How to run a review of Shopify app-generated URLs
You install a Shopify app to add one neat feature. A few weeks on, a crawl turns up pages nobody on the team recognises, with odd parameters and thin content. Were they meant to be indexed? Who knows.
Shopify apps can add real functionality and leave behind URLs nobody intended to expose. The answer isn’t to block everything that looks unfamiliar. First find out what the app creates, why those pages exist and whether customers or search engines need them.
Inventory URL patterns
I use a crawl, site search and the app’s documentation to identify the routes, parameters and templates tied to the app. Compare example URLs with live navigation, and check response codes, canonicals, robots directives and internal links. The same path can behave differently depending on settings, so inspect the rendered pages and responses instead of guessing from a URL pattern.
Decide the intended treatment
Sort each pattern into useful, necessary for the app, duplicate, private or accidental. Confirm whether a page has a customer purpose and carries distinct information. Only then consider canonicalisation, noindex, authentication, redirects or removal, and check the app’s instructions and the crawl behaviour you want first. Robots.txt isn’t access control, and blocking crawling can hide page-level directives.
Retest after app changes
Record the app, its configuration and some example URLs. Review the patterns after updates or uninstalls, because routes and templates can change. Crawl representative paths and check Search Console where it’s relevant. Keep customer-facing functionality working while you cut the unwanted routes.
Write down what you decided
For each pattern, keep a short record: the app, an example URL, what it does, the treatment you chose and why. Add who confirmed it and when you’ll look again. It doesn’t need to be fancy. A shared sheet will do.
This matters most when the app gets replaced. A new app often brings new routes, and without the record nobody remembers which old ones were handled deliberately. Keep the notes close to the store’s other technical documentation, and link to the app’s own instructions instead of copying them.
Don’t forget the other direction. When you add a new app, ask the vendor up front which public URLs it creates and whether they can be switched off. Ten minutes of questions before installing is far cheaper than untangling indexed pages months later.
Apply the review to one real case
Say what a finished review changes for the person using the page, not just what work the team did. When you collect routes and canonical behaviour, use the same scope and definitions across examples and note what you excluded. A lopsided sample can make one app’s behaviour look like a pattern across the whole store.
Follow one real URL from the app setting that creates it to the published page, and check what a visitor can understand or do there. Before you act, confirm which source can verify the underlying fact. A copied export may describe an earlier app version or configuration.
If more than one team is involved, name the decision owner and the person who’ll verify the result. Separate the pages that share a template from the ones that share a user need, since a bulk rule may suit the first group and not the second.
Check routes and canonical behaviour in the live environment afterwards. An approved fix can still be altered by imports, other apps, caching or theme behaviour. If a question can’t be answered with the evidence you have, write the limit down and assign a follow-up.
When sources disagree, compare timestamps, definitions and ownership before choosing a value. If the information is incomplete, I’d go for a small reversible test or ask the app vendor over a sitewide rule, especially when a wrong change could affect several customer journeys. Allow an exception only for a genuine user or operational problem, and document its boundary.
Whoever owns the review should know where the current guidance lives and how to restore the previous state if a release creates a new problem. And before you repeat the approach with another app, check the setup and templates really are comparable. When you close the review, check nearby routes that depend on the same facts, such as feeds, support answers and linked destinations.
Related reading
- A Shopify catalogue review that starts with customer questions
- How to review organic landing pages against Shopify conversion paths
- Choosing a meaningful SEO review for a Shopify release
Which app-generated URL would you investigate first, and what evidence would decide its treatment?