Common SEO Issues Faced by eCommerce Sites: Myths, Facts, and a Prioritised Fix List
Why your store underperforms organically - and which issues to fix first based on crawl budget, catalog size, and revenue impact.

Image via 123RF
In This Article
Myths vs. Index Reality
Diagnose Before You Fix
Catalog-Size Triage
Why eCommerce SEO Breaks Differently at Scale
eCommerce SEO failures rarely look like classic SEO failures. On a 50-page brochure site, a missing meta description is a minor inconvenience. On a 4,200-SKU Shopify store, the same mistake can be invisible against a backdrop of 610,000 indexed URLs - 92% of which are filter combinations nobody intended to rank. The damage is not a ranking drop on one page; it is a structural distortion of how Google allocates crawl budget, consolidates signals, and decides which URLs are worth surfacing. That is why checklist SEO breaks at scale: it treats symptoms in isolation instead of diagnosing the system generating them.
This article works through the most persistent myths about eCommerce SEO, checks each against how Google actually behaves, and then hands you a prioritised triage framework. Every issue below is something we at Prebo Digital have found in live US catalogs - not a theoretical list.
The Myth That More Pages Means More Traffic
The most common assumption we hear from growth managers is that every product variation, every filter combination, and every tag archive deserves its own indexable URL because "more pages means more chances to rank." It sounds logical. It is also the single largest driver of index bloat in eCommerce.
The Fact: Index Bloat Is a Crawl Budget Problem, Not a Content Problem
A 10,000-SKU catalog with five filter dimensions (colour, size, material, price band, availability) can generate millions of mathematically valid URL permutations. Google will discover a large share of them through faceted navigation links. Each URL consumes crawl budget that would otherwise be spent re-crawling your high-revenue product and collection pages - the pages where prices, stock status, and content actually change.
Indexed URLs on a 4,200-SKU Shopify catalog where 560,000 were filter combinations
Warning: Index bloat rarely announces itself with a penalty. It shows up as slower indexing of new products, stale prices in search results, and collection pages that lose rankings to thin filter URLs.
The fix is not to delete content. It is to control which URL patterns are crawlable, which are canonicalised, and which are excluded via robots.txt or meta directives. That control layer is what separates catalogs that scale from catalogs that slowly suffocate.
Myth 1 - 'Duplicate Content Will Get My Store Penalised'
This myth refuses to die. It is repeated in every SEO Slack channel and echoed in most audits. It is also, in eCommerce, misleading.
The Fact: Google Chooses a Canonical You Didn't Intend
Google has stated there is no "duplicate content penalty" in the sense of a ranking punishment. The real cost is subtler: when product descriptions are copied from a supplier feed and repeated across 1,800 SKUs, Google chooses which URL to show - and it may not be the one you want. Ranking signals, internal link equity, and any reviews or schema attached to your preferred URL get consolidated onto a variant, a paginated URL, or a competitor's page entirely.
How to Detect Canonical Mismanagement
Run a crawl of the catalog and compare self-referencing canonicals against the URL Google has actually selected. Then cross-check in Google Search Console under the Page Indexing report for the "Duplicate, Google chose different canonical than user" reason. If that count is above roughly 5% of your indexed URLs, you have a canonical problem, not a content problem.
# Screaming Frog custom extraction - flag mismatched canonicals
Canonical Link Element | HTML | //link[@rel='canonical']/@href
# Filter: Canonicalised URL != Address
# Then cross-reference against GSC 'Google chose different canonical'
Myth 2 - 'Out-of-Stock Products Should Be Deleted'
When a product runs out, the instinct is to remove it from the catalog to keep the store tidy. For a discontinued SKU, that instinct is expensive.
The Fact: Deletion Destroys Accumulated Link Equity
A product page that has been live for two years accumulates backlinks, internal links, branded search history, and sometimes user reviews. Deleting it returns a 404, drops those signals, and clears the page from Google's index. The brand never gets that equity back on a replacement URL without a redirect.
Google's guidance: keep out-of-stock products live with an availability status. Use structured data to reflect availability rather than removing the page.
A Decision Framework for Out-of-Stock and Discontinued SKUs
| Scenario | Action | Schema / Technical Note |
|---|---|---|
| Temporarily out of stock | Keep live, show availability | Offer availability = OutOfStock; add restock alert |
| Permanently discontinued, no replacement | 301 to closest category | Remove Product schema; keep category schema |
| Discontinued with direct replacement | 301 to replacement PDP | Carry over Reviews and Breadcrumb schema where relevant |
The default should be keep-and-signal, not delete. Deletion is a last resort reserved for SKUs that will never return and have no logical successor.
Myth 3 - 'Category Pages Rank Automatically if Products Are Optimised'
The Fact: Collection Pages Are the Commercial-Intent Winners
Head and mid-tail queries - "linen bedding US", "organic whey protein", "stainless steel water bottle" - are overwhelmingly commercial-investigation. Google surfaces collection and category pages for these, not individual PDPs. If your collection pages are auto-generated thumbnails with no unique copy, no internal links to high-margin products, and no FAQ or buying-guide content, you are ceding the highest-value organic real estate in your store.
What Thin Category Copy Actually Costs You
Thin category pages do not just fail to rank. They also fail to convert the traffic that arrives from brand queries. A US home-goods store we reviewed had 42 collections with under 80 words of copy each. After expanding the top 12 by revenue with buying-guide content, comparison tables, and internal links to best-sellers, organic sessions to those collections rose measurably within a quarter - and average order value from organic traffic rose too, because the pages did the job of an in-store salesperson.
Myth 4 - 'Page Speed Is a Developer Problem, Not an SEO Problem'
The Fact: Core Web Vitals Are Confirmed Ranking Signals
Speed is not a vanity metric. LCP, INP, and CLS are part of Google's page experience signals, and at scale they intersect directly with crawl efficiency: slow servers and heavy JavaScript reduce how many URLs Googlebot can crawl per session.
INP Replaced FID in March 2024 - Is Your Audit Stale?
Any audit quoting FID is out of date. Interaction to Next Paint became the responsiveness metric in March 2024, and it is stricter: it measures the worst interaction latency across the page visit, not just the first input. On Shopify and WooCommerce PDPs, the usual culprits are third-party review widgets, chat scripts, and tag managers loading on the main thread.
The 'good' INP threshold on mobile - anything above 500ms is marked poor in CrUX
Myth 5 - 'Structured Data Is Optional Garnish'
The Fact: Product, Offer, and Review Markup Drive Rich Results and AI Surfaces
Product schema is no longer decoration. It is the machine-readable layer that powers price, availability, rating, and shipping annotations in search results - and increasingly the layer that AI shopping answers lean on to describe your product accurately. Missing or invalid markup means your product is described by a competitor's data or omitted entirely.
Required vs. Recommended Properties in Google's Product Schema
Google's documentation distinguishes between properties required for a valid Product snippet and those needed for richer treatments. The short version: name and at least one of offers or review are the baseline. Price, priceCurrency, and availability are required inside Offer. Aggregated rating powers star ratings. Low-price and shipping details trigger further enhancements. The most common failure we see in audits is valid Product markup with missing Offer priceCurrency - which silently disqualifies the page from the richest treatments.
A Prioritised Triage Framework for Common eCommerce SEO Issues
Not every issue deserves equal urgency. The framework below ranks issues by revenue impact, not by how alarming they look in an audit. Tier 1 issues block revenue already owed to you. Tier 2 limits future revenue. Tier 3 is upside you capture once the first two tiers are stable.
Tier 1 - Revenue-Blocking Issues (Fix First)
These issues are actively suppressing indexed, crawlable, rankable pages. Treat them as fires.
| Issue | Detection | Fix |
|---|---|---|
| Faceted-nav index bloat | GSC Pages report; crawl URL count vs. SKU count | Disallow or canonicalise filter params; block in robots.txt |
| Broken canonicals | Screaming Frog; GSC 'Google chose different canonical' | Self-referencing canonicals on all PDPs and collections |
| Accidental noindex | GSC URL Inspection; crawl meta robots | Remove noindex from templates; redeploy |
| Soft 404s on OOS pages | GSC 'Soft 404' reason; content-size checks | Return 200 with availability schema, not thin content |
Tier 2 - Revenue-Limiting Issues (Fix Next)
These issues do not block indexing but cap how much you can rank and convert. Thin category copy, orphan pages, weak internal linking between PDPs and parent collections, missing Product schema, and slow LCP on PDPs all sit here. They take longer to fix and take longer to show results, but they compound.
Tier 3 - Upside Issues (Fix When Tier 1 and 2 Are Stable)
Pagination best practice, hreflang for cross-border, log-file-informed crawl tuning, and Google Merchant Center feed optimisation live here. They are real opportunities, but fixing them before Tier 1 or 2 is rearranging furniture in a burning building.
How Catalog Size Changes Your Priority Order
Small Catalogs (Under 500 SKUs): Content and Schema First
Under 500 SKUs, index bloat is rarely the dominant problem. Your constraint is that you have too few URLs to compete for head terms. Priorities invert: build deep, useful collection content, add FAQ and buying-guide depth, get Product and Breadcrumb schema right, and use internal linking to push authority into revenue collections.
Large Catalogs (10,000+ SKUs): Crawl and Index Control First
At 10,000 SKUs and above, index bloat, canonical management, and crawl budget dominate. Content depth is still important - but only on the handful of collections and PDPs that can realistically rank. The other 90% of your effort should go to making sure Googlebot spends its time on pages that change and pages that earn.
The Detection Toolkit: What to Use for Each Issue Type
Google Search Console Reports You Should Check Weekly
- Page Indexing report - canonical mismatches, soft 404s, excluded-by-noindex
- Crawl Stats - host status, crawl frequency by Googlebot type
- Core Web Vitals report - URLs grouped by LCP, INP, CLS status
- Enhancements / Merchant Listings - Product schema errors and warnings
- URL Inspection - per-URL live check including rendered HTML
- Sitemaps - submitted vs indexed ratio
Crawlers, Log File Analysis, and Speed Tools
GSC alone will not show you crawl waste. Pair it with Screaming Frog or Sitebulb for structural crawls, and a log file analyser (Screaming Frog Log File Analyser, GoAccess) against your Cloudflare or CloudFront logs. Log analysis is the only way to see what Googlebot actually fetches, how often, and what proportion of requests hit low-value URL patterns. For performance, benchmark PageSpeed Insights against CrUX field data - lab scores and real-user scores diverge meaningfully on PDPs with heavy third-party scripts.
# Quick crawl-waste check on a Cloudflare log export
grep -i 'Googlebot' access.log \
| awk '{print $7}' \
| grep -E '\?|/filter/|/tag/|orderby=' \
| sort | uniq -c | sort -rn | head -20
Platform-Specific SEO Issues: Shopify vs. WooCommerce vs. Magento
Shopify: Filter Apps, Variant URLs, and .json Endpoints
Shopify's URL architecture is clean by default, but two things break it. First, filter and search apps often create parameter URLs that are crawlable and indexable. Second, every product and collection exposes a .json endpoint that returns 200 - crawlers and scrapers can reach it, though it is normally not linked. Variant URLs (?variant=) are canonicalised correctly by default, but app-injected filters are not.
WooCommerce: Tag Taxonomies, ?orderby= Parameters, and Pagination Bloat
WooCommerce produces /product-category/ and /product/ URLs, plus tag archives and attribute taxonomies that multiply indexed URLs fast. Pagination (/page/2/) is often indexable and thin, and ?orderby= parameters create sort-order duplicates.
Magento and BigCommerce: Layered Navigation as a Crawl-Budget Risk
Magento and BigCommerce layered navigation generates a large number of parameter-based URLs by design. On Magento especially, the default filter configuration can expose hundreds of thousands of permutations on a 10,000-SKU catalog. Both platforms need deliberate configuration of which filter values are crawlable.
Three Real-World Scenarios: Before and After
Scenario 1 - A US Home-Goods Shopify Store With 610,000 Indexed URLs
A 4,200-SKU US home-goods brand had 610,000 indexed URLs, 92% of them filter combinations from a faceted navigation app. New products took three to four weeks to index. Fixes: robots.txt disallow on the filter parameter pattern, self-referencing canonicals on PDPs, and removal of the filter app's sitemap submissions. Within 12 weeks, indexed URLs fell to roughly 58,000 and new-product indexation dropped under 72 hours.
Scenario 2 - A WooCommerce Supplements Brand With 1,800 Near-Duplicate Descriptions
A supplements brand had 1,800 SKUs with descriptions copied verbatim from a supplier feed, all with valid Product schema but identical body copy. Canonicals were self-referencing and correct, but Google consistently chose category pages over PDPs for mid-tail queries. Fix: rewrote the top 120 SKUs by revenue, added unique FAQ blocks per product, and built comparison tables linking to sibling SKUs. Organic sessions to those 120 PDPs rose 34% in a quarter - the rest of the catalog stayed untouched until the template proved out.
Scenario 3 - A Brand That Deleted 300 Discontinued SKUs and Lost Backlinks
A US outdoor brand deleted 300 discontinued SKUs, returning 404s and freeing the URLs from the CMS. Webmaster Tools showed a steady drip of backlinks pointing to dead URLs for months. The fix involved mapping the old URLs to appropriate collection pages via 301, restoring partial link equity, and retroactively applying the keep-and-signal pattern to remaining OOS SKUs. The lesson: deletion is a decision with a six-to-twelve-month tail.
AI Search and Product Feed Optimisation: The Emerging Issue Category
Why Google Merchant Center Feeds Now Affect Organic Visibility
Merchant Center feeds and organic structured data increasingly share the same underlying product data. If your feed lists a different price, GTIN, or availability than your PDP schema, you create entity conflict - and AI shopping surfaces lean toward the more consistent source. Feed hygiene is no longer a paid-only concern.
Crawlability of JavaScript-Rendered PDPs for AI Crawlers
AI crawlers frequently do not execute JavaScript the way Googlebot does. A PDP whose price, description, and availability only render client-side may be invisible to those crawlers even if Googlebot handles it. For any PDP critical to organic performance, server-render the essential content and validate by fetching the raw HTML.
Tip: Fetch a PDP with curl and grep for price and GTIN. If they only appear in scripts, your AI visibility is weaker than your Google visibility.
US Compliance Context: CCPA, Consent, and Measurement Completeness
California's CCPA and CPRA give US consumers meaningful control over data collection. Consent management affects how much traffic is attributable in GA4, which in turn affects how confidently you can correlate organic investment with revenue. For pure-US stores, this typically means adopting consent mode, using server-side tagging through a first-party endpoint, and modelling the measurement gap rather than pretending it doesn't exist. For a complementary view on how tracking completeness interacts with performance reporting, see Prebo Digital's services overview.
One nuance worth flagging: hreflang matters relatively little for a US-only store. It becomes relevant if you also sell into Canada, the UK, or the EU - in which case locale and currency signals need to align with PDP content and feed data, not just URL structure.
Sources
- Google Search Central - Crawling and Indexing Documentation
- Google Search Central - Canonical URLs and Duplicate Content Guidance
- Google Search Central - Product Structured Data (Product, Offer, Review, Merchant Listing)
- Google Search Central - Core Web Vitals and Page Experience Guidance
- web.dev - Core Web Vitals (LCP, INP, CLS) Definitions and Thresholds
- Google Search Console Help - Page Indexing Report


