Short answer. Search Console reports Soft 404 when a URL answers with HTTP 200 but its content reads to Google like an error or an empty page, which makes it ineligible for Search. On Shopify, three things produce it: collection sub-paths that answer 200 for any string, including tags that do not exist; filters and collections with nothing left in them; and an app or theme script that renders an empty page for Googlebot and a normal one for you.
A Soft 404 is the least intuitive status in the Page indexing report, because every part of the store it names looks healthy. The URL loads. The theme renders. The server returns a success code. Shopify admin has nothing to say about it. And Google has decided the page is not worth keeping in the index.
On a Shopify store the flagged URLs are frequently ones nobody created on purpose: collection paths with a stray segment on the end, filter combinations that match nothing, pages an app emptied out. That is what makes this a monitoring problem rather than a one-off repair. The failure is produced by the interaction of Shopify's routing, your theme, an installed app and Google's judgement, and no single one of those four reports it.
What Search Console means by "Soft 404"
Google's definition is short. The Page indexing report documentation gives it as "The page request returns what we think is a soft 404 response." The phrase "what we think" is doing real work there: this is a judgement about content, not a reading of a header.
The rule is stated more mechanically in Google's HTTP status codes documentation. On a 2xx response, Google "considers the content for processing", and then, "If the content suggests an error for Google Search, an empty page or an error message," then "Search Console will show a soft 404 error." So the status code and the content are evaluated separately, and the content can overrule the code.
Two consequences follow, and the second one is the one merchants miss.
- A Soft 404 page is not indexed. It sits in the Page indexing report among the reasons a URL is absent from Search, not among the warnings on an indexed page. If a product page is in this list, it is not competing for anything.
- Google's recommended fix depends on what you meant. The documentation gives both halves at once: "returning a 404 response code for truly “not found” pages and adding more information on the page to let us know that it is not a soft 404." If the page should be gone, make the server say so. If it should rank, the page has to carry enough content to look like a page.
It is worth separating this from the hard-404 case, because the two get discussed interchangeably and behave differently. A genuine 404 or 410 is unambiguous: Google's crawlers "inform the next processing system that the content doesn't exist", and "the indexing pipeline removes the URL from the index if it was previously indexed." That is a clean removal. A Soft 404 is a page Google keeps fetching and keeps declining. For the deletion case, and why a removed Shopify product can linger in results for weeks, see a deleted Shopify product still showing on Google. For how this status sits against the rest of the report, every status in the Page indexing report covers the full list.
Cause 1: your collection URLs answer 200 for anything
This is the cause that generates volume, and it is structural rather than accidental.
On a Shopify storefront, the path segment after a collection handle is where tag filtering lives. Shopify's own documentation describes the mechanism from the Liquid side: current_tags holds "The currently applied tags", and "product tags can be used to filter a collection page to show only products with specific tags." The consequence is that /collections/shoes/sandals is a legitimate URL if sandals is a tag.
The problem is what happens when it is not a tag. The URL still resolves. A merchant documented this on the Shopify Community on 2 March 2023, listing affected URLs including https://shoezero.com/collections/staff-favorites/flip-flopssid, the real collection plus three stray characters. Their description of the symptom was that "we identified several pages that are serving a soft 404 code", and that redirects they had set up "are not taking effect because these pages are still serving a 200OK code." A reply the next day confirmed the behaviour in one line: "All tag filters will load without a 404 so you're correct that anything would return a 200."
Two things make the surface larger than it first looks. Shopify's product documentation puts the per-product tag ceiling high: "You can apply up to 250 tags to each product", and notes that "Because your online store theme uses the tags as filters on collection pages," those tags turn into navigable combinations. The same page tells merchants that "Tags aren't used by search engines, so don't use tags to try to improve search results for your online store." That advice is about ranking, and it is sound. It is also where the confusion starts: tags are not a ranking input, but tag URLs are very much a crawling and indexing input, and they are the ones arriving in your Soft 404 list.
StoreCanary checks this on your store
Not sure what Google is actually being served? We fetch your public catalogue the way Googlebot does and flag pages whose indexing directives, status codes and content disagree with what admin shows: read-only, no admin access, about 2 minutes.
Scan my store for freeCause 2: a filter or a collection with nothing left in it
Google has documentation specifically about filter URLs, and it is the single most useful page to read on this subject. Its faceted navigation guidance explains why the crawl side matters: with filter URLs, "the crawlers will typically access a very large number of faceted navigation URLs", and "if crawling is spent on useless URLs, the crawlers have less time to spend on new, useful URLs." On a store whose products are not all indexed yet, that second sentence is the expensive one.
Buried in the same document is the instruction that explains most Shopify Soft 404 reports: "Return an HTTP 404 status code when a filter combination doesn't return results."
Read that against the previous section and the shape of the problem appears. An empty tag filter is the exact state Google says should return a 404, and it is the one state Shopify's collection routing will not produce. The Soft 404 label is Google filling that gap with a judgement of its own. This is the case where "just fix it" has no clean answer, which is worth knowing before you spend an afternoon looking for the setting that does not exist.
The same logic reaches ordinary collections that have emptied out. A seasonal collection whose products are all unpublished, a tag that no product carries any more, a filter menu still linking to combinations that have gone empty: each is a 200 response rendering a near-empty page, which is the content pattern in Google's 2xx rule. Nothing is broken. There is simply nothing on the page.
Cause 3: an app renders an error you never see
The third cause is rarer and much more damaging, because it hits pages you care about rather than filter URLs you do not.
A Shopify Community thread from mid-2022 records it from two stores at once. The original merchant reported that "Google began switching many of my core pages" to Soft 404, and that "my core pages are no longer available in organic Google search results." When he looked at Google's render of the page, rather than at the page in his own browser, he found third-party code he did not recognise; he later reported that "one of our apps was hacked and serving this code." A second merchant on the same thread reported, on 27 July 2022, that "On the 12th June Google started to record soft 404's on our site", counted 354 affected pages over the following weeks, named the app responsible, and reported that after removing it his URLs "all started to have green ticks." Those are two merchants' own accounts of their own stores, which is what makes them useful: nobody's dashboard told either of them, and the common factor was an installed app.
Mechanically this is the client-side-rendering case Google writes about separately. Its JavaScript basics documentation acknowledges that in a client-rendered app "using meaningful HTTP status codes can be impossible or impractical", and offers two ways to signal an error page: "Use a JavaScript redirect to a URL for which the server responds with a 404 HTTP status code", or "Add a <meta name="robots" content="noindex"> to error pages using JavaScript." An app that empties or replaces page content in the browser is doing the first half of that pattern by accident, without the second half.
Which is why this one is invisible to the obvious test. Loading the page yourself proves nothing, and neither does a raw fetch: the mismatch between served HTML and rendered HTML is the whole failure mode here, and it cuts both ways, as the structured data warnings article covers at more length. The only instrument that answers the question is Google's own render: in URL Inspection, run the live test and, as the documentation puts it, "click View tested page to see a screenshot showing how Google renders the page."
The fix Google documents, and the one Shopify leaves you
Google's faceted-navigation page lists the levers it recommends for filter URLs. It is a short list, and what is missing from it matters as much as what is on it:
- A 404 on empty results — the primary recommendation, and on Shopify collection sub-paths, not available. A developer reported this in a Shopify Community thread dated 29–30 December 2025: Online Store 2.0, in their words, "does not allow returning a real 404 HTTP status" for collection pages. They built a theme-level error state and tested it, and the outcome was that "Shopify still returns a 200 HTTP status even for invalid tag paths like /collections/valid/invalid."
- robots.txt — "Use robots.txt to disallow crawling of faceted navigation URLs." This is the lever Google calls most effective for the crawl-budget problem, and on Shopify it is editable through
robots.txt.liquid. It is also the lever most likely to do collateral damage, so read what a Disallow actually blocks on a Shopify store before adding a pattern. A rule written slightly too wide takes real product pages with it, and a crawl block is not a removal request. - Canonical tags — Google says using
rel="canonical"on faceted URLs may reduce crawl volume over time, while calling the approach "generally less effective in the long term" than the robots.txt route. Shopify already emits canonicals on filtered collection views, which is why this often appears to be handled and the Soft 404 count still climbs. Canonical selection is a separate negotiation with its own failure mode, covered in user-declared canonical versus Google-selected. - URL fragments and
rel="nofollow"— both listed, both requiring theme-level control over how every filter link on the store is emitted, and nofollow only working if every anchor pointing at the URL carries it.
The notable absence: that document never mentions noindex. And noindex is precisely what the December 2025 thread landed on after everything else failed: keeping collection pages fully functional and adding noindex, nofollow whenever current_tags is present, which the developer described as the best balance available under Shopify's constraints. It is a reasonable decision. It is also a decision to solve an indexing symptom because the documented fix is out of reach, and that is worth recording as such rather than presenting as best practice.
One caution from the same thread, and it is the kind of thing that turns a tidy fix into an outage. Another participant pointed out that a sub-path which looks invalid can be a real tag: a string you would read as a typo may be a filter a navigation menu or a customer link depends on. Check before you block or noindex a pattern. The same warning applies to the blunt version of this fix: disallowing collections/*/* wholesale was raised in that thread only as a last-resort measure, not as a recommendation.
For a product or collection page that genuinely should rank, none of the above applies and the answer is the plain one from Google's own guidance: the page needs enough content for Google to tell it is a page. A collection with no products, or a product page stripped down to a sold-out notice, is being judged on what it shows, not on what it means.
How to check your own store
- In Search Console, open Indexing → Pages and click the Soft 404 row. Export the list rather than reading it on screen — the next step is pattern work.
- Sort the URLs by shape: collection sub-paths, product pages, search or filter URLs, everything else. Each group maps to one of the three causes above, and they do not share a fix.
- Request one flagged URL and read the status code it actually returns. A 200 means Google's objection is to the content. A 404 or 410 means the label is stale and will clear by itself.
- For anything in the product-page group, run the live test in URL Inspection and open View tested page. This is the only step that catches cause 3, and your own browser cannot substitute for it.
- Before noindexing or blocking a collection sub-path pattern, confirm the segment is not a tag in use. One check against your tag list, and a filter your menu depends on stays alive.
- When the work is done, click Validate Fix on the issue — and then leave it: Google's documentation says to not click it again until validation has succeeded or failed, and that it can take around two weeks.
None of these steps is visible from inside Shopify. The admin has no Soft 404 view, no list of tag URLs Google has discovered, and no record of what an app rendered to Googlebot last Tuesday. Search Console is the only place the symptom appears, which is also why it tends to be found weeks late, during a check prompted by something else.
Why this one comes back
Soft 404 is not a defect you repair once. It is a steady-state property of a store with tags, filters, seasonal collections and installed apps, and every one of those keeps changing after you have finished cleaning up.
Add a tag and the navigable URL surface grows. Unpublish the last product in a collection and a page that was fine becomes near-empty. Install an app with theme access and the rendered output can change for Googlebot without changing for you. Update the theme and the empty-state template may be rewritten by someone who never saw your Search Console. Each of those is an ordinary operation that nobody would think to announce, and each can put a new URL into the Soft 404 list, or quietly put a product page there.
The useful question is therefore not "how do I clear this list", but "what tells me the next time it happens". Right now, for most stores, the honest answer is a report nobody opens between crises. That is the gap, and it is the same gap behind "Crawled – currently not indexed" and the rest of the indexing statuses: Google knows, and will tell you if asked, and will never volunteer it. If you want the fuller map of how a Shopify product goes missing from Google, the 11 silent failures collects them in one place.
Is this happening on your store right now?
Search Console will not page you when this happens. The product page loads fine in a browser, the sitemap looks clean, and Google quietly stops crawling or indexing it anyway. The only way to know is to check the pages the way Google sees them.
- 90+ checks, the way Google crawls and indexes your store. Noindex tags, canonicals, redirects, robots.txt, sitemap coverage.
- Read-only. No app to install, no admin access, nothing changes on your store.
- The exact fix for each page, in plain English, so you or your developer can act the same day.
Across full-catalog scans of more than 90,000 Shopify product pages, about 1 in 4 stores had at least one critical Google visibility issue.
Scan my store for freeFree scan: issue types, counts and first fixes. Full report on every product page, not a sample: $49 once, no subscription.