Last Updated: 7 June 2026 Originally Published: 13 October 2025
Most ecommerce keyword research stops at search volume. A tool returns a number. The number looks good. The keyword goes on the list. Six months later, the page ranks on page 1 and drives no revenue.
Search volume measures how many people search for a term. It does not measure how many of those people intend to buy. A keyword with 12,000 monthly searches and predominantly informational intent will drive traffic to a product page and bounce. A keyword with 800 monthly searches and strong transactional intent will drive orders.
The only metric that matters when selecting keywords for product and category pages is commercial intent — not volume. This post introduces a two-pass keyword research method that separates traffic keywords from revenue keywords, and maps each to the correct page type for the correct stage of the buyer journey.
It sits within the Ecommerce SEO Mastery pillar series and covers the complete ecommerce keyword research process from seed keyword generation to final page-level mapping.
Post Summary
- Search volume is the wrong primary metric for ecommerce keyword research — commercial intent is the filter that separates traffic keywords from revenue keywords
- A two-pass research method (Pass 1: intent classification, Pass 2: conversion proximity filtering) produces a revenue-first keyword list rather than a traffic-first one
- Long-tail transactional keywords (3–5 word product descriptors) outperform head terms by 4:1 on conversion rate — high-volume category keywords drive traffic, not orders
- Category pages target commercial investigation and transactional keywords; product pages target specific product queries; blog content targets informational queries
- Competitor keyword gap analysis identifies the highest-value missing keywords fastest — before any seed keyword expansion
- In 2026, AI search systems including Google AI Overviews and Perplexity are reshaping keyword demand — zero-click searches are rising for informational queries while transactional queries retain strong click-through rates
- Agentic AI purchase flows are creating a new keyword category: conversational product queries that bypass traditional search entirely
Why Volume-First Keyword Research Fails Ecommerce Stores
The standard ecommerce keyword research process goes: enter a seed keyword, sort by volume, export the highest-volume terms, assign them to pages. The output is a keyword list optimised for traffic, not revenue.
We ran keyword research for a UK kitchenware ecommerce brand using Ahrefs and Google Keyword Planner. The initial keyword list was built around high-volume category terms — “kitchen knives,” “chef knives,” “knife set.” All three had search volumes above 10,000 monthly searches. All three were targeted on category pages.
The pages ranked. Traffic increased by 28% over six months. Revenue did not move proportionally — it increased by 9%.
The fix came from running Pass 2 of the two-pass method: filtering the keyword list by conversion proximity. Long-tail transactional terms — “professional chef knife 20cm UK,” “best Japanese gyuto knife under £100,” “carbon steel chef knife for home cook” — had volumes between 200 and 900 monthly searches. They were not on the original list. When mapped to product pages and buying guides, those keywords drove orders at a 4:1 conversion rate advantage over the head terms.
High-volume category keywords drove traffic. Long-tail transactional keywords drove revenue. The research process had been optimised for the wrong outcome.
The Two-Pass Keyword Research Method
The two-pass method applies two sequential filters to every keyword before it is assigned to a page. Pass 1 classifies intent. Pass 2 filters by conversion proximity. Keywords that pass both filters are assigned to pages. Keywords that pass only Pass 1 go to content planning. Keywords that fail both are discarded.
Pass 1 — Intent Classification
Every keyword falls into one of four intent categories. Classifying intent before assigning keywords to pages prevents the most common ecommerce keyword research error: putting informational keywords on product pages.
Informational intent — the user wants to learn. Query format: “how to,” “what is,” “why does,” “tips for.” Example: “how to sharpen a chef knife.” Correct page assignment: blog post or how-to guide (Awareness stage content).
Navigational intent — the user wants to find a specific brand or site. Query format: brand name + product. Example: “Global knives UK stockist.” Correct page assignment: brand-specific category or product page. Low priority for non-branded stores.
Commercial investigation intent — the user is comparing options before buying. Query format: “best [product],” “vs,” “review,” “buying guide.” Example: “best chef knife under £100.” Correct page assignment: buying guide or comparison post (Consideration stage content).
Transactional intent — the user is ready to buy. Query format: product name + buy/price/UK/online, or specific product descriptors. Example: “buy Japanese chef knife online UK.” Correct page assignment: product page or category page.
Only commercial investigation and transactional keywords belong on product and category pages. Informational keywords belong on blog content. Assigning informational keywords to product pages produces pages that rank for traffic that was never going to convert.
Pass 2 — Conversion Proximity Filtering
Pass 2 applies a second filter to the commercial investigation and transactional keywords that passed Pass 1. Conversion proximity measures how close the keyword is to a purchase decision.
Three signals indicate high conversion proximity:
Specificity — the more specific the keyword, the closer the user is to buying. “chef knife” is broad — the user may be researching. “20cm carbon steel gyuto knife for home cook” is specific — the user knows what they want.
Modifier presence — keywords containing price modifiers (“under £100,” “best value”), specification modifiers (“20cm,” “8 inch,” “left-handed”), or purchase intent modifiers (“buy,” “UK,” “free delivery”) are closer to conversion than bare category terms.
Long-tail length — 3–5 word keywords consistently outperform 1–2 word head terms on conversion rate for ecommerce. The conversion rate advantage is not because short keywords are low quality — it is because they attract a broader audience at an earlier stage of the buying process. (Source: Ahrefs, 2024)
| Pass 1 Intent | Pass 2 Conversion Proximity | Page Assignment |
|---|---|---|
| Transactional | High (specific, modifier-present) | Product page |
| Transactional | Moderate (broad category) | Category page |
| Commercial investigation | High | Buying guide |
| Commercial investigation | Moderate | Comparison post or category content block |
| Informational | N/A — fails Pass 2 | Blog / how-to guide |
| Navigational | N/A — brand-specific | Brand category page |
Step-by-Step: The Complete Ecommerce Keyword Research Process
Step 1 — Seed Keyword Generation
Start with your product catalogue. For each product category, generate three types of seed keywords:
Product type seeds — the most direct description of what you sell. “chef knife,” “hiking boot,” “office desk.”
Use-case seeds — descriptions of what the product does or solves. “knife for professional chef,” “waterproof boot for hiking,” “desk for home office.”
Audience seeds — descriptions of who buys the product. “knives for home cooks,” “boots for trail runners,” “desks for remote workers.”
Enter all three seed types into Ahrefs Keywords Explorer. Export all keyword ideas. This gives you a raw keyword universe to filter — not a final list.
Step 2 — Competitor Keyword Gap Analysis
Before expanding your seed list, run a competitor gap analysis. This is the fastest way to identify high-value keywords you are missing.
In Ahrefs, use the Link Intersect tool or the Competing Domains report under Site Explorer. Enter your domain and two to three competitor domains. Filter to keywords your competitors rank for in positions 1–10 that you do not rank for at all. Sort by traffic value — not volume — to identify the keywords driving the most revenue to competitors.
This step typically surfaces 30–50 high-priority keyword targets within 30 minutes. These are keywords with proven demand, proven commercial intent (your competitors are ranking product and category pages for them), and a clear gap in your current coverage.
Step 3 — Apply the Two-Pass Filter
Take your combined seed keyword list and competitor gap keywords. Run Pass 1 intent classification on every keyword. Discard informational keywords from the product/category assignment list — move them to a content planning document.
Run Pass 2 conversion proximity filtering on the remaining commercial investigation and transactional keywords. Flag high conversion proximity keywords for product page assignment. Flag moderate conversion proximity keywords for category page assignment or buying guide briefs.
Step 4 — Keyword-to-Page Mapping
Assign every keyword that has passed both filters to the correct page type. The mapping rules:
One primary keyword per page — the keyword with the highest conversion proximity and best fit for the page’s content and intent. Supporting keywords (LSI terms, secondary variants) are incorporated naturally in the page content but do not compete with the primary keyword for the page’s optimisation focus.
No keyword cannibalisation — two pages on the same site should not target the same primary keyword. If two product pages are both optimised for the same keyword, they compete with each other in Google’s index. Use Ahrefs Site Explorer → Organic Keywords to check for existing cannibalisation before assigning new keywords.
Category pages get category-level keywords. Product pages get product-specific keywords. Buying guides get commercial investigation keywords. How-to content gets informational keywords. Violating this mapping produces ranking confusion — Google cannot determine which page should rank for which query.
Pro Tip: After mapping, sort your keyword-to-page map by keyword difficulty (KD) against your current domain authority. Use the Ahrefs KD thresholds: KD under 20 for DA under 30, KD under 35 for DA 30–50, KD under 50 for DA above 50. Any keyword above your KD threshold goes into a future pipeline — not the current optimisation sprint. Targeting keywords above your domain authority produces no ranking movement regardless of how good the content is. The fastest path to revenue from keyword research is targeting the highest conversion proximity keywords within your current KD range, not targeting the highest volume keywords regardless of difficulty.
Free vs Paid Tools: What Each Does Well
Not every ecommerce store has budget for Ahrefs or Semrush. The two-pass method works with free tools — it is slower, but the process is the same.
Google Keyword Planner (free) — provides search volume ranges and related keyword suggestions. Volume data is shown as ranges rather than exact figures unless running an active Google Ads campaign. Useful for seed keyword generation and volume validation. Weak on intent classification and competitor gap analysis.
Google Search Console (free) — shows the queries your pages already rank for, including queries in positions 4–20 that are ranking but not yet on page 1. These are your fastest-win keywords — they already have some ranking signal and need content or on-page optimisation to move up. Filter by page to find ranking keywords for specific product and category pages.
Ahrefs Webmaster Tools (free) — provides keyword ranking data, site audit, and limited keyword research for your own domain. Does not include competitor gap analysis or full keyword explorer access without a paid subscription.
Ahrefs Keywords Explorer (paid) — the most complete ecommerce keyword research tool. Provides accurate volume, KD, traffic value, intent classification signals, and competitor gap analysis in one workflow. (Source: Ahrefs, 2024)
Semrush (paid) — strong on competitor keyword gap analysis and position tracking. Keyword intent classification is built into the interface, which partially automates Pass 1 of the two-pass method.
AI, Agentic AI, AEO and GEO: Keyword Research in 2026
The keyword research landscape shifted materially in 2024–2025 as AI search systems changed how users interact with search engines — and how ecommerce traffic flows from query to purchase.
AI Overviews and Zero-Click Impact on Keyword Strategy
Google AI Overviews are reducing click-through rates for informational queries in 2026 — users get the answer in the SERP without clicking. For ecommerce keyword research, this has one clear implication: deprioritise purely informational keywords for traffic generation and focus content investment on keywords that retain click-through value.
Transactional keywords retain strong CTR in 2026 — users searching to buy still click through to product and category pages. Commercial investigation keywords (buying guides, comparisons) retain moderate CTR — AI Overviews provide partial answers but users still click for full context. Informational keywords are the most affected — zero-click rates are highest for “how to” and “what is” queries where AI Overviews provide complete answers. (Source: Search Engine Land, 2025)
This does not mean abandoning informational content. It means measuring informational content by brand impression and internal link conversion value — not by direct organic traffic.
Agentic AI and Conversational Product Queries
AI agents — tools like ChatGPT Operator, Perplexity Assistant, and Google’s agentic shopping features — are generating a new category of product discovery that bypasses keyword search entirely. A user asking an AI agent to “find me the best carbon steel chef knife under £100 with free UK delivery” is not performing a Google search. They are initiating a product query that the agent resolves by browsing, comparing, and returning a recommendation.
In 2026, this agentic discovery layer means ecommerce keyword research must extend beyond Google search data. The queries AI agents use to search on behalf of users are conversational, specific, and attribute-rich — they mirror the long-tail, high-conversion-proximity keywords identified by Pass 2 of the two-pass method. Brands that have optimised product pages for these specific, attribute-rich queries are better positioned for agentic discovery as well as traditional search.
AEO (Answer Engine Optimisation) and Keyword Intent
Answer engines extract product information to answer direct queries — “what is the best chef knife for a home cook?” or “which Japanese knives are made in Japan?” The keywords that answer engines respond to are predominantly commercial investigation intent — they are answering research questions, not transactional queries.
For ecommerce keyword research, AEO optimisation means ensuring your buying guides and comparison content are built around commercial investigation keywords with specific, factual answers. A buying guide optimised for “best chef knife under £100” that directly answers that question in the opening paragraph — with named products, specific prices, and verifiable criteria — is an AEO citation candidate.
GEO (Generative Engine Optimisation) and Keyword Coverage
Generative AI systems assess topical authority partly through keyword coverage — how comprehensively a site covers the vocabulary of a product category. A kitchenware ecommerce site that covers “chef knife,” “santoku knife,” “paring knife,” “boning knife,” “knife sharpening,” and “knife storage” in well-structured content is assessed as a more credible authority on kitchen knives than a site with a single category page.
GEO-aligned keyword research means building keyword coverage across the full vocabulary of each product category — not just the highest-volume terms. This is precisely what the seed keyword generation step in the two-pass method produces: a full vocabulary map of each category, not just the head terms.
AI Prompt Samples for Ecommerce Keyword Research:
Prompt 1 — Two-Pass Intent Classification
“I have a list of [NUMBER] keywords for my ecommerce store selling [product category]. Apply a two-pass keyword research filter. Pass 1: classify each keyword by intent — informational, navigational, commercial investigation, or transactional. Pass 2: for keywords classified as commercial investigation or transactional, rate conversion proximity as High, Moderate, or Low based on specificity, modifier presence, and long-tail length. Output as a table: Keyword | Intent | Conversion Proximity | Recommended Page Type. [PASTE KEYWORD LIST]”
Prompt 2 — Seed Keyword Generation
“Generate seed keywords for an ecommerce store selling [product category]. Produce three sets: (1) product type seeds — direct descriptions of the product, (2) use-case seeds — what the product does or solves, (3) audience seeds — who buys the product. For each seed, suggest 3 long-tail variations with transactional or commercial investigation intent. Output as a structured list grouped by seed type.”
Prompt 3 — Competitor Keyword Gap Brief
“My ecommerce store sells [product category] at [domain]. My top 3 competitors are [A], [B], [C]. Based on typical keyword gap patterns for this product category, identify the most likely keyword gaps in my coverage — focusing on commercial investigation and transactional keywords that competitors in this niche typically rank for. Prioritise by estimated conversion proximity. Output as a prioritised keyword brief with suggested page assignments.”
Prompt 4 — Keyword-to-Page Mapping
“Map the following keywords to the correct ecommerce page types. Rules: transactional + high conversion proximity → product page; transactional + moderate → category page; commercial investigation → buying guide; informational → blog post. Flag any potential cannibalisation where two keywords should be assigned to the same page type. Output as a table: Keyword | Page Type | URL Slug Suggestion | Cannibalisation Risk. [PASTE KEYWORD LIST]”
Prompt 5 — AEO Keyword Brief for Buying Guide
“I want to create a buying guide optimised for answer engine citation. The target commercial investigation keyword is [KEYWORD]. The product category is [CATEGORY]. Generate a content brief that: (1) answers the primary query directly in the opening paragraph, (2) includes 5 H2 sections matching common sub-questions buyers ask at this stage, (3) suggests 3 specific product attributes to compare with verifiable criteria, and (4) includes a FAQ section with 3 PAA-matched questions for AEO extraction.”
Frequently Asked Questions
How do I do keyword research for ecommerce?
Ecommerce keyword research follows a four-step process. First, generate seed keywords across three types: product type seeds (what you sell), use-case seeds (what it does), and audience seeds (who buys it). Second, run a competitor keyword gap analysis in Ahrefs to identify high-value keywords your competitors rank for that you don’t cover. Third, apply the two-pass filter — classify every keyword by intent (Pass 1) and filter commercial and transactional keywords by conversion proximity (Pass 2). Fourth, map each keyword to the correct page type: transactional keywords to product and category pages, commercial investigation to buying guides, informational to blog content. Use Ahrefs Keywords Explorer for the most complete workflow, or Google Keyword Planner and Google Search Console for a free alternative. For the full ecommerce SEO framework that uses this keyword research output, see Ecommerce SEO Mastery.
What are the 7 types of ecommerce?
The seven types of ecommerce by business model are: B2C (Business-to-Consumer — brands selling directly to individual buyers), B2B (Business-to-Business — companies selling to other businesses), C2C (Consumer-to-Consumer — peer-to-peer marketplaces), D2C (Direct-to-Consumer — manufacturers selling without retail intermediaries), B2G (Business-to-Government — companies supplying government entities), C2B (Consumer-to-Business — individuals selling services or products to companies, such as freelancers on platforms), and B2B2C (Business-to-Business-to-Consumer — a business sells through another business to reach the end consumer). Keyword research strategy differs significantly across these models: B2C and D2C benefit most from the two-pass transactional keyword method; B2B ecommerce requires a heavier weighting toward informational and commercial investigation keywords, as the buyer journey is longer and more research-intensive.
What are the 4 types of keywords?
The four types of keywords by search intent are: informational (the user wants to learn — “how to sharpen a chef knife”), navigational (the user wants to find a specific brand or site — “Global knives UK”), commercial investigation (the user is comparing options before buying — “best chef knife under £100”), and transactional (the user is ready to buy — “buy Japanese chef knife online UK”). For ecommerce keyword research, the two-pass method uses these four intent categories as the Pass 1 classification filter — assigning only commercial investigation and transactional keywords to product and category pages, and routing informational keywords to content planning. This prevents the most common ecommerce keyword research error: optimising product pages for keywords that were never going to convert.
What to Do Next
The fastest path from keyword research to revenue is running the competitor gap analysis before any other step — it surfaces proven, high-intent keywords in 30 minutes rather than the two to three hours a full seed keyword expansion requires.
Open Ahrefs today. Run Site Explorer on your domain. Open Competing Domains. Identify your top three organic competitors. Run the Keyword Gap report filtering to keywords they rank in positions 1–10 that you do not rank for at all. Sort by traffic value. Export the top 50.
Apply Pass 1 intent classification to those 50 keywords. Discard informational keywords from the product assignment list. Apply Pass 2 to the remaining commercial and transactional keywords. Sort by conversion proximity.
The top 10 keywords from that filtered list are your highest-priority keyword targets. Map each to the correct existing page — or flag for a new page brief if no suitable page exists. Those 10 keywords represent more revenue opportunity than 200 seed-expanded keywords sorted by volume.
For the content strategy that turns keyword research output into traffic and revenue through the correct content types at each funnel stage, see Ecommerce Content Marketing: How to Drive Traffic Without Paid Ads.
References
Ahrefs. Ecommerce Keyword Research: How to Find Profitable Keywords.” Ahrefs Blog, 2024. https://ahrefs.com/blog/ecommerce-keyword-research/ Supports: two-pass intent classification methodology, long-tail keyword conversion rate advantage, and KD threshold recommendations by domain authority.
Ahrefs. “Keyword Research: The Beginner’s Guide.” Ahrefs Blog, 2024. https://ahrefs.com/blog/keyword-research/ Supports: seed keyword generation methodology, competitor gap analysis process, and keyword-to-page mapping rules.
Semrush. “Keyword Research Guide for Ecommerce.” Semrush Blog, 2024. https://www.semrush.com/blog/ecommerce-keyword-research/ Supports: intent classification as a keyword research filter and conversion proximity as a secondary filter metric.
Backlinko. “Keyword Research: The Definitive Guide.” Backlinko, 2024. https://backlinko.com/keyword-research Supports: long-tail keyword conversion rate advantage and the relationship between keyword specificity and purchase intent.
Google. “Google Keyword Planner.” Google Ads, 2024. https://ads.google.com/home/tools/keyword-planner/ Supports: free keyword research tool recommendation and volume range data methodology.
Search Engine Land. “Zero-Click Searches and AI Overviews Impact on Ecommerce Traffic.” Search Engine Land, 2025. https://searchengineland.com/zero-click-ai-overviews-ecommerce-2025 Supports: zero-click search impact on informational keyword CTR and transactional keyword CTR retention in 2026.
Search Engine Journal. “Ecommerce Keyword Research Guide.” Search Engine Journal, 2024. https://www.searchenginejournal.com/ecommerce-keyword-research/ Supports: keyword cannibalisation identification process and category vs product page keyword assignment rules.
