Shopify Search and Discovery Filters: Setting Up Filters Shoppers Actually Use
In Shopify's Search and Discovery app, keep the filter list short and data backed: availability, price, and one or two attributes your catalog actually maintains, like size or a product metafield. Filters built on tags you stopped maintaining, or on metafields filled for a third of the catalog, produce empty results that read as out of stock. Test every filter combination on the largest collection, on a phone, before calling it done.
Filters fail quietly. A store adds fourteen of them because the app allows it, shoppers tap two, get a zero results page for a product that exists three pages deeper, and conclude the store does not carry their size. The filter told the truth about bad data, and the shopper heard it as an inventory problem.
The fix is restraint backed by an audit: filter on fields you can swear are complete, and nothing else. Three honest filters outperform a dozen speculative ones on every metric that matters, including the only one the shopper sees, which is finding the product.
Audit the data before touching the app
Export the catalog and check completeness for each candidate field. Vendor and product type are native and usually clean. Tags are usually not: years of imports and seasonal naming produce synonyms, plural drift, and dead values that each spawn their own filter option. Metafields can be excellent filter sources, a fabric or fit field maintained at product creation, or a trap, filled enthusiastically for one season and abandoned.
A workable rule: do not ship a filter whose underlying field is complete on less than about 90 percent of the products it will appear over. Below that, the filter lies by omission, hiding products that simply lack the data. Fixing the data for a field you believe in is a legitimate project; shipping the filter first and the data later is not a sequence, it is a complaint generator.
Configuration choices that matter
Order filters by how shoppers think, not by how the catalog is organized: availability and size before vendor, price where it can narrow a large collection early. On collection pages with deep inventory, set the default layout so filters are visible without a click on desktop and behind a clearly labeled button on mobile, then check the results count updates are announced and the page does not lose the shopper's scroll position on every tap.
Use the app's synonym and spelling handling in search itself while you are in there: add the misspellings and alternate terms your own search logs show. The search results page your analytics already records is the cheapest research you own; the queries returning zero results are a to do list written by customers.
What to leave off
Leave off filters for attributes shoppers cannot verify at a glance and cannot act on: internal collection codes, supplier names, warehouse locations. Leave off color filters built on free text color names until they are normalized to a standard set, unless you enjoy forty shades of near black listed as separate options.
Revisit quarterly. A filter that earned its place in January can rot by September as new products arrive without the field filled. The audit is the maintenance; the app settings are the easy part everyone photographs for the launch announcement.
Related reading
Related reading: Shopify Markets vs Separate Stores for Selling Abroad, What Shopify Flow Can Automate Without a Developer, and Shopify Search Settings: What Shoppers Actually Find.
Frequently asked questions
How many filters should a collection page have?
Three to five well maintained filters cover most catalogs: availability, price, and one to three attributes central to how the product is chosen, like size or fabric. More filters mostly add empty combinations and slower decisions.
Why do my filters show no products when the products exist?
Almost always a data gap: the products lack the tag or metafield value the filter reads, or the value is spelled differently. Check the specific products against the filter's source field before changing any app settings.
Can filters use metafields in Shopify?
Yes. The Search and Discovery app supports filters based on product metafields, variants, price, availability, vendor, and tags. A metafield filter is only as good as the completeness of that field across the catalog, so audit before enabling.
Do filters help SEO?
Filtered URLs are usually parameterized and best kept out of the index to avoid thin duplicates. The SEO value of filters is indirect: shoppers who find products buy them, and collections structured around real attributes make better landing pages than a wall of filtered links.