What is the difference between filters and facets?
A filter limits a collection of products based on a condition, such as brand, price or size. Facets are dynamic filter groups that are made up of the properties of the current results. They can also show and change numbers when the visitor makes a choice. In practice, the terms are often used interchangeably, but dynamic facets give more context and prevent irrelevant filtering options.
How does Smart Filters determine which filters are relevant?
The available category, search results and product attributes are the basis. Per product group, it can be set which properties are important and in what order they appear. A search can also provide additional context. For example, a filter for shoe size does not have to appear with bags and screen diagonal only gets a prominent place where that property really helps with the choice.
Can the order of filters be sent by category?
Yes. Filters that strongly influence a purchase decision can be at the top, while less used properties are placed lower. The order you want may vary by category. In clothing, size and color are often important, while with a laptop price, screen size and working memory make more sense. Usage data helps to improve that order later targeted.
Which product data is needed for good filters?
Each filter requires consistent product information. Consider unambiguous values for size, color, material, brand, capacity or technical characteristics. Values such as “dark blue”, “navy” and “blue dark” must be normalized where necessary, otherwise double or unclear choices arise. Not every product has to have the same properties; the data that visitors use to compare must be especially reliable.
What happens if product characteristics are missing or inconsistent?
Products without the relevant property cannot be correctly classified within that filter. This can make a selection look incomplete or out of the picture a product. It is therefore wise to first check the main categories and purchasing characteristics. Filter usage can then show which missing data has the greatest influence, so that the team improves product data in a targeted manner.
Can visitors use multiple filters at once?
Yes. For example, a visitor can combine brand, size, color and price to make a targeted selection. Within one filter group, the logic may vary per situation: multiple brands can often be chosen side by side, while a maximum price works as a limit. The active choices must always remain visible and separately removable, so that the visitor understands why certain products are shown.
How do you prevent a combination of filters from giving zero-result searches?
By only offering values that still produce products within the current selection. Result numbers indicate in advance how many articles remain after a choice. If a combination cannot give a result, the option can be disabled or hidden. It is important that the visitor can easily reverse existing choices and not remain stuck in an empty selection.
How does Smart Filters work on mobile devices?
On mobile, filters are usually opened in a separate panel or an extendable layer. Key groups are at the top and long lists can be folded. Active filters and the expected number of results remain clearly visible. Large touch surfaces, simple labels and a fixed button to view results ensure that visitors do not get lost or accidentally lose their selection.
Can filters be combined with search intent and AI Search?
Yes. A natural search question can already contain preferences, such as “waterproof walking shoe size 43”. The search function uses that context to find relevant products; Smart Filters then helps the visitor to refine further by, for example, brand, price or weight. Both parts complement each other: search reduces the starting point and filters give the visitor control over the final selection.
How do you measure whether the filters work properly?
Take a look at the use per filter group, frequently chosen combinations, the number of remaining products, the removal of filters and click through to product pages. Also compare conversion and failure between sessions with and without filter use, without directly deducing a causal link. Remarkably little used may indicate an unclear label, a low position or a trait that is not relevant to visitors.
Do smart filters automatically help with conversion?
Good filters can shorten the route to appropriate products and make comparison easier, especially with large or technically complex assortments. That is not a guarantee of more sales. The result also depends on product range, prices, product information and the rest of the online store. Therefore, measure whether visitors view relevant products more often after filtering, add to the shopping cart and complete.
Does the existing filter structure need to be completely replaced?
Not always. Often the existing layout can serve as a starting point and be improved step by step. For example, start with categories where visitors filter a lot or where the current structure demonstrably leads to empty choices. After that, labels, sequence, product values and mobile operation can be adjusted and measured per part. This way, the change remains manageable and the team can learn from real use.