Most merchants we talk to upgraded under duress. An agency or an app vendor did the minimum to keep checkout functioning, the deadline passed, and nobody went back to ask what the new stack could actually do. That is the gap this article covers: what quietly broke, and what is now on the table.
What quietly broke if you did the minimum
The upgrade path was framed as "click the button and you're done". For plenty of stores, that was not true.
Shopify Scripts died on 30 June this year. If you were on Plus and using Scripts for tiered discounts, payment method reordering or shipping rate tweaks, that logic stopped executing. Some stores are still discovering this through margin erosion or support tickets, because nothing visibly errors - the logic just isn't there any more.
Tracking is the other silent casualty. Additional scripts on the Thank you and Order status pages went view-only in August 2025, and for non-Plus stores they stop firing entirely on 26 August 2026. That field is where a decade of Google Ads conversion tags, Meta pixel purchase events and affiliate postbacks accumulated. When it goes, your ads keep spending but your platforms stop seeing revenue. Reported ROAS sags over weeks and it looks like a performance problem, not a tracking one. The fix is moving everything to web pixels and app pixels, and validating events end to end before the cutover, not after.
Smaller losses add up too. Order status page content boxes built with addContentBox are gone. Script-tag apps that injected widgets post-purchase have been progressively blocked since early 2025. If your post-purchase experience feels thinner than it used to, this is why.
What the new stack actually enables
The honest upside: the replacement is better than what it replaced, because it is upgrade-safe. Customisations built on the new primitives survive every checkout update Shopify ships, which was never true of checkout.liquid hacks.
There are three building blocks. Checkout UI extensions render your own components inside checkout - banners, custom fields, dynamic content - in sanctioned placement slots. The checkout branding API gives programmatic control over typography, colour, corner radius and layout well beyond the theme editor. And Shopify Functions replace Scripts with server-side logic for discounts, cart and checkout validation, delivery customisation and payment customisation, running fast enough to sit in the hot path of checkout.
Customisations actually worth doing
Some examples from builds we would defend on conversion grounds, not novelty.
Duties and taxes messaging. If you sell internationally, ambiguity about landed cost is a killer. A UI extension that tells a New Zealand or US buyer exactly what is and is not included - DDP versus DDU, GST handling, who pays what at the border - removes the hesitation that shows up as abandoned checkouts from overseas IPs.
PO number and order reference fields. B2B buyers often cannot get an invoice paid without a purchase order number on it. A custom field at checkout that writes to order metafields (or the native B2B PO field if you are using company profiles) means their accounts payable process works first time, and you stop fielding "can you add our PO to invoice 4471" emails.
Delivery logic. Delivery customisation Functions let you hide, rename or reorder shipping options based on cart contents and destination. Dangerous goods that cannot go to a PO box, bulky items restricted to metro carriers, express options hidden when the warehouse cutoff has passed. This logic used to live in Scripts or apps; now it is native and fast.
Validation that protects margin. Checkout validation Functions can block orders that break your rules - quantity limits during a launch, trade customers dipping below minimum order value, mismatched shipping regions - with a clear message instead of a silent cancellation email two days later.
Upsells that don't tank conversion. The pattern that works is post-purchase offers, shown after payment is captured and before the Thank you page. The buyer has already converted, so there is zero risk to the primary sale. In-checkout upsells can work as a single quiet line item suggestion, but treat anything that adds friction before the pay button with suspicion.
When you need Plus and when you don't
The line is clearer than most vendors let on. On any paid plan you get UI extensions on the Thank you and Order status pages, post-purchase offers, customer account extensions, web pixels, and every Functions category - discounts, validation, delivery and payment customisation - via public apps.
Plus buys you three things: UI extensions on the information, shipping and payment steps themselves, the checkout branding API for deep design control, and the ability to run custom Functions apps rather than off-the-shelf ones. If your differentiation lives inside checkout proper - B2B fields, bespoke delivery logic tied to your ERP, brand-critical design - that is the Plus case. Plus currently starts at about USD $2,300 a month on a three-year term, so the maths has to be about revenue, not vanity.
Where to start
Audit before you build. Pull the list of what your old checkout did - every script, every app, every discount rule - and check what survived. Most stores that "upgraded" in 2025 are running at maybe 70 per cent of their old capability and 10 per cent of what the new stack allows.
If you'd like a second pair of eyes on your checkout - what broke, what is worth rebuilding properly, and whether Plus stacks up for you - that is exactly the work Jaboo does for Australian merchants. See our Shopify services, or start an enquiry.