Start with the questions a customer needs to answer: where to go, when to travel and which requirements matter for the stay. Choose filters that support those decisions and write down what each one means in your inventory. This keeps the interface tied to the information the site can actually provide.

Use a small set of representative searches before extending the filter list. Include a broad search, a specific requirement and a combination that returns no matches. These scenarios reveal whether the controls help someone refine a decision or simply add more choices.

Agree the meaning of each filter

For a hotel facility filter, ask which supplier field supports the label and how missing information is handled. Keep “not supplied” separate from a verified absence in the working specification. This avoids making an unsupported promise from incomplete data.

Write one example for each important filter. Include the selected travellers and dates where they affect the offer. The example gives the supplier and website teams a shared case to review.

Inspect the addresses that filters create

Google describes faceted navigation as controls that change a list by attributes. It warns that parameter-based combinations can generate very large spaces of URLs, increasing crawling of pages whose usefulness has not been established. Source: Google: faceted navigation URLs

Ask the website team to show the addresses produced by your sample searches. Record which choices change the address and which combinations produce the same content. This makes the search discussion concrete without requiring the content team to design technical rules itself.

Keep customer usefulness and search discoverability as separate decisions. A filter may help someone use the site without needing a dedicated landing page in search.

Select the pages worth maintaining

For any filter combination proposed as a search landing page, write its audience and purpose. Check that the result is useful on its own and that someone is responsible for its content and availability. This gives you a reason to maintain the page beyond generating another address.

Use a finite review list at first. Compare those pages with existing destination and product guides so a new landing page adds a useful decision rather than repeating the same explanation.

Choose crawling controls with the technical team

Google’s faceted-navigation guidance sets out different approaches depending on whether the resulting URLs need to be crawled. It also explains that canonical signals may reduce crawling over time but are less effective for this purpose than its recommended controls. Source: Google: faceted navigation URLs

Do not apply a broad block without checking the important pages it would affect. Give the team your intended landing pages and sample unwanted combinations, then ask it to document and test the chosen policy.

A robots.txt rule controls crawling. Google warns that a blocked address can still appear in search, so it is not a substitute for an indexing decision or protection of private content. Source: Google: introduction to robots.txt

Test the customer’s recovery route

When no matches appear, check that the page explains the result and lets the customer remove a filter or change the search. Preserve the other useful choices where your design supports it. This gives the customer a clear next action instead of a dead end.

Repeat the tests on a narrow mobile screen and with keyboard navigation. Use the accessibility guide for a broader initial review of the interaction.

Measure the issue you actually have

Google’s detailed crawl-budget guidance is mainly for large or frequently changing sites. Do not assume every small travel website needs advanced crawl-budget work. Start with observed discovery or indexing problems and investigate specific addresses. Source: Google: crawl budget management

Keep the filter definitions, example searches and approved URL policy together. Revisit them when inventory or site behaviour changes, so the search setup continues to reflect the service customers are using.

Written and updated by Travelgenix. Sources checked on 16 October 2026. Worked examples and checklists are editorial illustrations and recommendations, not customer results.