UX / Navigation
How Many Categories Should an Ecommerce Store Have?
// group by how people shop, not how you stock
By Fracto Solutions
October 8, 2026
The short answer
Have as many categories as it takes to cover your range without overwhelming the menu, usually a short set of top-level categories with subcategories carrying the detail. Group products by how customers shop, keep the structure shallow, and let search handle the long tail. The right number follows the catalog, not a fixed rule.
60%
76%
0
Is there a right number of categories?
Not a fixed one. The goal is a menu a shopper can scan quickly at the top level, with enough subcategory depth that products aren’t buried. A short, clear top-level set with well-organized subcategories beats both a sprawling menu and an over-flattened one.
Chasing a magic number misses the point, and the number most often quoted does not survive contact with its source. The five-to-seven rule comes from Miller’s 1956 paper on short-term memory span; Miller himself said it had nothing to do with comprehending printed text, Jakob Nielsen calls it misleading for menu design, and Edward Tufte noted the design conclusion could only be reached by not reading the paper. Items visible on screen do not need to be memorised. Baymard names no optimal count either. This is the detailed companion to our navigation pillar.
What the benchmark data does say is that the failure is grouping, not counting. Baymard assesses 180+ leading US and European sites against more than 16,000 UX elements manually reviewed and scored, and finds 60% of sites do not divide their categories and subcategories into manageable chunks. A menu fails because its groupings are undifferentiated, not because it has eleven top-level items rather than seven. Its only public threshold is a suggestion to subdivide a list once it reaches around ten items, which is about list length rather than an optimal category count.
What happens if you have too few or too many?
Too few categories force products too deep or dump unrelated items together, so shoppers can’t narrow effectively. Too many turn the menu into an unscannable wall and split your range so thinly that categories feel empty. Both hurt findability.
Problem
Symptom
Fix
Too few categories
Unrelated products lumped together; deep burial
Add meaningful subcategories by real buying criteria
Too many categories
Unscannable menu; thin, near-empty categories
Consolidate into a shorter top-level set
Wrong grouping
Shoppers can't predict where a product lives
Regroup by how customers shop, test the labels
Fix grouping and depth around how customers actually narrow their choice, not around your internal catalog.
One symptom is easy to miss because it looks like a design detail rather than a structural one: 95% of sites fail to highlight the shopper’s current scope in the main navigation. If your categories are poorly grouped and the menu also never shows people which section they are in, the two compound, and shoppers navigate by guesswork.
How should I decide what the categories are?
Start from how customers describe and choose your products, then group accordingly. Use their language for labels, and let the real decision criteria in your category (type, use case, size, audience) define the structure.
Map the words shoppers actually use and the ways they narrow a choice, then build categories around those. Test the labels on someone outside your team: if they can predict where a product lives, your structure works. Run that test literally. Give someone a list of ten products and ask which category each one lives in. If they cannot predict it, the labels are wrong however defensible the structure looks on a whiteboard.
What about very large catalogs?
Large catalogs still keep a short top level, but lean harder on subcategories, strong filtering, and search. The top-level menu stays scannable; depth and faceted filtering carry the detail so shoppers can drill down without a giant menu.
The mistake on big catalogs is exposing everything in the menu. Instead, keep the top level tight, use subcategories for structure, and make filtering and search do the heavy lifting for the long tail. That keeps browsing manageable no matter how many SKUs you carry.
The page that does this work is the intermediary category page, the one a shopper lands on after picking a top-level category, and it is the most neglected page in the benchmark: 76% of sites do not feature subcategories as the primary content there, and 55% are missing or have hard-to-interpret subcategory thumbnails. That is precisely where a large catalog should be handing people their next narrowing step. On mobile the gap is wider still, with 42% of sites providing no “View All” option at each level of the product catalog.
We size your category structure around how customers shop, not your SKU list.
We regroup categories around the words and decision criteria your buyers actually use, keep the top level scannable, and push detail into subcategories and filters, then check in the data whether products get found faster.
It’s the same measurement-first approach behind our conversion optimization work: we agree the metric before the test runs and report the result at the confidence level the test actually reached.
// we would rather show up in your reporting than in your inbox
Frequently asked questions
Is there an ideal number of top-level categories?
No fixed number, and the five-to-seven rule often quoted is Miller’s 1956 memory-span research misapplied to menus, which Miller and Nielsen both rejected for interface design. Keep the top level short and scannable with subcategories carrying the detail. What the data shows is that grouping matters more than counting: 60% of sites fail to divide categories and subcategories into manageable chunks.
Should every product be reachable from the menu?
Not necessarily every product, but every category should be. Let subcategories, filtering, and search handle the long tail, especially on large catalogs.
How do I know if I have too many categories?
If the menu is hard to scan, categories feel thin or near-empty, or shoppers can’t predict where a product lives, consolidate into a shorter top-level set.
How do large catalogs stay usable?
Keep a tight top level and rely on subcategories, strong faceted filtering, and search for depth, rather than exposing thousands of items in the menu itself.
Sources
Navigation benchmark and compliance gaps: Baymard Institute, 2025 homepage and category navigation UX benchmark. 180+ leading US and European sites, 16,000+ UX elements manually reviewed and scored, 13,000+ categorized best and worst practice examples. 58% of desktop and 67% of mobile sites “mediocre” to “poor”. Figures are quoted as Baymard states them and are not device-split unless noted: 95% do not highlight the user’s current scope in the main navigation; 76% do not feature subcategories as the primary content on intermediary category pages; 61% do not have a hover delay; 60% do not divide categories and subcategories into manageable chunks; 59% do not provide the full scope for links on mobile homepages (mobile-specific); 55% are missing or have hard-to-interpret subcategory thumbnails; 51% do not make it clear where hit areas in visual elements lead; 33% do not make headers clickable. Published 6 August 2024, last updated 30 September 2025. https://baymard.com/blog/ecommerce-navigation-best-practice
Navigation and category UX research: Baymard Institute, homepage and category navigation usability research. Benchmark of 344 top-grossing US and European sites against 40+ guidelines, with 900+ usability issues encountered during moderated testing of 19 sites across 8 verticals. Reports 76% of sites at “mediocre” or worse. This page carries no publication or update date and is not device-split; where it disagrees with the dated 2025 benchmark we use the dated figures. Note its 76% measures overall performance and is a different statistic from the 76% intermediary-category-page figure above. https://baymard.com/research/homepage-and-category-usability
Mobile ecommerce UX: Baymard Institute, mobile UX benchmark. 150+ mobile ecommerce sites, 71,000+ manually reviewed and rated UX elements. 75% rate only “Mediocre” on overall ecommerce UX and 95% are “Mediocre” or “Decent”, with one site rated “Good”; this scores the whole mobile experience, not navigation alone. 94% of mobile sites do not highlight the current scope in the main navigation. 42% do not provide a “View All” option at each level of the product catalog. Last updated 14 July 2026. https://baymard.com/blog/mobile-ux-ecommerce
The five-to-seven menu rule, and why we do not use it: Miller’s 1956 paper measured short-term memory span and one-dimensional absolute judgment, not on-screen reading. Miller: the limits have “nothing to do with a person’s capacity to comprehend printed text”. Jakob Nielsen calls it “misleading to use it for menu design”. Edward Tufte: the seven-item design conclusion “can be sustained only by not reading the paper”. Items visible on screen need no memorisation, so the limit does not bind. UX Myths: choices limited to seven plus or minus two
Stated as Fracto’s position, not research: that the count should follow the catalog and how customers shop, and the label-prediction test. These come from client work. Baymard names no optimal number of top-level categories in its public material, and part of its guideline set sits behind a paid tier we have not reviewed.