Catalog structure for demand: how to build categories that actually bring traffic

Catalog structure is the single decision at launch that is most expensive to redo later. Not the design, not the platform, not the copy. It decides which searches you can even compete for, and changing it means changing page addresses, redirects, and the risk of a ranking drop.

Two catalog logics

Warehouse logic: categories match how stock sits on your shelves and how it is convenient to count. "Kitchen," "Decor," "Textiles."

Demand logic: categories match how people actually search. "Salt and pepper sets," "Vases for dried flowers," "Doormats."

Almost every store starts with the first, because it is obvious. The problem is nobody googles "decor" intending to buy — it is too broad. People search specifics, and whoever has a dedicated page for that specific wins.

The right answer is not "replace one with the other," it is to layer them: warehouse logic stays the top level, demand logic becomes the second and third.

How to gather keyword data without expensive tools

Step 1. Search Console — the most valuable source

Queries → 3-month period → sort by impressions. Anything with impressions and zero clicks is demand Google is already showing you, with no page to answer it. It is the warmest list there is, because it is about your actual assortment.

Step 2. Search suggestions

Type a category name into Google and look at autocomplete and the "related searches" box at the bottom. Free, and shows the actual phrasing people use.

Step 3. Competitors

Open two or three stores already ranking well in your niche and look at their menu. Do not copy — but notice which subcategories they broke out separately. It is usually not an accident.

Step 4. Ahrefs or a similar keyword tool, if you have the budget

Gives you search volume and lets you filter out queries not worth a dedicated page. Useful, but not required if you have done the first three steps.

When a subcategory is worth it

Check three conditions before creating a page. You need all three:

• There is demand. The query exists and has meaningful volume — not "two impressions a month."
• There is product. You have at least 5–7 items that genuinely fit. A category with two products looks empty, both to a person and to Google.
• There is a real difference. This page differs from its neighbor in a way you can put into words. If "glass vases" and "vases made of glass" are the same thing, make one page.

Breaking the third condition causes cannibalization: your two pages compete with each other, Google cannot tell which to show, and shows neither.

How many subcategories you need

For a catalog of 300–800 products, 15–30 meaningful subcategories is typical. Fewer, and you are leaving demand on the table. More, and you start spawning two-product pages, which makes the catalog awkward for shoppers — a direct hit to conversion.

Simple rule of thumb: every subcategory should have enough product that the page looks complete.

Filters are not a substitute for categories

A common assumption: "why do we need subcategories, we have filters." Filters are convenient for a shopper already on the site, but filter pages are usually blocked from indexing — correctly so, since they generate thousands of variants.

The smart middle ground: turn your most popular filter combinations into real static pages with their own URL, title, and copy. Leave the rest as filters, blocked in robots. This is standard practice on most platforms.

What to write on a category page

Not an "SEO text" at the bottom nobody reads. Two or three paragraphs above or right below the products that actually help someone decide:

• how the variants differ from each other;
• how to choose the right size, material, or volume;
• what people buy most often and why;
• what is worth knowing about care or compatibility.

That kind of text gives search engines the page’s meaning and genuinely removes a shopper’s objections at the same time. "In our online store you can buy vases at a great price" does neither.

How to restructure without losing rankings

1. Record your current rankings and traffic per page — you will need something to compare against.
2. Create new subcategories additively, without breaking old URLs.
3. If you do remove an old category, always 301-redirect it to the closest match by meaning.
4. Update your menu and internal links: new pages need to be reachable within two or three clicks.
5. Check your sitemap and request a recrawl in Search Console.
6. Watch the trend for 6–8 weeks before drawing conclusions.

If you are not sure whether your current structure is worth rebuilding, a free audit will answer that: I look specifically at the gap between what people search for and which pages you actually have.