known, we’d rather show no date than a wrong one.
Which country’s holidays apply when customers order from abroad?
The holidays of the store that ships. The parcel leaves that store’s warehouse on that country’s working days, wherever the shopper is. Key the calendar on the store, not on the visitor’s country or language.
Takeaway
The warehouse already knows what’s on the shelf, and its wall calendar knows which days it’s closed. Put both on the product page, and “when will it arrive?” is answered before anyone has to ask.
Working on something similar? I’m always happy to compare notes: get in touch or find me on LinkedIn.