Your Products Are Live. So Why Can't Google Find Them?
Here's a scenario that plays out in ecommerce teams more often than most people admit. The store has a few thousand products live, the catalog team ha...

Here's a scenario that plays out in ecommerce teams more often than most people admit. The store has a few thousand products live, the catalogue team has done a solid job with photos and descriptions, and the merchandising calendar is full. Yet when someone types the exact name of a new product into Google, the page doesn't appear. Sometimes it's not even in the index at all—meanwhile, a competitor with a smaller catalogue ranks for those same product names without breaking a sweat.
The instinct is to blame content or backlinks. Sometimes that's right. But very often, the real issue is technical, and it's hiding in plain sight. The store's platform is quietly generating URLs nobody asked for, the filter menu is producing tens of thousands of near-identical pages, the canonical tags are sending mixed messages, and discontinued products are sitting in limbo with no clear signal about what happened to them.
This is the central question we want to answer: how can an ecommerce store have thousands of active products and still struggle to get those products discovered and indexed by Google? The answer almost always comes down to a gap between what the store thinks it's publishing and what Google actually finds when it crawls the site. Google needs to find each product page, understand which version of that page matters, and decide it's worth storing in the index. Every technical quirk that muddies those three steps makes the job harder.
We'll work through five practical areas: product variants and duplicate URLs, faceted navigation and crawl budget waste, canonical tags and internal linking, out-of-stock and discontinued product handling, and a practical Shopify technical SEO audit checklist you can run yourself. Along the way, we'll keep the focus on what you can actually check and change, rather than abstract theory.
Let's start with a useful mental model, because it makes every later problem easier to understand.
Live Does Not Mean Indexed
A product being "live" on your store simply means it's accessible to a visitor who knows where to look. Being "indexed" means Google has crawled the page, processed it, and stored it in its database as a candidate to show in results. Those are two very different states, and the gap between them is where a lot of ecommerce SEO problems live.
Google describes its process roughly as discovery, crawling, and indexing. A URL has to be found (through a sitemap, an internal link, an external link, or a feed), then fetched and rendered, then evaluated for whether it's a unique, worthwhile page that should be stored. At each stage, things can go wrong, and the store owner usually only sees the result, not the reason.
Google Search Console makes some of these reasons visible through its page indexing report. A few statuses come up constantly for ecommerce sites. "Discovered, currently not indexed" means Google knows the URL exists but hasn't bothered to crawl it yet, which often signals that Google doesn't think the URL is important enough to prioritise. "Crawled, currently not indexed" is more interesting: Google visited the page and decided not to keep it, which frequently happens when pages look thin, near-identical to others, or low in value. "Duplicate, Google chose different canonical than user" tells you that you declared one URL as the preferred version, but Google disagreed and picked another. And "Duplicate without user-selected canonical" means Google found multiple versions of a page and you never told it which one to prefer.
Every one of these statuses points back to a technical or structural decision. That's good news, because technical decisions can be changed. It's also why the first step in any ecommerce indexing problem should be reading the indexing report carefully rather than guessing.
With that foundation in place, let's look at the most common culprit for product-level confusion: variants.
Product Variants and Duplicate URLs
If you sell clothing, shoes, furniture, cosmetics, or electronics, you almost certainly deal with variants. A single t-shirt might come in six colours and five sizes, which makes thirty purchasable options. A single lamp might come in three finishes. The merchant sees one product. The platform sees a handful of combinations that each need to be represented on the page somehow.
On Shopify, the default behaviour is that variant selection happens on the main product page. When a shopper picks a colour or size, the page updates, and the URL often changes to include a variant parameter identifying that specific option. These parameterised URLs are useful for the shopper and for things like tracking a specific selection. For search, they create a problem. Each variant parameter is technically a distinct address that a crawler can find, fetch, and potentially treat as its own page.
Most of the time, these variant URLs display almost identical content. The title, the description, the reviews, and the main imagery are the same. The only real difference is which option is pre-selected. From Google's perspective, that's a lot of duplicate material competing for the same search intent. Ideally, Google would consolidate all of those versions into one canonical product page. When the store sends clear signals, that usually happens. When the signals are muddled, you get a mess of indexing decisions that don't match what you want.
Here's where it gets more nuanced. Not every variant deserves to be ignored. Some variants correspond to real search demand. People do search for "navy leather belt size 34" or "walnut dining table 72 inch," even though the base product might be simply "leather belt" or "dining table." Google's own ecommerce guidance acknowledges that how you handle variants should depend on whether they represent meaningfully different products for shoppers.
We use a framework called the Variant Index Exposure Model to sort variants into three tiers.
Tier one: consolidate. These are variants that differ only in a minor way, such as a slightly different shade of the same neutral colour, or a size that shoppers rarely search for independently. They should be consolidated under the main product URL. The variant selector stays on the page, the canonical points to the main product, and the variant URLs don't need to be indexed. Most variants on most stores fall into this tier.
Tier two: selectively index. These are variants with proven or plausible search demand that deserve their own visibility. A colour that's a major product distinction, a material that changes how the item is used, or a configuration that people specifically search for often belongs here. For these, you may want a dedicated, indexable page with a unique title, a description that explains what makes this version different, and a clear path from the main product. This isn't something to do for every variant, because creating dozens of thin variant pages dilutes your site and wastes crawl attention.
Tier three: isolate. These are variants that are effectively separate products, such as a different model with a different spec sheet or a different use case. They shouldn't be treated as variants at all. Give them their own product entries, their own handles, and their own structured data, and let them stand on their own.
A common mistake is treating variants as either all-indexable or all-blocked. Both extremes waste something. Indexing every variant floods the index with near-duplicates. Blocking everything can hide a legitimately searched product configuration from Google entirely. The right answer depends on search demand, and you can check that with keyword research on variant-specific terms and by reviewing the queries that already bring traffic to your site in Search Console.
One more point about variants that often gets overlooked: the content problem. If all twelve variants of a sweater share the same description, the same bullet points, and the same generic alt text, you've created a duplicate content issue that the variant structure didn't cause but certainly amplified. Writing a short, specific description for variants in the second tier, and making sure your main product description covers the variant range honestly, goes a long way toward making each page worth indexing.
A Hypothetical Illustration
Consider a store with 800 products, where the average product has six colour options and five sizes. If every combination generated a separately crawlable URL, the store would have around 24,000 variant URLs before any filters or collections were even counted. In reality, most of those URLs would show the same content with a different pre-selected option. If the store's search data showed that only about thirty of its products had variant-specific search demand, the Variant Index Exposure Model would point toward consolidating nearly everything and building dedicated pages for a small, targeted set of high-demand configurations. The numbers here are invented for illustration and aren't drawn from a real store.
Faceted Navigation and Crawl Budget Waste
If variants are a problem of a few hundred extra URLs, faceted navigation is where things can spiral into tens of thousands.
Facets are the filters on a category or collection page: size, colour, brand, price range, material, availability, rating, and so on. They help shoppers narrow a large catalogue down to something manageable, and they're genuinely valuable for users. The trouble starts when each filter choice creates its own crawlable URL, and those filters can be combined.
Suppose a women's footwear collection has filters for brand (fifteen options), size (twelve options), colour (ten options), width (three options), and price range (five options). Even if a shopper only ever uses one or two filters at a time, a crawler exploring every combination could be looking at a number of URLs that runs into the tens of thousands. Add sort parameters like "price low to high" or "newest first," and the number grows again. Google's documentation on faceted navigation describes this as a potential infinite space problem, where a site can generate practically unlimited URL combinations with little or no unique content in them.
Why does this matter for product indexing? Because crawl budget is finite, at least in practical terms. Google's documentation explains that crawl budget is mostly a concern for large sites, and it's a combination of how much Google wants to crawl your site and how much your server can handle. For a small store with a few hundred products, crawl budget is rarely the bottleneck. For a large catalogue with heavy filtering, it can become one. When Googlebot spends its time working through filter combinations that have no search value, it has less time and attention left for the product pages you actually want indexed, especially new ones.
That's the core mechanism behind "crawl budget waste." It's not that Google is being punished. It's that the crawler is busy with low-value URLs when your high-value product pages are waiting in line.
To deal with this systematically, we use what we call the Facet Waste Audit. It sorts every facet on your site into one of three categories.
Demand facets are filters that correspond to real, searched phrases. "Waterproof hiking boots women," "queen size organic cotton sheets," and "stainless steel water bottle 1 litre" are examples. These combinations can be worth indexing because they match a specific search intent and often have their own unique intro text or curated product selection. Be selective here, though. A single demand facet like "waterproof" might be worth a page, while adding "women's" and "size 8" usually isn't.
Utility facets are filters that help shoppers navigate but don't represent a meaningful search. Sort orders, "show 48 per page," in-stock toggles, and most price range slider positions fall here. These should typically not be indexed, and you usually want to keep them from being crawled at scale.
Infinite facets are the combinatorial problem: multiple filters stacked in ways that produce near-endless unique URLs with nearly identical content. These are the main source of crawl waste and should be controlled aggressively.
Once you've classified your facets, you have several tools to manage them. Google's guidance mentions a few approaches, and each has ttrade-offs
Robots.txt can prevent crawlers from requesting certain URL patterns. It's effective at reducing crawl waste, but it has a catch: a URL blocked in robots.txt can still appear in search results if other pages link to it, and Google can't see any noindex tag on a page it isn't allowed to fetch. So robots.txt is best for facet patterns you're confident should never be crawled, not as a general index control. Shopify allows merchants to customise the robots.txt file through a template, and its default file already restricts some filtered collection patterns. Because those defaults can vary with your theme and apps, check the live file rather than assuming.
Noindex tags tell search engines not to keep a page in the index, but they still allow crawling, which means crawl budget can still be spent on those URLs. Noindex is useful for pages that are crawlable and useful for users but not worth indexing, like certain sorted or paginated views.
Canonical tags point to a preferred version of a page. Canonicalising filtered URLs to their parent collection is a common approach, but remember that canonicals are hints rather than directives. Google may ignore them if other signals contradict them, which is why the Canonical Signal Alignment Check later in this article matters so much.
Internal link handling is the quieter lever. If your filter links are plain, crawlable links on every page, you're effectively advertising the infinite space to Googlebot. Some stores address this by making filter interactions load through JavaScript without generating crawlable links, though this requires care so the content remains accessible and so that Google can still reach your important product pages through other paths.
The right combination depends on your platform, your theme, and which facets have real demand. The key principle is to decide deliberately rather than letting the defaults decide for you.
Watching for Pagination Traps
Collections with many products also paginate, and pagination can create its own version of the problem. Google stopped using rel next and rel prev annotations as an indexing signal back in 2019, so you shouldn't rely on those tags to tell Google how pages relate. What matters more is that every product is reachable through crawlable links, and that paginated pages carry enough unique context to be worth indexing. Deep pagination, where product 400 is only reachable by clicking through twenty pages, is a discoverability problem. Good sorting, sensible collection design, and strong links to the products themselves usually work better than hoping Google follows the pagination chain.
Canonical Tags and Internal Linking
Canonical tags and internal links are the two signals most directly under your control that tell Google which product page matters. They're also the two most frequently contradictory.
A canonical tag is an HTML element placed in a page's head that suggests which URL is the preferred version of that content. Google treats it as a strong hint, but not an absolute command. If you canonicalise a product variant to the main product, Google will usually follow that. If your sitemap lists one URL, your internal links point to another, and your canonical points to a third, Google has to pick, and it may pick something you didn't expect.
This is why we use the Canonical Signal Alignment Check. It asks whether four signals agree for any given product URL.
The canonical tag. Does the page declare itself or another URL as canonical, and is that declaration correct? Self-referencing canonicals on unique product pages are a healthy default. Canonicals on variant or filtered URLs should point to the appropriate parent page.
Internal links. Do the links across your site point to the same URL you're trying to index? If your collection pages link to a product using a variant URL, or if your blog links to an old handle, you're sending mixed signals. Internal links are among the strongest hints you control, because they reflect how you actually organise your site.
The sitemap. Does your XML sitemap list only the canonical versions of your products? Shopify generates sitemaps automatically, but they still include whatever URLs the platform considers public, so it's worth checking whether non-preferred versions are appearing.
Redirects and status. If a product has changed handles or been removed, do the redirects point to the right place, and do they resolve cleanly? Redirect chains and loops weaken signals and waste crawl activity.
When all four agree, Google tends to index the page you intended and consolidate duplicates correctly. When they disagree, you get the "Google chose a different canonical than the user" status and a lot of confusion about why a product isn't ranking.
Shopify generally sets canonical tags on product pages to point to the main product URL, and it typically handles product paths under collections in a way that points back to the primary product address. But custom themes, apps, and manual edits can override these defaults, and it's easy to end up with a canonical pointing to a filtered collection URL that you never intended. Inspect a sample of your product pages and filtered pages directly, looking at the canonical element in the page source, rather than assuming the platform is handling it correctly everywhere.
Internal Linking as a Discovery Engine
Canonicals determine which version of a page gets credit. Internal links determine whether a page gets found at all. A product with no internal links pointing to it is an orphan, and orphans are often the reason new products sit in "Discovered, currently not indexed" for weeks.
Think about how a product is actually reached. A shopper might arrive on the homepage, click a featured collection, scroll through a grid, and open a product. Googlebot does something similar, but it doesn't have intuition about which products matter. It follows links. If your strongest links flow toward older, high-traffic categories and your newest products sit three or four clicks deep with no supporting links, the crawler may treat them as low priority.
We find it useful to think about a metric we call Product Reachability. For each important product, ask how many clicks it takes to reach it from the homepage through crawlable links, and how many distinct internal links point to it from other pages. Products that are close to the homepage, linked from relevant collections, and mentioned in related content tend to get crawled and indexed faster. Products that are buried, rarely linked, or only accessible through a search box or JavaScript interaction tend to lag.
There are practical ways to improve this without rebuilding your site. Feature new products in relevant collections and on the homepage for a defined period. Link to them from related blog posts, buying guides, and category introductions where they genuinely fit. Use breadcrumbs consistently, since they create clear hierarchical links back to parent collections. Make sure related product modules link to actual product URLs rather than variant URLs. And avoid building navigation that only exists in a menu requiring interaction Googlebot can't perform, since links that aren't in the rendered HTML may not be followed reliably.
Internal linking also interacts with facets. A well-linked product should be discoverable through a clean, stable path that doesn't depend on a shopper or crawler stacking filters. If the only way to find a product is to select three specific filters, you've made it harder for both people and search engines to reach it.
Out-of-Stock and Discontinued Product Handling
Every ecommerce store eventually deals with products that aren't available anymore. The mistake most stores make is treating all unavailable products the same way, usually by deleting them or letting them sit in an ambiguous state. Each of those choices sends different signals, and some of them quietly cost you rankings and traffic.
Google's guidance generally distinguishes between products that are temporarily unavailable and those that are gone for good. A temporarily out-of-stock product that will return should usually keep its page, because that page has accumulated links, history, and search visibility that you don't want to throw away. Structured data can help here. Product markup allows you to specify availability, and Google's documentation for product structured data includes an out-of-stock value you can use to indicate that the item isn't currently purchasable. The page can still give shoppers useful information, like an expected restock date, a back-in-stock signup, or links to similar items.
For products that are gone permanently, the right answer depends on whether a close replacement exists. Google's guidance accepts both 404 and 410 status codes for removed pages, and it's worth knowing that Google treats 410 as a slightly more explicit signal that content is gone, though both are valid. The more important issue is avoiding soft 404s, where a removed product returns a page that looks like an error or an empty page while still returning a success status. Those pages confuse crawlers and can linger in the index.
Redirects are the other option, and they're powerful when used well. If a discontinued product has a clear successor, a 301 redirect to that successor is usually the right move, because it passes along the relevance the old page earned. The key word is relevance. Redirecting every discontinued product to the homepage or to a generic category is a common mistake. Google's guidance has long warned against redirecting unrelated content in bulk, and mass redirects to the homepage can look like a soft 404 in disguise, passing little value while confusing users who expected a specific product.
To make these decisions consistently, we built the Product Lifecycle Status Map, which sorts every product into one of five states.
Active. The product is in stock and publicly available. It should have a self-referencing canonical, be included in the sitemap, and be linked from relevant collections.
Temporarily unavailable. The product will return, perhaps seasonally or after restocking. Keep the page live, mark availability accurately in structured data, offer a restock notification or similar product suggestions, and avoid removing it from internal links unless the gap is very long.
Seasonal. The product is available only at certain times of the year. Consider keeping the page accessible and indexed, since it often regains traffic when the season returns, and make sure the content explains the seasonal window clearly.
Discontinued with a successor. The product is gone, but a clearly equivalent replacement exists. Set up a specific 301 redirect to the successor, update internal links to point to the new page, and remove the old URL from the sitemap.
Discontinued without a successor. The product is gone, and there's no close replacement. Return a proper 404 or 410, remove it from the sitemap and internal links, and consider linking to the parent category as a helpful alternative within the error experience, without redirecting everything to the homepage.
Shopify gives merchants the option to unpublish or delete products, and it allows URL redirects to be set up in the admin. Be deliberate about which path you take. Deleting a product that has accumulated backlinks and search history without setting up a redirect throws that value away. Keeping thousands of unavailable products live with no context creates thin, confusing pages. Setting up redirects in bulk without checking relevance can create the same problem as blanket homepage redirects.
Handling Out-of-Stock Products Without Losing Shoppers
There's a practical side to this that goes beyond search. Out-of-stock pages are often the page where a shopper is most motivated, since they've already found what they want and just can't buy it yet. A page that simply says "sold out" with nothing else wastes that intent. Better options include offering a notification signup, showing similar in-stock items, displaying an estimated restock date if you have one, and making sure the page still answers the questions the shopper had, such as sizing, materials, and care instructions. Keeping that information on the page supports both the user and the search engine's understanding of what the product is.
A Practical Shopify Technical SEO Audit Checklist
Everything above is only useful if you can check it against your own store. This checklist is designed to be worked through in an afternoon or two, depending on catalogue size. It's grouped by theme rather than by priority, so you can tackle the high-impact items first based on what you find.
Indexing health
Open the page indexing report in Google Search Console and note the counts for each excluded and not-indexed status. Pay special attention to "Discovered, currently not indexed," "Crawled, currently not indexed," and any duplicate statuses. Export a sample of affected URLs and categorise them by type: product, variant, collection, filtered, blog, or other.
Check whether the number of indexed pages is roughly in line with the number of products and content pages you actually want indexed. A large gap in either direction is worth investigating.
Variants
Identify the URLs generated by variant selection on a sample of products. Confirm that variant URLs canonicalise to the main product page. Run each product with meaningful variant differences through the Variant Index Exposure Model and decide which variants belong in each tier. Check whether your top-performing variant-specific search queries have dedicated, indexable pages or are being served by the main product page, and make sure the page targeting each query is the one you want Google to show.
Facets and filters
Map your filter types on collection pages and classify each as demand, utility, or infinite using the Facet Waste Audit. Check your live robots.txt file to see which filtered patterns are restricted. Check the canonical tags on a sample of filtered collection URLs. Review how filter links are rendered and whether they generate crawlable URLs. Look at your server logs or crawl statistics in Search Console, if available, to see how much crawling is going to filtered URLs compared to product URLs.
Canonicals
Sample a range of product, variant, collection, and filtered URLs and inspect the canonical element in each. Confirm that unique product pages are self-referencing, that variants point to the main product, and that filtered pages point to the appropriate parent. Flag any canonical that points to a URL returning a redirect, a 404, or a page blocked by robots.txt, since these contradict themselves.
Compare the canonical URL against the URL listed in your sitemap and the URL used in internal links. All three should match for every important product.
Internal linking
Confirm that every active product is reachable through at least one crawlable internal link. Identify new products that have no inbound internal links and link them from relevant collections or content. Check that breadcrumbs are consistent and that they link to canonical collection and product URLs. Look for internal links pointing to outdated handles, variant URLs, or redirected pages, and update them to the final destination.
Product lifecycle
Pull a list of products currently marked unavailable, discontinued, or unpublished. Assign each to a state in the Product Lifecycle Status Map. For discontinued products with successors, verify the 301 redirect resolves to a relevant page. For discontinued products without successors, verify they return a proper 404 or 410 and not a soft 404. Remove discontinued URLs from the sitemap and from internal links. Check for redirect chains where a redirect points to another redirect.
Structured data
Test a sample of product pages with Google's Rich Results Test or the Schema Markup Validator. Confirm that product markup includes accurate name, image, price, and availability information. Make sure the availability value matches the actual stock state, since mismatches between structured data and what's shown on the page can cause problems. Check that variant information is represented accurately if your markup includes it.
Sitemap
Review your sitemap to confirm it includes canonical, indexable product and collection URLs only. Confirm that discontinued products and non-preferred variant URLs are excluded. Check that the sitemap is submitted in Search Console and that it's being read without errors.
Performance and rendering
Check that product pages load their core content in the initial HTML response or render reliably for crawlers. Content that only appears after complex client-side interactions can be harder for search engines to process. Review page speed on mobile for your highest-traffic product templates, since slow templates affect both user experience and how efficiently pages are crawled and processed.
Ongoing monitoring
Set a monthly routine to review the indexing report, check for new duplicate or excluded statuses, and verify that new product launches are being indexed within a reasonable window. Catching a faceted navigation change or a theme update early is much easier than discovering months of crawl waste after the fact.
Key Takeaways
Live products and indexed products are different things, and the gap between them is where many ecommerce SEO problems hide. The Search Console indexing report is the best place to start diagnosing that gap.
Product variants create duplicate URLs that should usually be consolidated under the main product page, with only a small, deliberately chosen set of high-demand variants given their own indexable pages. The Variant Index Exposure Model helps sort which variants belong in which tier.
Faceted navigation can generate enormous numbers of low-value URLs, which wastes crawl attention that should go to your product pages. The Facet Waste Audit separates demand facets worth indexing from utility and infinite facets that need to be controlled through robots.txt, noindex, canonicals, or careful link handling.
Canonical tags are hints, not commands, so they only work reliably when internal links, the sitemap, and redirects agree with them. The Canonical Signal Alignment Check helps confirm that all four signals point to the same URL for every important product.
Internal linking determines whether a product gets discovered at all. Products with no internal links or buried several clicks deep often sit unindexed. Improving product reachability through collections, breadcrumbs, and relevant content links can make a noticeable difference.
Out-of-stock and discontinued products need deliberate handling. Keep pages for temporarily unavailable items, redirect discontinued products to genuinely relevant successors, and return proper 404 or 410 responses for products with no replacement. Avoid blanket redirects to the homepage and avoid soft 404s. The Product Lifecycle Status Map gives you a consistent way to make these decisions.
A regular technical audit, built around the checklist in this article, catches problems before they compound and turns a vague frustration about rankings into a specific list of fixes.
About GEO SEO Lab
GEO SEO Lab is a research and content publication focused on the practical side of search visibility, from technical SEO and content strategy to the emerging demands of generative search. We write for marketers, developers, and business owners who want clear, verifiable guidance grounded in platform documentation and real diagnostic methods. Where we introduce original frameworks, we label them as ours, and where we rely on official guidance, we point you to the source so you can check it yourself.
References
Google Search Central, ecommerce and product site guidance (developers.google.com/search/docs)
Google Search Central, faceted navigation best practices documentation (developers.google.com/search/docs)
Google Search Central, canonicalization and duplicate content guidance (developers.google.com/search/docs)
Google Search Central, product structured data documentation (developers.google.com/search/docs)
Google Search Central, crawling and indexing overview and crawl budget documentation (developers.google.com/search/docs)
Google Search Central Blog, announcement on rel next and rel prev indexing signals (2019)
Google Search Console Help, page indexing report documentation (support.google.com/webmasters)
Shopify Help Centre, robots.txt customisation and URL redirect documentation (help.shopify.com)
Schema.org, Product and Offer type definitions including availability values (schema.org)
Tags
Frequently Asked Questions
Find answers to common questions about this topic
About the Author
Anubhav
SEO Expert & Content Creator
Experienced digital marketing professional specializing in SEO strategies, content optimization, and data-driven marketing solutions. Passionate about helping businesses grow their online presence and achieve better search rankings.
Related Articles
View all posts
Why Is My Website Not Ranking on Google? A Practical Diagnostic Path
A step-by-step way to find out whether Google can't see your site, doesn't trust it yet, or simply prefers someone else's page.

Which Platform Reaches the Right Audience? Why the Context Matters More Than the Logo
Which Platform Reaches the Right Audience? Why the Context Matters More Than the Logo. Google, Meta, and ChatGPT ads compared by what people are doing when your brand appears,

AI Content Is Cheap to Produce. Why Is It Still Expensive to Scale in 2026
Ask a room full of marketers what AI has done to content production, and you'll usually hear some version of the same answer: it's made things faster ...