E-commerce SEO
Most product pages fail on a sentence the retailer never wrote, copied from the manufacturer along with the photograph. Ivan Sp works with e-commerce catalogues on technical fixes, content gap analysis, and on-page corrections at product and category scale. One e-commerce client doubled organic traffic within 90 days through this combination, with no paid promotion behind the number.
Where the Growth Actually Came From
The 90-day result came from three combined areas, not one single change. Technical fixes cleared indexation and crawl faults. Content gap work identified what the catalogue was missing against real demand. On-page corrections brought existing pages up to a real standard. No single lever produced the result alone.
Manufacturer Copy Is the Common Starting Point
Most e-commerce sites arrive carrying manufacturer product descriptions, identical to every competitor selling the same item. Replacing that copy at catalogue scale is usually the single highest-leverage content fix available to a store still running on shared manufacturer text. That work is covered under e-commerce product content at scale.
Category and Collection Structure
Category pages carry their own technical requirements: correct canonicalization across faceted navigation, schema reflecting the actual category hierarchy, and internal linking connecting related categories. A catalogue with strong product pages and a weak category structure still underperforms its potential.
What This Work Does Not Include
Page speed engineering, checkout customization, and platform migration all sit outside this work, the same boundary that applies across every industry this practice serves. A technical audit finding a genuine hosting or platform limitation reports it honestly, rather than promising a fix that is not actually available.
Getting Started
An e-commerce engagement usually starts with a technical crawl and a content gap analysis together. A catalogue’s real opportunity is rarely visible from either angle alone.
What Decides E-commerce Search Performance
- Product markup
- Product schema, carrying offers, price, availability, and any visible rating.
- The duplication problem
- Manufacturer descriptions repeated verbatim across every retailer selling the item.
- Faceted navigation
- Filter combinations that can generate more crawlable URLs than the catalogue holds products.
- Category pages
- Usually the highest-value pages, and usually the thinnest.
- Out-of-stock handling
- A decision between keeping, redirecting, or removing, with different index consequences.
- Merchant Center
- The feed behind shopping surfaces, separate from organic indexing.
- Crawl efficiency
- The constraint that decides whether new products get discovered quickly.
Where the Standard E-commerce Advice Breaks Down
How many of your product descriptions were written by the person who made the product?
- The usual advice
- Write unique descriptions for every product, since duplicate manufacturer copy is the core problem.
- What holds up in practice
- Rewriting ten thousand descriptions by hand is not a project most retailers can finish. Category pages carry more commercial intent per page, and a rules-based system handles product copy at a scale hand-writing never reaches.
Imagine a retailer commissioning unique copy for 8,000 products. The first 400 are written well, the budget runs out, and the remaining 7,600 stay as they were.
In the same period the 40 category pages, which carry the highest-volume commercial queries, received no attention at all.
I start with the category pages and move product copy to a generated system. Forty pages written properly beats four hundred out of eight thousand.
The half-finished rewrite is the worst outcome available. The budget is spent, the catalogue is inconsistent, and the pages with the most demand were never touched.
If your category pages are thinner than your product pages, your highest-intent queries are landing on the weakest content you own.
Four Things Commonly Assumed About E-commerce SEO
Which of these is currently true of your catalogue?
Ivan Sp meets these four assumptions on nearly every catalogue engagement.
| Commonly assumed | What actually holds |
|---|---|
| Every product needs a unique hand-written description. | Every product needs content that differs meaningfully. A rules-based system using real attributes achieves that at a scale hand-writing cannot. |
| More products indexed is always better. | A catalogue full of near-identical variants can dilute rather than help. Consolidating variants under one indexable product is often stronger. |
| Filters should all be crawlable so Google finds more pages. | Filter combinations commonly consume the majority of crawl requests while holding no indexable content. Most should be closed off. |
| Out-of-stock products should be deleted. | A product returning permanently should be removed or redirected. A seasonal product deleted each year loses its accumulated history every time. |
| Product schema improves rankings. | Structured data is not a ranking factor. Product markup governs eligibility for rich results, which changes click-through rather than position. |
Category pages are usually where the money is and where the content budget never reaches. That single misallocation costs more than any other decision on this list.
A Worked Example: Where the Budget Should Go
Consider a 8,000-product store with 40 categories and a fixed content budget. The allocation below is illustrative, written to show the arithmetic. No client engagement is described here.
| Allocation | Product-first plan | Category-first plan |
|---|---|---|
| Pages hand-written | 400 products | 40 categories |
| Share of catalogue covered | 5 percent | 100 percent of categories |
| Commercial query volume addressed | Long tail only | Head and mid tail |
| Remaining products | 7,600 unchanged | 8,000 through a generated system |
| Consistency across the catalogue | Mixed | Uniform |
| Budget exhausted before completion | Yes | No |
I allocate to the pages with the most demand per page first. Forty categories is a project that finishes, and 8,000 hand-written products is a project that stops.
Rewriting products before categories is repainting every bedroom while the front door stays boarded up. The rooms look better and nobody gets in.
The generated system is not a shortcut. Rules-based copy built on real attributes produces genuinely different pages, which is what the requirement actually asks for.
How an E-commerce Engagement Runs
I establish crawl efficiency before any content work, since discovery gates everything else.
- Crawl the store and count how many URLs the filters generate against the product count.
- Read the logs or Crawl Stats to see what share of requests reach filter URLs rather than products.
- Close off filter paths through robots.txt or parameter handling, keeping the useful ones crawlable.
- Audit category pages for content depth, since they usually carry the strongest commercial queries.
- Compare product descriptions against manufacturer copy to measure the real duplication level.
- Design a rules-based content system using real product attributes rather than spun synonyms.
- Validate Product schema against the Rich Results Test, and remove any rating with no visible reviews.
- Set an out-of-stock policy, and apply it consistently across seasonal and discontinued items.
- Check the Merchant Center feed against the on-page data, since disagreement causes disapprovals.
- Re-crawl after deployment and confirm crawl requests have shifted toward product URLs.
Tools and Primary Sources
The references below define the markup and the crawl behaviour. Each opens in a new tab.
- Google ecommerce SEO documentation covers product markup, pagination, faceted navigation, and out-of-stock handling.
- Product structured data documents the required and recommended fields for product markup.
- Faceted navigation guidance explains how filter URLs consume crawl capacity and how to manage them.
- Google Merchant Center documents the feed specification that shopping surfaces read.
- Screaming Frog SEO Spider counts generated URLs and exports duplicate content signals across a catalogue.
How This Engagement Is Scoped
I scope catalogue work by SKU count and by the number of filter dimensions.
- What arrives
- A crawl efficiency plan, category page content, a product content system, and validated schema.
- Typical duration
- Six to twelve weeks, depending on catalogue size and platform.
- What the client supplies
- Platform admin access, product attribute data, and Search Console access.
- Included
- Crawl audit, filter handling, category content, the generation system, schema, and verification.
- Not included
- Paid shopping campaign management, and product photography.
- What changes the estimate
- Catalogues above roughly fifty thousand SKUs, or many filter dimensions.
- How it is quoted
- After a short scoping call, once SKU count and platform are known.
Measure Your Own Duplication
I suggest a check that takes about five minutes and settles the duplication question.
- Open one of your product pages and copy a full sentence from the description.
- Search Google for that sentence inside quotation marks.
- Count how many other retailers return the identical sentence.
- Repeat on two more products from different categories.
A dozen retailers returning your exact sentence means that page has nothing of its own to offer. The photograph is usually shared too, which leaves price as the only differentiator.
What This Looks Like on a Product Page
Consider a shopper comparing the same item across four stores. Every page shows the same photograph and the same paragraph, supplied by the manufacturer.
The retailer wants to be chosen for something other than price. Being indistinguishable from three competitors is a slow way to lose margin.
I open the product feed and find nine unpublished attribute fields. The compatibility data alone gives that page something none of the other three can say.
Common Questions
How large does a catalogue need to be before this approach makes sense?
Roughly 50 products is where the effort starts paying back clearly. Below that, catalogue-wide systems cost more to build than the manual alternative.
Does this work apply to Shopify and other hosted platforms?
Yes. Implementation adapts to whichever platform the store runs on, covered under SEO by platform.
How is the 90-day result measured?
Through Google Analytics and Search Console data, comparing the 90-day period against the equivalent prior period. The comparison isolates organic search traffic specifically from paid and direct channels.
Should filter URLs be blocked in robots.txt or set to noindex?
Blocking in robots.txt saves crawl capacity and prevents Google reading any tag on the page. Noindex allows crawling and prevents indexing. The choice depends on whether crawl waste or index bloat is the actual problem.
Discuss This Work
The first call is a free 30-minute consultation with no pitch attached. You receive an honest assessment of where the site stands and what can realistically be built.