Six Shopify Changes That Already Broke Something on Your Store

Platform changes that break loudly get fixed. Your checkout goes down, someone calls, it gets sorted the same day. The expensive ones are the changes that fail quietly, where the store keeps taking orders and something underneath has stopped working. Six of those landed in 2026. Here is what each one did, how to tell in a few minutes whether it hit you, and which ones actually matter for an Indian D2C store.

One, Shopify Scripts stopped executing on 30 June 2026

Not a future deadline. Editing ended on 15 April 2026 and execution ended on 30 June 2026, per Shopify's changelog.

If your store used Scripts for discount logic, shipping rules or payment customisation, that logic stopped applying on 1 July. Nothing failed visibly. Orders kept completing, just without your rules. The fingerprint is a discount that quietly stopped applying, or a shipping rule nobody has verified since June.

Check: if you ever had Scripts, place a test order that should trigger the rule. Rebuilding these as Shopify Functions is the path forward.

Two, Additional Scripts stops on the thank you page on 26 August 2026

The one still ahead of most non-Plus stores. Script tags on the Thank you and Order status pages sunset for non-Plus stores on 26 August 2026. Plus stores went through this on 28 August 2025.

This breaks tracking, not checkout. Google Ads conversion tags, Meta pixel fires and affiliate postbacks pasted into Additional Scripts stop firing, and your reporting develops a hole you will notice in October when the numbers stop making sense.

Check: open Settings, Checkout, and look at the Additional Scripts field. If there is anything in it, you have work to do. We covered what can and cannot be rebuilt, which differs sharply by plan.

Three, checkout validation moved to Functions on 2 July 2026

Shopify deprecated the useBuyerJourneyIntercept hook and the block_progress capability in favour of server-side validation Functions. No removal date has been announced and existing extensions still work, so this is the least urgent item here.

It is on the list because of a problem that predates the deprecation. A rule enforced in the checkout UI never fired when a shopper used an express wallet, so COD caps and minimum order rules have had a hole in them all along. The migration closes it.

Check: place a test order through Shop Pay or another express wallet that should be blocked by one of your rules. If it goes through, you have found the gap. More on rebuilding those rules.

Four, inventory buckets shifted on 5 August 2026

Shopify ran a one-time migration moving draft order and transfer holds from the reserved state to committed. Sellable quantity did not change, so this is genuinely nothing for most stores.

It matters if something reads inventory by bucket. A B2B operation quoting through draft orders, an ERP reconciliation, or a dashboard that splits committed against reserved will report differently than it did on 4 August.

Check: search your integration code for the string reserved. No hits means no exposure. Full detail here.

Five, delivery profile APIs were deprecated on 1 August 2026

Eight merchant-owned delivery profile APIs including deliveryProfileCreate and deliveryProfileUpdate were deprecated in favour of Contextual Product Feeds, app-owned delivery profiles, or Markets APIs.

Purely a developer surface, so most merchants are unaffected. It matters only if a custom app or agency-built integration sets your shipping rates programmatically.

Check: ask whoever maintains your custom code whether it touches delivery profiles. If you have no custom code, skip this one.

Six, invalid metafield queries now return errors

From 24 July 2026 the GraphQL Admin API returns errors for invalid metafield queries rather than handling them quietly. This is a correctness improvement and a good change, but any integration that was relying on the previous forgiving behaviour now gets an error instead.

Check: look at error logs for your custom apps and integrations since late July. If nobody looks at those logs, that is the more important finding.

The pattern worth noticing

Five of these six fail silently. That is not an accident of 2026, it is what platform maturity looks like: changes get rolled out carefully enough that nothing crashes, which also means nothing alerts you.

It also means the usual signal merchants rely on has stopped working. Historically, if something broke on your store a customer complained. None of these six produce a customer complaint. A shopper whose conversion event failed to fire has a perfectly good experience and buys the product. The only person harmed is you, months later, when you allocate budget using numbers that were wrong.

The practical consequence for a merchant is that "the store is working fine" has stopped being evidence that everything works. Tracking, discount logic, stock reporting and shipping rules can all be wrong while orders flow normally. The stores that catch these early are not the ones with better developers. They are the ones where somebody reads the changelog once a quarter and places a few test orders.

The test order sweep, in about an hour

Reading a changelog tells you what changed. Test orders tell you what your store actually does. This sweep catches five of the six items above without anyone opening code.

• Place an order that should trigger an automatic discount. Confirm it applies. This catches dead Scripts logic and broken promotion rules.
• Place an order that should be blocked. Over your COD cap, to a pin code you do not serve, below your minimum value. Then do the same through an express wallet, which is where the rule usually fails.
• Complete a real order and watch your tracking. Check that the conversion event arrives in Google Ads and Meta with the correct value and currency, not just that something fired.
• Check the shipping rates a customer actually sees for two or three destinations, including one you rarely ship to.
• Compare stock for one fast moving SKU against your warehouse or ERP figure.

Refund the test orders afterwards and keep a note of what you checked and when. Doing this once a quarter, plus once before your peak season, is the whole preventive programme. It costs an hour and a few transaction fees.

The reason it works better than a code review is that it tests the thing you actually care about, which is whether the store behaves the way you believe it does. Code can be correct while the configuration around it is wrong.

Which of these actually matter for a small store?

Two of them. The 26 August Additional Scripts change, because it breaks your tracking and almost every store has something in that field. And the Scripts execution change from June, if you ever used Scripts. The other four require custom code or draft order workflows that most small stores do not have.

How do I know if my store used Shopify Scripts?

Scripts were Plus only, so if you have never been on Shopify Plus you never had them. If you are on Plus or migrated from it, check with whoever managed your discounts before mid-2026, and test any promotional logic that is meant to apply automatically.

Will Shopify tell me if something breaks?

Not reliably, and that is the core problem. These changes are announced in the changelog and the developer changelog rather than pushed into your admin as alerts. Deprecation notices go to app developers, who may or may not pass them on to you.

How often should I audit for this?

Quarterly is enough for most stores, plus a check before peak season. The audit is not technical. Place test orders covering your discount rules, your shipping rules, your checkout restrictions and your tracking, and confirm each one still does what you think.

What is the most common thing you find?

Tracking that stopped working and nobody noticed, usually followed by a discount rule that has not applied for months. Both are quiet, both cost money continuously, and both are found in an afternoon by someone who thinks to look.

Get a platform change audit

Free check. Email hello@exactwhy.com with subject "Platform audit" and your store URL, your plan, and whether you have any custom apps. We tell you within 4 hours which of these six apply to you. For most stores it is one or two, and we will say so rather than inventing a project.

Paid audit and fixes, Rs 25,000 to Rs 1 lakh. We test every rule your store is supposed to enforce, verify tracking against real orders, rebuild what has broken, and give you a list of what to check before the next peak.

Ongoing Shopify development, Rs 20,000 to Rs 50,000 a month. Including reviewing platform changes against what your store actually runs, which is the only version of this that is genuinely preventive.

None of these six was hidden. All of them were published. The gap is not information, it is that nobody whose job includes reading the changelog also knows which parts of your store depend on what.