UX / Navigation
Ecommerce Navigation: Menus That Help People Buy
// if they can't find it, they can't buy it
By Fracto Solutions
October 8, 2026
The short answer
Good ecommerce navigation gets a shopper from any page to the product they want in as few clear steps as possible. Use plain category labels your customers would recognize, keep the structure shallow, make it work on mobile, and pair it with visible search. Navigation is findability, and findability is revenue.
58%
67%
95%
Why does navigation make or break a store?
Because a shopper who can’t find a product can’t buy it. Navigation is the map of your store; if it is confusing, mislabeled, or buried on mobile, people leave even when you stock exactly what they wanted.
This is measured, not assumed. Baymard’s 2025 homepage and category navigation benchmark, covering 180+ leading US and European sites with more than 16,000 UX elements manually reviewed and scored, rates 58% of desktop sites and 67% of mobile sites “mediocre” to “poor”. Its separate, undated research page reports 76% of 344 top-grossing sites at “mediocre” or worse; that figure is not device-split and looks like the older snapshot, so we lead with the dated one. Baymard says there is “great room for improvement” specifically within category taxonomy, main navigation and intermediary category pages, and also within the homepage.
The upside is that this is a durable advantage rather than a one-off win. Clear navigation helps every visitor on every visit, it pairs with site search as the two ways people find products, and on the benchmark evidence most of your competitors are not doing it well.
How should I structure my main navigation?
Around how your customers think, not your internal org chart. Use plain, recognizable category labels, keep the hierarchy shallow so most products are reachable in about two levels, and make the primary menu consistent across every page.
The common failure is labeling navigation in company language, clever category names or internal product groupings, instead of the words shoppers actually use. Name categories the way a customer would search for them. Then keep the tree shallow: every extra level is another decision and another chance to drop someone. Reaching most products within about two levels is our own working rule rather than a published finding, and we say so rather than dressing it up as research.
Two structural gaps show up again and again in the benchmark data, and both are cheap to fix. 95% of sites fail to highlight the shopper’s current scope in the main navigation, so people lose track of which section they are in. And 76% do not feature subcategories as the primary content on intermediary category pages, which is the page a shopper lands on precisely when they want to narrow down. A third of sites do not make drop-down headers clickable at all. None of these are redesigns; they are corrections.
Nav element
Purpose
Best practice
Primary menu
Route to main categories
Plain labels; consistent on every page
Subcategories
Narrow within a category
Shallow; reachable in about two levels
Search
Direct path for known-item shoppers
Visible open box; pairs with browsing
Utility nav
Account, cart, help
Top corner, out of the way but findable
Footer
Catch-all and trust links
Policies, support, full category list
Name categories the way customers would, and keep the path to any product short.
Mega menu or simple dropdown?
Use a mega menu when you have a broad catalog with many categories to show at once; use a simple menu when your range is narrow. The deciding factor is catalog breadth, not fashion, a small store with a mega menu just looks empty and over-engineered.
Both can work; the mistake is copying a big retailer’s mega menu onto a small catalog, or cramming a broad catalog into a thin dropdown. We cover the trade-off in detail in mega menu vs simple navigation, but the short version is to match the pattern to how much you actually need to show.
How many top-level categories should I have?
Enough to cover your range without overwhelming the menu, usually a handful rather than a long list. Too many top-level items turn the menu into a wall; too few bury products too deep. Group by how customers shop, not by how you stock.
There is genuinely no magic number, and you should be suspicious of anyone who gives you one. The “five to seven items” rule you will see quoted comes from Miller’s 1956 paper on short-term memory span, which Miller himself said had nothing to do with reading printed text; Jakob Nielsen calls it “misleading to use for menu design”. Items visible on screen do not need to be memorised, so the limit does not bind. Baymard names no optimal count either.
What the data does say is that the failure is usually grouping rather than counting: 60% of sites do not divide their categories and subcategories into manageable chunks, so the menu becomes a wall of undifferentiated links whatever its length. Baymard’s only public threshold is a suggestion to subdivide a list once it reaches around ten items. The full reasoning, including what to do with a large catalog, is in how many categories should a store have.
What belongs in the header, footer, and utility nav?
The header carries primary navigation, search, and the cart. Utility nav holds account and help. The footer is for policies, support, trust signals, and a fuller category list. Each region has a job; problems start when everything competes in one place.
A clean header keeps the primary menu, search, and cart obvious and uncluttered. The footer does quieter but important work: it is where shoppers look for shipping and returns policies, contact details, and reassurance before buying, which ties into building trust on your store. Don’t crowd the header with links that belong in the footer.
How do I know my navigation is working?
Watch behavior, not opinion. High use of search can signal that browsing is failing. Category pages with high exits, or products people reach only via search, point at navigation gaps. Session recordings show where people hunt and give up.
Your analytics will tell you where navigation leaks: categories with poor engagement, products that only ever get found through search, pages where mobile users pogo-stick back and forth. Watching a handful of session recordings of people trying to find something usually makes the problem obvious. It is the same find-the-leak discipline behind our growth-strategy pillar: measure the behavior, then fix the worst gap.
Do breadcrumbs and secondary navigation still matter?
Yes. Breadcrumbs show shoppers where they are and let them step back up a level without starting over, and secondary navigation, filters, related categories, in-page links, helps people move sideways through your catalog. Both reduce the dead ends that send shoppers to the back button.
Breadcrumbs are a small feature with an outsized usability payoff: they give shoppers a sense of place in a deep catalog and a one-click way back to a broader category. They also reinforce your structure for search engines. Keep them on category and product pages where the hierarchy runs more than a level deep.
Secondary navigation picks up where the main menu leaves off. Once a shopper is inside a category, filters and links to related categories or complementary products keep them moving without returning to the top-level menu. The aim across all of it is the same: never leave a shopper on a page with no obvious next step toward the product they want.
One small check worth running on your own site: 33% of sites do not make drop-down headers clickable, and 95% do not show the shopper which section they are currently in. Both leave people navigating by memory. If your breadcrumb trail and your menu disagree about where someone is, the breadcrumb is doing half a job.
How is navigation different on mobile?
On mobile the menu is collapsed behind a tap, vertical space is scarce, and fat fingers need big targets, so structure and labeling matter even more than on desktop. The same categories must be reachable, just through a flow designed for a small screen rather than a squeezed desktop menu.
How much of ecommerce traffic is mobile depends on how you count. Siege Media’s analysis of 13.4 billion sessions across 120 brands puts its ecommerce row at 55.8% mobile measured across all traffic, and 65.3% for the “typical site”, where each brand counts equally, because a few desktop-heavy giants drag the first figure down. Baymard puts it at roughly three quarters of retail store visits. Whichever fits your store, the mobile menu is the one that decides whether people find things, and it is usually an afterthought inherited from the desktop design.
The benchmark data says so plainly. Baymard rates 67% of mobile sites mediocre to poor on homepage and category navigation against 58% on desktop. Its separate mobile benchmark of 150+ sites finds 75% rate only “Mediocre” on overall ecommerce UX, with exactly one site rated “Good”, though that is a whole-experience score rather than a navigation one. On navigation specifically, 94% of mobile sites do not highlight the shopper’s current scope and 59% do not give the full scope for links on mobile homepages. A further 51% of sites, desktop and mobile together, do not make it clear where hit areas in visual elements lead.
So give the mobile menu real attention: a clear menu trigger, shallow levels, large tap targets, labels that say where a link goes, and visible search. Don’t bury the catalog behind an unlabeled icon or an endless accordion. This is the same mobile-first discipline covered in mobile conversion optimization.
We structure navigation around how your customers shop, then prove it in the data.
We map your categories to the words and paths real shoppers use, flatten the structure so products aren’t buried, and fix the mobile menu where most stores quietly lose sessions, then measure whether people find products faster and buy more.
It’s the same measurement-first discipline behind our conversion optimization and ecommerce operations work: we agree the metric before the test runs, change one thing at a time, 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
How many top-level categories should an ecommerce site have?
There is no established optimal number, and the five-to-seven rule you will see quoted is Miller’s 1956 memory-span research misapplied to menus; Miller and Nielsen both said it does not transfer. Keep the top level short and scannable with subcategories carrying the detail, and group by how customers shop rather than how you stock. What the benchmark data does show is that 60% of sites fail to divide categories and subcategories into manageable chunks, so grouping matters more than counting.
Should I use a mega menu?
Use one when you have a broad catalog with many categories to surface at once. For a narrow range, a simple menu looks cleaner and avoids an empty, over-engineered mega menu.
How deep should my category structure be?
Shallow. Aim to let shoppers reach most products within about two levels. Every extra level is another decision and another place to lose people.
What is the biggest navigation mistake?
Labeling categories in company language instead of the words customers use. If your labels don’t match how people think about your products, they won’t find them, no matter how logical the structure seems to you.
Does navigation affect SEO as well as conversion?
Yes, indirectly. Clear structure and internal linking help both shoppers and search engines understand your catalog, which connects to ecommerce SEO, but design navigation for shoppers first.
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
Mobile share of ecommerce traffic: Siege Media, mobile vs desktop traffic share. 13.4 billion sessions across 120 brands, 1 January 2023 to 30 June 2026, last updated 11 August 2026. The ecommerce row reads 55.8% mobile across all traffic and 65.3% for the “typical site”, where each brand is weighted equally. Two caveats the study states: 2026 is a partial year of six months, and tablet and smart TV traffic are excluded. https://www.siegemedia.com/research/mobile-vs-desktop-traffic-share
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: reaching most products within about two levels, grouping categories by how customers shop rather than how you stock, and the order in which we fix navigation problems. These come from client work rather than a published study. Baymard publishes no public benchmark comparing mega menus against simple drop-downs and names no optimal number of top-level categories; part of its guideline set sits behind a paid tier we have not reviewed.