Two people type into Google within the same minute. One searches “hiking boots.” The other searches “waterproof hiking boots men’s size 11.” On a blog, you might chase both with a single article about choosing footwear. In a store, they are not the same customer at all, and treating them as one is where a lot of ecommerce keyword research quietly goes wrong.
The first shopper is browsing. The second already has a boot in mind, a foot size, and probably a card in their wallet. Send them to the same page and you will disappoint at least one of them.
That gap is what this whole exercise is really about.
Why a store searches differently than a blog
For a typical blog, keyword research is a hunt for topics to write about. Find a question people ask, answer it, publish. For an online store, the job is different: you are matching the words people search with the specific pages that let them buy. Those are not the same problem, and the store version is the harder one.
Think again about our two searchers. The shopper who typed “waterproof hiking boots men’s size 11” is close to a purchase. They should land somewhere they can pick a color and add the thing to a cart, not on a 2,000-word explainer about boot materials. The person who searched “how to break in hiking boots” is nowhere near checkout. Drop a bare product grid in front of them and they will bounce.
Point every visitor at the same kind of page and you lose both directions. Buyers get essays when they wanted a Buy button. Researchers get a store aisle when they wanted an answer. So ecommerce keyword research is as much about sorting terms by what the searcher actually wants as it is about digging the terms up in the first place. Get the sorting right and every page gets to do the one job it is good at.
None of this needs an expensive toolkit to start. A free account in Google Keyword Planner, the queries already sitting in Search Console, and your own product catalog will carry you a long way. Paid tools like Ahrefs make the work faster and show you more of what competitors rank for, but they are an accelerator, not a prerequisite. The thinking is the part that matters, and the thinking is free.
Search intent is the whole ballgame
Search intent is just the reason behind the search. Nothing fancier than that. But for a store it pays to keep three rough buckets in your head.
Informational. This is research. How does something work, how do I choose between two options, how do I care for the thing after I own it. “How to waterproof leather boots” lives here.
Commercial. One step closer to a decision. Comparisons, roundups, the endless “best” queries where someone is weighing options and hasn’t committed. “Best waterproof hiking boots” is the classic shape.
Transactional. Buying. A specific product, a category, or any phrase that reeks of readiness to spend. “Waterproof hiking boots men’s size 11” is about as transactional as it gets.
The lines between these blur, and that is fine. “Waterproof hiking boots” on its own sits somewhere between commercial and transactional depending on who is asking. You are not grading an exam. You are making a good-enough call about what the searcher most likely wants so you can point them at the right kind of page.
Here is why the buckets matter so much. Google is constantly trying to serve the kind of page that satisfies each kind of query, and it has gotten good at it. If the first page of results for a term is wall-to-wall product listings, that is Google telling you the searcher wants to shop. A long article is not going to rank there, and even if it snuck in, it would not convert. Flip it around: if the results are all guides and explainers, a naked product page will feel out of place and get ignored.
So before you marry a keyword to a page, go look at what already ranks for it. Open the results, actually read the top handful, and ask what type of page keeps showing up. Are they product pages, category pages, or articles? Is there a shopping carousel bolted across the top, or a block of questions people also ask? That lineup is Google’s answer, spelled out for free. In our experience this one habit prevents more wasted pages than any tool ever will.
Build your seed list from stuff you already own
You do not have to conjure keywords out of thin air. The good ones are usually sitting in plain sight, in the language you and your customers already use every day. Your product names are a seed list. Your category names are a seed list. So are the questions your support inbox and your sales calls surface for the hundredth time, because those are real phrases from real buyers, not guesses.
A few sources are worth mining hard when you run a store:
- Your own site search box. What people type into it is a direct window into demand and vocabulary. If shoppers keep searching “waterproof” inside a boots store, that word belongs in your titles and your category structure, not buried three clicks deep. It even shows you things customers want that you do not stock yet, which is its own kind of gift.
- Autocomplete and related searches. The suggestions Google offers as you type, and the related terms it lists at the bottom of the results, tell you how people actually phrase things and what they mention in the same breath. Type “waterproof hiking boots” and watch what it wants to finish for you.
- Competitor and marketplace listings. How rivals and the big marketplaces name and file their products will surface phrasing and angles you missed. You do not have to copy them, just read them. The filters they expose on a category page are a tidy map of how buyers slice a decision.
- Reviews and customer questions. The words shoppers use to describe what they loved or what went wrong are very often the exact words they typed into a search bar a week earlier. A review griping that a boot “leaked in the first puddle” is a keyword in disguise.
Dump all of it into one place. Do not filter hard yet. The seed stage is about breadth, and the instinct to tidy up too early kills good candidates before you have looked at them properly. Narrowing comes next.
Expand, then group into clusters
Now take that raw pile and stretch it. Add the natural variations, the neighboring products, the questions that orbit each topic. “Waterproof hiking boots” pulls in “waterproof trail shoes,” “are hiking boots waterproof,” “waterproof boots for wet weather,” and a dozen more once you start pulling the thread.
Then, and this is the part people skip, group the results into clusters of closely related terms instead of treating every phrase as its own lonely target. A lot of keywords are just different ways of asking for the same thing. “Waterproof hiking boots,” “water resistant hiking boots,” and “hiking boots for rain” can usually be handled by one genuinely strong page, not three thin ones elbowing each other for the same spot.
A quick gut check when you are unsure whether two terms belong together: would the same page make both searchers happy? If yes, cluster them. If one wants to buy and the other wants to learn, split them, no matter how similar the words look. “Buy waterproof hiking boots” and “are waterproof hiking boots worth it” share four words and want completely different pages.
Clustering does something else, too: it shows you the shape of the site you should be building. A cluster full of transactional variations is pointing at a category page. A cluster of comparison and how-do-I questions is pointing at a guide. Group first, assign pages second. Do it the other way around and you end up with dozens of near-duplicate pages that split your own relevance and confuse shoppers and search engines in equal measure.
Prioritize on fit and intent, not on volume
The tempting move is to chase the terms with the fattest search numbers. For a store, that instinct is usually wrong.
The huge, broad terms are two things at once: brutally competitive and loosely connected to any single product you sell. “Hiking boots” is a battlefield, and even winning it drags in browsers who may want kids’ sizes, or reviews, or something you do not carry. The more specific phrase pulls fewer searches on its own, sure. But it often carries far clearer intent and converts better, because the person typing it knows exactly what they are after. “Waterproof hiking boots men’s size 11” almost sells itself.
When you decide what to go after first, weigh a handful of things together rather than fixating on any one number:
- Relevance. How tightly does the term match something you genuinely sell? A dead-on match on a smaller term beats a loose match on a giant one, every time.
- Intent fit. Does the query signal buying readiness that lines up with a page you can actually offer, or is it pure research that will never touch a cart?
- Competition. Are the current results locked down by big, entrenched names, or is there daylight for a focused store to slip in?
- Business value. Would ranking here bring buyers for products that matter to your margins, or just a crowd of curious clickers?
Hold those four in balance and you stop pouring weeks into terms that look impressive in a spreadsheet and almost never turn into orders. The specific, slightly boring keyword that quietly sells boots week after week is worth more than the glamorous head term you will never outrank.
Map each cluster to the page that should own it
All of this pays off in one plain artifact: a keyword map. It is just a record of which term or cluster each important page is meant to target. Nothing exotic, and honestly more useful for being simple.
The rule is one primary focus per page, so your pages are not quietly competing with each other for the same query. Transactional clusters map to product and category pages. Research and comparison clusters map to guides, buying advice, and the rest of your content. A map can be nothing more than a table:
| Keyword cluster | Intent | Target page type |
|---|---|---|
| Specific product variations (e.g. “waterproof hiking boots men’s size 11”) | Transactional | Product page |
| Broad product category terms (e.g. “hiking boots”) | Transactional or commercial | Category page |
| Comparison and “best” queries (e.g. “best waterproof hiking boots”) | Commercial | Buying guide or roundup |
| How-to and explainer questions (e.g. “how to break in hiking boots”) | Informational | Blog article or resource |
Once the map exists, two problems you could not see before jump straight out. Gaps: places where there is real demand and you have no page that fits it. Overlaps: several pages all chasing the same thing, quietly cannibalizing each other, begging to be consolidated into one. Both are hard to spot in your head and obvious on a single sheet.
Keep the map somewhere everyone can reach it, because it earns its keep over months, not in one sitting. When someone proposes a new blog post or a new category, the first question is easy: what cluster does this own, and does anything already own it? If the answer is that a page already covers it, you have just saved yourself a duplicate.
Treat research as a habit, not a project
One thing keyword research is not: a task you finish once, file away, and never reopen.
Shopper language drifts. You add products and retire others. New questions bubble up around your category that nobody was asking a year ago. The stores that get the most out of this treat research as a standing habit rather than a one-off sprint. They circle back to the site search data every so often. They check which pages are actually pulling search traffic and which have gone quiet. They watch Search Console for queries that are bringing people in by accident, then build the page those searchers deserved. They refresh content when the intent behind a term seems to have shifted under their feet, which it does more often than you would expect.
Kept up, this compounds in a way that is easy to underrate. Every round sharpens the keyword map a little more, closes one more gap, tightens the fit between what people search for and what your pages hand them. And that fit, far more than any single clever keyword you managed to find, is what steadily grows the kind of search traffic that ends in a sale instead of a bounce.
