Ecommerce On-Page SEO That Actually Drives Sales

Stop guessing which page should rank. A page-by-page framework for category, product, and article SEO built to convert, not just rank.

Richard

Richard

Richard leads SEO at RankSpiral. He spent years ranking his own niche sites before helping anyone else with theirs.

November 10, 2026 · 17 min read

Ecommerce On-Page SEO That Actually Drives Sales

Why Most Ecommerce SEO Checklists Fail

Try this with any "ultimate ecommerce SEO checklist." For each item, ask who on your team actually does it. Within three lines you'll hit "fix crawl budget," "build backlinks," and "rewrite title tags," and none of them belong to the same person.

Most checklists mix on-page, technical, and off-page tasks into one long list with no owner.

The easy stuff gets done twice.

The hard stuff never gets done.

Even when tasks do get finished, they get finished page by page, as if each URL lived on its own little island.

It doesn't.

The core premise of this guide: your category pages, product pages, and articles should be optimized as one connected system, not as isolated URLs competing for attention.

So instead of another 87-point list, this guide organizes ecommerce on-page SEO by page type and search intent. Every fix ties back to three questions: can Google crawl it, will searchers click it, and does it make money?

Rankings are nice.

Revenue pays rent.

Who Should Rank: Category, Product, or Article

Which page do you want Google to show for a given keyword?

Most store owners never decide.

If you don't, Google decides for you, and it often picks the wrong one.

Before you assign keywords, split the work into two buckets. On-page SEO covers everything you can change directly in a page or its template: title tags, headings, body copy, internal linking, image alt text, and structured data fields.

Technical SEO is the plumbing underneath. That means crawl rules, rendering, Core Web Vitals, canonical logic, and pagination, which usually need a developer.

Why does the split matter?

A merchandiser can rewrite a category intro this afternoon. Changing how filter URLs get generated needs a ticket, a sprint, and probably a snack bribe.

The rule for search intent mapping fits on a sticky note.

Broad commercial terms go to category pages. Specific SKU or model queries go to product pages.

How-to and comparison queries go to articles.

Tape that to your monitor.

Seriously.

Category Pages Own Broad Intent

Someone searching "women's trail running shoes" hasn't picked a shoe yet. They want options, filters, and a feel for the price range, which is exactly what a category page delivers.

Google knows this too.

For broad commercial investigation keywords, the first page is usually full of collection pages from retailers, not individual product URLs.

Your category page SEO job is simple to describe and hard to do: be the most useful version of that grid.

Product Pages Own Specific Intent

"Brooks Cascadia 18 women's size 8" is a different animal. That searcher has transactional search intent and a credit card within arm's reach.

Product page SEO wins these queries because the page matches the search almost word for word: model, variant, price, availability, reviews.

Sending this person to a category page is like answering "where's the ketchup?" with "somewhere in this building."

When an Article Should Outrank Both

Some searchers aren't ready to buy anything yet. "How to choose trail running shoes" and "Cascadia vs Speedgoat" are research questions, and Google will usually rank a guide above a product grid.

Fighting that with a category page is a losing battle.

Instead, write helpful buying content that answers the question honestly. Then link to the category and product pages that finish the journey.

Quick check-in: can you name which page type owns each of your top 20 keywords? If not, pause here and map them.

Everything else in this article builds on that map.

Category and Product Page Templates

Templates sound boring.

They're where ecommerce SEO scales: one good category template improves 400 pages at once, while one good blog post improves... one blog post.

The challenge is that templates have to please two audiences who want opposite things.

Shoppers want products immediately.

Google wants enough unique text to understand what the page is about.

Category Pages Without Burying Products

Here's a repeatable template for collection page optimization that keeps both camps happy.

It has three zones, and each one has a single job.

Above the product grid, write a 50 to 100 word intro. Name the category in plain language, mention the main subtypes or use cases, and stop. On mobile that's three or four lines, so the first row of products stays visible.

Below the product grid, add the expanded buying guide copy: how to choose, key differences between options, sizing or compatibility notes, and common questions. This is where the page earns its topical depth, usually 300 to 600 words, without pushing a single product down the screen.

Throughout the page, link to subcategories with crawlable HTML links, not JavaScript-only buttons that crawlers may skip. Add breadcrumb navigation at the top so users and Google both understand where the page sits in your hierarchy.

Diagram, Anatomy of a Ranking Category Page. Breadcrumbs — Shows the page hierarchy; Short intro — 50 to 100 words above…

What about hiding copy in a "read more" accordion? Google indexes collapsed content under mobile-first indexing, so it's fine for user experience.

Just don't use it to hide 2,000 words of keyword soup.

Nobody's fooled, least of all your bounce rate.

For titles, the category title tag should match the broad term, like "Women's Trail Running Shoes | Free Returns | StoreName." On categories, title tag optimization comes down to clarity plus one reason to click.

Product Pages That Beat Manufacturer Copy

If you sell the same blender as 300 other retailers and paste in the same manufacturer description, why would Google rank you over Amazon?

It won't.

Duplicate product descriptions rarely trigger a penalty. They just give Google zero reason to prefer your page over a bigger site with more authority.

Buyer Questions Over Spec Sheets

Spec dumps answer questions nobody asked.

"Motor: 1,200W" means very little until you say it crushes ice for smoothies without stalling.

Build every product description around three buyer questions. The first is fit: who is this for, and will it work for my size, skill level, or setup?

The second is use case: what does it do well, and where does it struggle? The third is comparison: how does it differ from the two or three similar SKUs on your own site?

That last one is gold.

A line like "Choose the Pro if you blend daily; the Lite is fine for weekend smoothies" helps shoppers decide and gives you text no competitor has. Keep the full spec table, but move it below the human-written part.

Product reviews do similar work. Real customer language about fit and durability adds fresh, unique content to the page every month, for free.

Avoiding AI Duplicate Language

Here's the 2026 twist.

Plenty of stores "fixed" manufacturer copy by running 5,000 SKUs through an AI writer. Now their descriptions all sound like the same overcaffeinated intern wrote them.

"Elevate your kitchen experience with this versatile, sleek blender."

Multiply that across a catalog and you've swapped one kind of duplication for another. Google's guidance on scaled content cares about whether a page adds value, not how it was produced, and templated filler adds none.

If you use AI, give it real inputs: return reasons, support tickets, review themes, and your own comparison notes. Then have a human edit the top 20% of products by revenue first.

That's where unique copy pays off fastest.

Crawl Control for Variants and Filters

A store with 500 products and a handful of filters can generate hundreds of thousands of crawlable URLs. Five filters with ten options each, combined freely, gets you there faster than you'd think.

Most of those URLs are near-duplicates that waste crawl attention and split ranking signals. Crawl control means deciding, on purpose, which URLs deserve a place in Google's index.

Variants, Canonicals, and Feeds

Product variants (colors, sizes, materials) are where most stores either index too much or too little. One question decides it: does each variant have meaningful search demand of its own?

  • Give variants separate indexable URLs when they have real search volume. If people search "black leather Chelsea boots" separately from "tan leather Chelsea boots," each color deserves its own URL with a self-referencing canonical and a unique title.
  • Consolidate variants with canonical tags when demand is minimal. Sizes almost never need their own URLs, because almost nobody googles "Chelsea boots size 9.5." Point those variant URLs to the main product with a canonical tag.
  • Keep Product structured data aligned with your variant logic. Google supports ProductGroup markup with hasVariant and variesBy properties, so you can describe variants accurately whether they live on one URL or several.
  • Match your Merchant Center product feed to the page. Price, availability, and variant IDs in the feed should match the landing page exactly. Mismatches cause item disapprovals and weaken trust signals across free listings and Shopping ads.
  • Use item_group_id consistently. Variants of the same product should share one item_group_id in the feed, matching the grouping you've already set up with canonicals and schema.

Faceted Navigation Rules

Faceted navigation is great for shoppers and risky for crawlers. The rule of thumb is to index filter combinations that match how people actually search, and keep everything else out of the way.

  • Index filter pages with distinct search demand. "Red dresses" and "waterproof hiking boots" are real queries. A color-plus-category or feature-plus-category page can become an indexable landing page with its own title, H1, and short intro.
  • Noindex or block sort orders. "Sort by price low to high" changes the order, not the content. These URLs add nothing to the index and should be blocked or canonicalized to the unsorted version.
  • Keep price ranges out of the index. Nobody searches for "$47 to $83." Block price parameters in robots.txt or apply noindex.
  • Limit stacked filters. Allow indexing for one or two filters at most. "Red + size M + cotton + under $50 + on sale" is a URL only a crawler could love.
  • Know that robots.txt and noindex behave differently. Robots.txt stops crawling but can leave linked URLs in the index. Noindex only works if Google can crawl the page and see the tag. Pick one approach per URL pattern rather than stacking both.
  • Handle pagination cleanly. Paginated category pages should use self-referencing canonicals and plain crawlable links between pages. If every page canonicalizes to page one, products on deeper pages may never get discovered.

Schema, Images, and Out-of-Stock Pages

Structured data and images are some of the cheapest visibility wins in ecommerce. They feed rich results, Shopping surfaces, Google Images, and Google Lens, which Google says handles nearly 20 billion visual searches a month.

  • Product and Offer schema: mark up name, image, brand, SKU or GTIN, price, currency, and availability. That's the minimum for showing price and stock information in results.
  • Review and AggregateRating schema: mark up genuine product reviews hosted on the page. Star ratings make a listing stand out, which tends to lift click-through rate alongside a strong meta description.
  • BreadcrumbList schema: mirror your visible breadcrumb navigation so Google shows a clean category path instead of a messy URL.
  • Descriptive image filenames: "brooks-cascadia-18-womens-blue.jpg" beats "IMG_4471.jpg" every time. Pair it with image alt text that describes what's actually in the photo.
  • Consistent dimensions and compressed files: use the same aspect ratio across a category, serve WebP or AVIF, and keep most product images under roughly 200KB. Lighter images help Core Web Vitals, especially Largest Contentful Paint.

Then there's out-of-stock SEO, where good pages often get deleted for no reason.

Removing a product page throws away its rankings, backlinks, and reviews.

Here are four options, in order of preference:

1
Keep the page live with a restock date.
Set availability to "BackOrder" or "OutOfStock" in schema, show the expected date, and offer a back-in-stock email alert.
2
Recommend a close replacement.
Add a prominent "similar items in stock" block so the visitor still has a path to buy today.
3
Preserve existing reviews.
If a new model replaces the product, keep the old page live as a reference with its reviews intact, and link clearly to the successor.
4
Redirect to the category only as a last resort.
Use a 301 to the closest category only when the product is permanently gone and no close replacement exists.
Step-by-step diagram, Out-of-Stock Options in Order. 1. Restock date — Keep the page live; 2. Replacement — Suggest a close…

Proving SEO Revenue, Not Just Rankings

Ranking third for a keyword feels great until finance asks how much money it made.

Then the room gets very quiet.

Proving revenue starts with internal linking, because links are how organic visitors move from research to purchase. Think of your internal links as an intent map: articles link to category pages with contextual anchors, and categories link to their top-performing products.

A guide on choosing trail shoes should link to "women's trail running shoes" in the sentence where that recommendation comes up naturally. It shouldn't stuff three random product links into an unrelated article about marathon nutrition.

Forced links confuse readers and blur the signal you're sending Google.

Next, open Google Search Console and filter the Performance report by page. You're looking for three patterns.

Pages ranking in positions 11 to 30.

These are close enough that a better title, a stronger intro, or a few internal links can push them onto page one, where the clicks are. Backlinko's analysis of 4 million search results found fewer than 1% of searchers click anything on page two.

Key insight: Fewer than 1% of searchers click a result on page two of Google — Backlinko analysis of 4 million search results

Pages cannibalizing each other. If a category and a blog post keep swapping positions for the same query, pick one owner and adjust the other page's targeting.

Pages ranking for the wrong intent. A product page showing up for "how to clean suede boots" is collecting impressions it can't convert.

Build the article that query is asking for.

Then track what matters beyond rankings: organic product-detail views, add-to-cart rate, assisted conversions, and organic revenue per page. Most teams get stuck on organic revenue attribution, because articles rarely get last-click credit even when they start the journey.

That's where click-to-checkout tracking helps. Tools like Rankspiral automate it with tracking and Retrolink features that connect an article click to the product click and checkout that followed, so nobody has to stitch spreadsheets together at 11pm.

Takeaway: once you can show that one buying guide sent 40 visits to product pages and led to 6 orders last month, defending the content budget gets a lot easier.

Frequently Asked Questions

What is the 80/20 rule in eCommerce?

The 80/20 rule in ecommerce says roughly 80% of revenue comes from about 20% of products. For SEO, that means optimizing your best-selling product and category pages first, because improvements there move revenue fastest.

Fix those titles, descriptions, and internal links before you touch the long tail.

How much does on-page SEO cost?

On-page SEO can cost nothing if you do it in-house, or several thousand dollars a month with an agency. As of 2026, agency retainers for mid-size stores commonly run $1,500 to $5,000 per month, while freelancers often charge per page or per hour.

The biggest hidden cost is usually internal time spent rewriting copy for priority pages.

Is SEO worth it for eCommerce?

Yes, SEO is worth it for most ecommerce stores because organic traffic keeps converting without a cost per click.

Paid ads stop the moment the budget does, while a well-optimized category page can drive sales for years. Expect three to six months before results show, and treat it as a compounding investment.

What is on-page SEO?

On-page SEO is the practice of optimizing a page's content and HTML so search engines understand it and searchers want to click it.

It covers title tags, meta descriptions, headings, body copy, image alt text, internal links, and structured data. It's separate from technical SEO (site infrastructure) and off-page SEO (links and mentions from other sites).

Are category or product pages better for SEO?

Category pages are usually better for broad, high-volume keywords, while product pages win specific model and SKU searches. They serve different intents, so neither is better across the board.

The strongest stores use categories to capture discovery traffic and pass authority to products through internal links.

How do I write product page titles and meta descriptions?

Write product titles as brand + model + key attribute + store name, kept under about 60 characters. For example: "Brooks Cascadia 18 Women's Trail Shoe | StoreName."

Keep meta descriptions around 150 characters, name one concrete benefit, and add a reason to click such as free shipping or easy returns.

Start With the Page That's Already Close

You don't need to fix everything this week.

You need to fix one page.

Tonight, pull your Google Search Console data, filter for positions 11 to 20, and find a category or product page ranking there for its intended intent.

Google is already paying attention to that page.

It probably needs a sharper title, a better intro, a few contextual internal links, and maybe a description that sounds like a human wrote it.

Fix that one first.

Measure it for four weeks.

Then pick the next.

When you're choosing what to work on after that, use this rule:

If a page already gets impressions, optimize it. If it doesn't exist yet, map its intent before writing a word.

That's ecommerce on-page SEO in two sentences.

The rest is doing it page by page until the revenue numbers make the case for you.

You probably have page-two articles earning nothing right now

Connect your site and RankSpiral shows every keyword you rank 11–30 for, priced by what closing it is worth — then writes the rewrite. Your first two articles are on us, no card.

Get started for free
Richard

Richard

Richard leads SEO at RankSpiral. He spent years ranking his own niche sites before helping anyone else with theirs.

Keep reading