Many stores struggle with over-filtering: hundreds of filter options, nested levels, sub-options within sub-options, and, in the end, the customer leaves the site. Not because they couldn't find anything suitable, but because they stopped looking.
The opposite of user-friendliness isn't chaos. It's too much choice without guidance.
The research: Hick's Law and the paradox of choice
Three research findings are particularly relevant here:
- Hick's Law (1952): The time it takes to make a decision increases logarithmically with the number of options. More filters = more decision effort = more drop-offs.
- Barry Schwartz, The Paradox of Choice (2004): More choice leads to less satisfaction and more decision paralysis. In the well-known jam experiment (Iyengar & Lepper, 2000), shoppers bought far more often when offered 6 flavors than 24.
- Baymard Institute (2025): 58% of e-commerce sites on desktop and 78% on mobile perform only "poor" to "mediocre" when it comes to product lists and filtering.
The goal: 3 steps to the product
The 3-click rule is a rule of thumb, not a proven law. It still works as a target: users should be able to get from the entry point to the target product in no more than 3 interactions. For the filter structure, this means:
max. 7 entries
Most important first
Direct and clear
Anything beyond these 3 levels increases the likelihood of drop-off exponentially. Deep filter structures with 4+ levels are almost always an information architecture problem, not a user problem.
How many filter levels: concrete recommendations
- Standard stores (fashion, household, electronics): 1 filter level + free-text search covers 80% of purchases. Maximum depth: 2 levels.
- Complex stores (B2B, industrial, MedTech): 2 filter levels are ideal. 3 levels as an exception for highly specialized product categories, but with a clear tree structure and breadcrumbs.
- Faceted search (multidimensional): Several filters at once (price + color + size), without adding depth. This is the most powerful and user-friendly combination.
The paradox of "complete" filters
Many stores build filters designed to cover the entire product range, "so the customer can really find everything." The opposite happens: customers choose the wrong filter first, end up with an empty results list, and leave the store.
According to the Baymard Institute, on sites with mediocre product lists and filters, 67–90% of test users abandoned their search. On slightly optimized sites, only 17–33% did, for the same products. Less is more, as long as what's left is the right thing.
What makes good filter UX
Do
Don't
"Unnoticed filters are — to the user — effectively the same as nonexistent filters." — Christian Holst, Baymard Institute, 2015
Test your store: how many clicks does a first-time visitor need to reach your best-selling product? If it's more than 3, start there. Not with a new design, but with categorization and filter prioritization.
Sources: Hick (1952), Quarterly Journal of Experimental Psychology, 4(1), 11–26 · Schwartz (2004), The Paradox of Choice · Iyengar & Lepper (2000), Journal of Personality and Social Psychology, 79(6), 995–1006 · Crowley (2025), Product List UX 2025, Baymard Institute · Baymard Institute, Product Lists & Filtering UX Research · Holst (2015), The Current State of E-Commerce Filtering, Smashing Magazine.
Published June 2026