Why Product-Name Keywords Are Not Enough
Most ecommerce stores start and end their keyword strategy with product names: "buy [product]," "[brand] [product]," "[product] price." These keywords are high-competition (every competitor targets them), relatively low-volume (fewer people search for a specific product name than for the problem it solves), and they only capture buyers who already know what they want. They miss the 70% of the buying journey that happens before someone decides which product to purchase.
The bigger opportunity lives in the research phase. "Best running shoes for flat feet," "ceramic vs stainless steel cookware," "how to choose a backpacking tent" — these are the queries with higher volume, lower competition, and direct influence on buying decisions. A store that only targets product-name keywords is only visible to buyers at the very end of their decision. A store that targets research-phase queries is visible throughout the entire journey, building trust and authority before the buyer is ready to purchase.
This is why long-tail keywords dominate ecommerce SEO. Each one is small — 50 to 500 monthly searches — but they convert at 2-5x the rate of broad terms because the intent is specific and the competition is thin.
The Four Types of Ecommerce Keywords
Every keyword maps to one of four search intent types, and understanding this map is the foundation of ecommerce keyword research. Navigational keywords are brand-specific searches: "Nike Air Max 90," "Allbirds tree runners." These go to product pages and you rank for them by existing. Informational keywords are questions: "how to clean suede shoes," "what thread count means for sheets." These go to blog posts and guides.
Commercial investigation keywords are where the money is: "best running shoes for wide feet," "Allbirds vs On Cloud," "top mattresses for side sleepers 2026." These searchers are ready to buy but have not decided what yet. They are comparing, evaluating, and looking for authoritative recommendations. The stores that appear here influence the decision. Transactional keywords are "buy," "price," "discount," "free shipping" — the final step where someone is ready to pull out their card.
Most ecommerce stores only target navigational and transactional keywords. The opportunity is commercial investigation — those searchers have the highest lifetime value because you catch them before they have made up their mind, and the content that serves them (comparison guides, buying guides, best-of lists) builds topical authority that lifts every other page on the site.
Where to Find Keywords for Free
Google Search Console is the single most valuable free keyword research tool because it shows you queries your site already appears for — including ones you never intentionally targeted. Sort by impressions to find queries where Google already thinks your site is relevant but you do not yet have dedicated content. These are low-hanging fruit: the intent exists, Google already associates you with it, and building a page for it converts an accidental impression into a deliberate ranking.
Google autocomplete reveals what real buyers type. Open an incognito browser, type your product category plus a space, and record every suggestion. Then type your product category plus each letter of the alphabet. "Running shoes a..." gives you "running shoes arch support," "running shoes ankle pain," "running shoes asics." Each suggestion is a validated query with real search volume. Competitor blog analysis surfaces what your competitors have already identified as traffic-driving topics — if they wrote about it, there is probably demand.
People Also Ask boxes in Google results expand into cascading question chains. Click through 3-4 levels deep and you will have 20-30 related questions for any seed query. Reddit and niche forums show you the language buyers use — which is often different from the industry jargon your product pages use. Amazon search suggestions reflect buyer language specifically tuned for purchase intent. Google Search Console ties all of these together with real impression and click data.
How to Evaluate a Keyword's Value
Do not chase volume. A keyword with 200 monthly searches and clear buying intent — "best titanium cookware for induction" — is worth more than a keyword with 10,000 searches and no commercial intent — "what is titanium." The evaluation framework is four questions. First: does a buyer search this? If the searcher is researching a purchase decision, the keyword has commercial value. If they are writing a school report, it does not.
Second: can I build a page that answers it better than what currently ranks? Search the keyword and read the top 5 results. If they are thin, generic, or outdated, there is an opening. If they are comprehensive and recently updated by authoritative domains, the effort may not be worth it yet. Third: does it connect to my product catalog? A page that answers the query should naturally lead to products you sell. If the connection is forced, the conversion rate will be near zero regardless of traffic.
Fourth: can my page earn an AI search citation for this query? Search the keyword in ChatGPT, Perplexity, and Google AI Overviews. If AI surfaces answer it, that is a citation opportunity. If no AI surface touches it, you are competing only in traditional search — still valuable, but with a lower ceiling. Keywords that trigger AI answers represent dual-channel opportunities: organic traffic plus AI citation exposure.
Building a Keyword Map for Your Store
Group your keywords into topic clusters. Each cluster has a pillar topic (broad, competitive, high-volume: "running shoes") and supporting pages (specific, long-tail, lower competition: "best running shoes for flat feet," "running shoes vs walking shoes," "how to size running shoes"). The supporting pages build authority for the pillar. The pillar links to and from each supporting page. Together, they signal to Google that your site covers this topic comprehensively.
Map each cluster to content types. Informational keywords become articles and guides. Commercial investigation keywords become comparison pages, buying guides, and interactive tools. Category keywords become collection landing pages. Transactional keywords are served by product pages you already have. This mapping becomes your content calendar — you are not brainstorming "what should we write about" anymore; you are executing against a research-backed keyword map.
Prioritize clusters by three factors: relevance to your catalog (does this cluster naturally lead to your products), competition (can you realistically rank within 6 months), and coverage (how many supporting keywords exist in this cluster — bigger clusters compound faster). A cluster with 30 supporting keywords is worth more than a cluster with 5 because each new page lifts every other page in the cluster.
Keyword Research for AI Search
AI search engines surface different content than Google's traditional results. The queries that trigger AI-generated answers tend to be question-format ("how does X work"), comparison-format ("X vs Y for Z"), and recommendation-format ("best X for Y"). Research which queries in your niche trigger AI Overviews in Google, conversational answers in ChatGPT Search, or cited responses in Perplexity. These are the queries where your content can earn citations — named references with links back to your site.
The research process is manual but high-value: search 20-30 of your target keywords across ChatGPT, Perplexity, Google (checking for AI Overviews), and Claude. Note which queries get AI answers, what format the answers take, and which sources get cited. Look for patterns: are certain content structures (step-by-step, comparison tables, definitive answers) getting cited more than others? Build your content in those formats for those specific queries.
AI search citation is the new layer on top of traditional SEO — not a replacement for it. A page that ranks well in Google and is formatted for AI citation captures traffic from both channels. The long-tail specificity that works for traditional search also works for AI citation because AI retrieval systems reward pages that answer one question definitively rather than pages that touch many topics shallowly.
The Keyword Research Workflow (Step by Step)
Step 1: List your product categories — not individual products, but the categories they belong to (running shoes, camping cookware, skincare serums). Step 2: Run each category through Google autocomplete (category + every letter), People Also Ask (search the category and expand 3-4 levels), and Amazon suggestions. Record every query that represents a distinct buying question. Step 3: Pull your Search Console data for the last 90 days, filter to queries with impressions over 100 and clicks under 10 — these are opportunities where Google sees you as relevant but you lack dedicated content.
Step 4: Group all collected queries into topic clusters. Each cluster shares a theme and would cross-link naturally. Step 5: Prioritize clusters by intent (commercial investigation first), competition (where can you win in 6 months), and catalog connection (which clusters lead to your products). Step 6: Assign content types to each keyword — article, guide, comparison, tool, or collection page. Step 7: Build your content calendar from the prioritized, typed keyword map. Execute the top cluster first, measure results at 60 and 90 days, then move to the next cluster.
Repeat this workflow quarterly. Keyword landscapes shift as competitors publish, as search behavior changes seasonally, and as AI search surfaces evolve which queries earn citations. A keyword map is not a one-time artifact — it is a living document that grows with your store's authority and the market's search behavior.