Key takeaways
- Category pages, not product pages, are where most online stores win non-brand organic traffic. Build them around how shoppers search.
- Google Merchant Center is now part of SEO. Free listings can show your products across Search, Shopping, Images, Lens, YouTube, Maps and Gemini.
- AI referrals are small but valuable: Adobe found traffic from generative AI tools to US retail sites rose 693.4% in the 2025 holiday season and converted 31% better than other traffic.
- Faceted navigation is the most common way large stores waste crawl budget. Decide which filter combinations deserve indexing and block the rest.
- Shipping costs, delivery times and returns drive abandonment, and they're also data Google and AI agents read. Publish them clearly in markup and feeds.
A home-decor store in Pune adds 300 new SKUs before Diwali. Six weeks later, Search Console shows 40,000 new "Discovered, currently not indexed" URLs, most of them filter combinations like ?colour=beige&size=queen&sort=price. The new products? Barely crawled. That's an ecommerce SEO strategy problem, not a content problem, and it's the kind we see in audits every month.
Online store SEO in 2027 has three jobs: rank category and product pages in classic search, feed clean product data to Google's Shopping surfaces, and make sure AI shopping assistants can understand and recommend what you sell.
Short answer: E-commerce SEO means making your categories, products and product data easy to find, crawl and trust. Build category pages around real search demand, write original product pages with Product schema, control faceted URLs, submit a complete Merchant Center feed for free listings, and publish clear price, shipping and returns data that AI shopping tools can read.
Why does an ecommerce SEO strategy differ from other SEO?
Scale changes everything. A services firm might have 40 pages; a mid-size fashion store has 4,000 products, each in five sizes and six colours, sorted and filtered a dozen ways. One bad canonical rule can deindex a whole collection.
You're also competing with marketplaces. Amazon, Flipkart, Myntra, Noon and eBay have huge authority and review counts. You won't outrank Amazon for "wireless earbuds". You can outrank it for "sweat-proof earbuds for cricket" if your category page is genuinely the best answer.
Trust matters too: pages showing price, delivery, returns and real reviews carry weight with people and Google. Google's product structured data documentation now covers return policies and loyalty programmes at the organisation level, which tells you where it's heading.
Which ecommerce SEO trends matter heading into 2027?
AI shopping has moved from answers to checkout
In November 2025 Google added agentic checkout in the US: shoppers set a price alert and, with permission, Google can buy the item on an eligible merchant's site using Google Pay. In January 2026 it announced the Universal Commerce Protocol, an open standard for AI agents to transact with retailers inside AI Mode and Gemini, with Shopify, Walmart, Target, Etsy and Wayfair among the partners named. Google says its Shopping Graph holds more than 50 billion product listings, 2 billion of them refreshed every hour.
ChatGPT went the other way
OpenAI launched in-chat Instant Checkout in 2025, then in March 2026 said it would let merchants use their own checkouts and focus on product discovery, as Retail Gazette reported. Discovery still needs data: OpenAI publishes a product feed specification so merchants can make products discoverable inside ChatGPT. The constant through all this churn is clean product data.
AI shoppers arrive ready to buy
How AI-referred shoppers behaved on US retail sites, holiday 2025
Visitors from generative AI tools compared with all other traffic sources (Nov–Dec 2025)
Source: Adobe Digital Insights, AI-driven traffic surges across industries, January 2026. US retail data; AI is still a small share of total traffic.
Adobe's holiday 2025 report also found 56.4% of US online transactions happened on a smartphone. Design product pages for phones first, and remember free listings can appear in Google Lens, which matters for fashion and furniture.
How do you build an ecommerce keyword strategy?
Map keywords to page types, not to individual pages. Broad and attribute-led terms ("men's linen shirts", "linen shirts for summer") belong on categories and subcategories. Model-level and spec-led terms belong on products. "Best" and "vs" queries belong on curated collections or buying guides.
The richest source is your own data. Site search logs tell you exactly what shoppers type on your store. Search Console's Performance report, filtered to URLs containing /collections/ or /category/, shows which categories already get impressions without clicks.
| Keyword / query type | Search intent | Page to build | Priority |
|---|---|---|---|
| "kurta sets for women", "men's running shoes" | Commercial, browsing | Main category page with intro copy and filters | High |
| "running shoes for flat feet", "cotton kurta sets for office" | Commercial, attribute-led | Indexable subcategory or curated filter page | High |
| "{brand} {model} price", "{model} buy online" | Transactional | Product page with Product and Offer schema | High |
| "best air fryer under ₹5,000", "best standing desk under $300" | Commercial investigation | Curated collection or buying guide | High |
| "{product A} vs {product B}" | Comparison | Comparison page linking both products | Medium |
| "{product} size guide", "{brand} size chart" | Pre-purchase support | Size guide linked from every relevant product | Medium |
| "{product} in stock near me" | Local, immediate | Store pages + local inventory in Merchant Center | Medium |
| "{brand} return policy", "{brand} delivery time Dubai" | Navigational, trust | Shipping and returns pages + policy markup | Medium |
| "how to clean suede shoes" | Informational | Care guide linking to relevant products | Low |
| "{brand} discount code" | Navigational | Offers page you control | Low |
How should you structure category and product pages?
Category pages
Keep the hierarchy shallow: home, category, subcategory, product. Each category page needs a unique title (not just "Shoes | Brand"), a short intro above the grid that answers what a shopper needs to choose, and a longer section below covering fit, materials or buying advice. Add BreadcrumbList markup. On Shopify, watch for products reachable at both /products/ and /collections/x/products/; most themes canonicalise correctly, but check yours.
Faceted navigation
Google's guidance on faceted navigation is blunt: filter URLs can create near-infinite URL spaces, and its first recommendation is to block them in robots.txt if you don't need them indexed. Pick the handful of filter combinations with genuine search demand ("linen shirts" plus "white"), give them static, crawlable URLs with their own titles, and keep everything else (sort orders, price sliders, multi-select combinations) out of the crawl.
Product page SEO
Manufacturer descriptions copied by fifty resellers give Google no reason to pick you. Write original copy covering who the product is for, materials, dimensions, care and what's in the box. Add your own photos and real reviews. For variants, decide whether colours deserve separate URLs (they often do in fashion, where people search "black Chelsea boots") or one URL with selectable options.
Mark up products with Product and Offer (price, priceCurrency, availability), plus aggregateRating and review where you have genuine reviews, and shippingDetails and hasMerchantReturnPolicy for merchant listing eligibility. Keep out-of-stock pages live with OutOfStock availability if the product returns; redirect only when it's permanently gone.
Picture a Jaipur block-print brand on Shopify with 1,200 products and 18 collections. Filters for colour, size, fabric and price produce roughly 9,000 crawlable URLs, and only about 1,500 product and collection pages actually deserve indexing. Blocking sort and price parameters, creating 25 curated filter pages for real queries ("indigo cotton bedsheets king size") and rewriting the top 100 product descriptions is a sensible first quarter. These numbers are illustrative, not a client result.
How do Google Merchant Center and marketplaces fit into SEO?
For an online store, Merchant Center plays the role a Google Business Profile plays for a restaurant. Free listings can show your products on Google Search, the Shopping tab, Images, Lens, YouTube, Maps and Gemini at no cost. You need a product ID, title, link, image, price, description, availability, brand and GTIN or MPN where they exist, plus colour, size, age group and gender for apparel.
Google says it wants both: structured data on the page and a Merchant Center feed "maximizes your eligibility" and helps it verify your data. Keep feed titles descriptive, prices identical to the page, and fix disapprovals weekly. If you run Shopping or Performance Max campaigns, the same feed powers them, so our Google Ads team and SEO team should share one feed strategy.
Marketplaces are both rival and channel. Selling on Amazon, Flipkart or Noon builds reviews and brand searches, but every brand search that lands on Amazon is a customer you rent. Keep your best bundles and loyalty perks on your own site.
Why US shoppers abandon checkout
Share of shoppers citing each reason, excluding those "just browsing".
Source: Baymard Institute, Cart Abandonment Rate statistics, updated September 2025. Orange bars: issues you can address before checkout with clear shipping, delivery and returns data.
Baymard puts the average documented cart abandonment rate at 70.22%. Several of the biggest reasons are information problems. Show delivery estimates by pincode or postcode on the product page, state the returns window in plain words, and mirror both in your feed and markup so shoppers see them in Google before they even click.
What content strategy works for online stores?
Ecommerce content should sell, help or reassure. Build clusters around each major category: buying guides ("how to choose a mattress firmness"), comparisons, care guides and fit guides, each linking down to the category and products. Skip trend listicles that bring traffic that doesn't buy.
E-E-A-T for a store looks like named experts (the in-house gemologist, the physio who tested the office chairs), original photos and testing notes, clear company details, and honest reviews including negative ones. Our guide to E-E-A-T covers the principle, and our older post on advanced ecommerce SEO tips goes deeper on tactics.
How do ecommerce sites earn links?
Product pages rarely attract links on their own. Categories and guides do, and internal links pass that value down. Tactics that work for stores: original data (a sizing survey, a "most-returned products" analysis from your own orders), product loans to reviewers and YouTube creators, gift guide inclusion pitched to editors six to eight weeks before Diwali, Black Friday or Christmas, and supplier or brand-stockist pages. Our link building team runs this kind of digital PR for retail brands.
How do you get products recommended in AI search?
AI assistants answer shopping questions by combining product data (feeds, structured data, the Shopping Graph) with what the web says about products (reviews, editorial roundups, forums). You need both. Keep feeds complete and accurate, because an AI agent won't recommend an item it thinks is out of stock or mispriced. Write product copy that states use cases plainly ("suitable for wide feet", "fits a 15-inch laptop"), since AI tools match those phrases to conversational queries.
Then earn mentions in the sources assistants trust: review sites, comparison articles, Reddit threads and YouTube reviews. Our playbook on ranking in AI search covers the method, and our GEO service handles it for larger catalogues.
Which ecommerce SEO KPIs should you track?
Track organic revenue, not just sessions. In GA4, set up the standard ecommerce events (view_item, add_to_cart, begin_checkout, purchase) and report revenue by landing page type: category, product, guide. Build a channel group for AI referrers (chatgpt.com, perplexity.ai, gemini.google.com, copilot.microsoft.com) so you can watch that line grow.
In Search Console, check the Shopping reports (Merchant listings and Product snippets) for markup errors, and the Pages report for indexing problems on key templates. In Merchant Center, watch disapprovals and clicks from free listings.
What are the most common ecommerce SEO mistakes?
- Letting every filter combination get crawled. Large stores burn crawl capacity on URLs nobody searches.
- Manufacturer copy on every product. Duplicate descriptions give Google no reason to rank you.
- Feed and page mismatches. A different price in Merchant Center and on the page triggers disapprovals.
- Deleting seasonal pages. Keep "Diwali gifts" or "Black Friday deals" at a permanent URL and update it yearly.
- Hiding shipping and returns. Shoppers and AI agents both look for them before buying.
What does a 90-day ecommerce SEO roadmap look like?
- Crawl and index audit
- Faceted URL rules in robots.txt
- Canonicals and variant handling
- Merchant Center diagnostics
- Keyword-to-page map
- Rewrite top 20 category pages
- Curated filter pages for real queries
- Breadcrumb and Product schema
- Rewrite top 100 product pages
- Shipping and returns markup
- Feed title optimisation
- GA4 revenue by page type
- Buying guides per category
- Seasonal gift guide PR
- AI answer testing monthly
- Quarterly technical audit
Need hands on this? Our ecommerce SEO services cover audits, content and feeds, and our Shopify development team can fix theme-level issues.
What should you do this week?
- Open the Pages reportIn Search Console, check how many "Crawled" or "Discovered, currently not indexed" URLs contain filter parameters.
- Pick your top five categoriesCompare each with the page ranking first and note what it covers that yours doesn't.
- Check Merchant CenterFix disapproved products and confirm free listings are enabled.
- Find a duplicate descriptionPaste a sentence from a bestseller into Google in quotes. If fifty sites share it, rewrite it.
- Ask an AI assistantAsk ChatGPT for the best product in your category for a specific need. See who's recommended and why.