Ecommerce keyword research: finding terms that actually convert

Most keyword research advice is written for content sites, where the goal is traffic. Ecommerce keyword research has a different goal: traffic that’s close enough to a purchase decision that it’s worth building or optimizing a page for. Those aren’t the same exercise, and treating them the same is why a lot of ecommerce keyword lists end up full of terms that drive visits and nothing else.

Start with intent buckets, not a keyword list #

A flat spreadsheet of keywords sorted by search volume tells you what people type, not what they need. The more useful starting point is intent, and the split is more lopsided than most people assume: research consistently shows roughly 70% of search queries in a given space are informational, 22% commercial, 7% navigational, and around 1% directly transactional — though this varies significantly by industry and how competitive a niche’s transactional terms already are.

That 70% isn’t wasted effort to ignore — it’s where blog and guide content earns its keep, feeding traffic toward commercial and transactional pages rather than competing with them directly. The mistake is assuming a keyword list is “done” once you’ve found the 1% of terms with a shopping cart icon next to them in your keyword tool.

The canonical buyer’s guide model #

For each major product category, the useful structure is one canonical guide — the definitive resource for that category — supported by product pages, comparison content, and how-to posts that all link back up to it, rather than a scattered set of blog posts each competing for pieces of the same territory.

Not this: five separate blog posts about hiking boots, each targeting a slightly different version of the same keyword, each mediocre because the topic got split five ways.

Do this: one comprehensive hiking boots buying guide, linked to from every relevant product and category page, with narrower supporting content (a boots-vs-trail-shoes comparison, a boot-care how-to) each linking back up to it. The guide earns the authority; the supporting content earns its own smaller wins and reinforces the guide.

Building an actual keyword-to-page map #

The output of ecommerce keyword research shouldn’t be a list — it should be a map: one row per URL (existing or planned), its primary target keyword, its intent bucket, and which other pages link to it. This is the single most effective way to prevent the cannibalization problem where a blog post and a category page both chase the same commercial term and Google can’t decide which one to rank.

A simple version:

Every new piece of content or page gets checked against this map before it’s built — if a proposed keyword already has a primary page assigned, that’s a signal to either strengthen the existing page or genuinely differentiate the new one, not launch a second page to compete with it.

Where the actual research happens #

Standard keyword tools (Ahrefs, SEMrush, Google’s own Keyword Planner) work fine for volume and difficulty data — the ecommerce-specific part isn’t the tool, it’s the discipline of sorting everything into intent buckets and a page map before writing anything, rather than working straight off a raw exported list.

Two sources worth checking that generic keyword tools miss entirely: your own site search data (what people type into your on-site search bar is often more honest than what they type into Google, because they’re already committed to your store) and your support inbox (recurring pre-sale questions are informational-intent keywords you haven’t targeted yet).

A proper keyword-to-page map is one of the first deliverables in any ecommerce SEO engagement here — everything else, content and technical work alike, gets built against it.

Related reading: Blog content vs. category pages · Ecommerce SEO services