What this checkpoint covers
The questions are set at Halcyon Ferry Holidays, an invented operator selling ferry holidays from several ports, with English editions for two markets, a French edition and booking offices in many towns. They draw on the faceted navigation guide, the hreflang guide and the guide to location pages and listings at scale.
Task
- Write a one-line answer and the action you would take for each question before reading on.
- Mark each answer confident or unsure.
- Compare with the explanations and note the signal you misread on every miss.
- Reread the linked section for any confident miss before starting the final module.
Questions
- Halcyon's holiday search lets visitors filter by destination region, pets allowed, cabin type, departure day and trip length. The example below shows two combinations. Which deserves a landing page, and what happens to the other?
- A developer proposes blocking all filter URLs in robots rules and adding a noindex to them as well, "to be safe". What will actually happen?
- The French edition lists the two English editions as alternates, but the English pages list only each other, and no group names a fallback for visitors who match no edition. What is wrong, and what is missing?
- After the second English edition moved to a new folder, its pages redirect correctly, but the hreflang sets on the other editions still name the old addresses. Separately, a template change made that edition's pages canonical to the first English edition. What does each problem do?
- The two English editions are identical apart from the currency symbol. Will hreflang keep both visible in their markets?
- Halcyon plans a page for every booking office, using one template where only the town name changes. What would you change before launch?
- Winter hours were updated on the website but not on the business profiles or the booking platform, and customers arrived at locked offices. What is the structural fix?
Halcyon Ferry Holidays: two filter combinations
| Combination | Query it matches | Monthly searches | Holidays listed |
|---|---|---|---|
| Brittany + pets allowed | pet friendly ferry holidays to Brittany | 1,300 | 46 |
| Brittany + sea-view cabin + Tuesday departures + 3 nights | none found | 0 | 2 |
Every filter selection currently creates a crawlable link, and the crawler already knows about tens of thousands of holiday search URLs for a programme of a few hundred holidays.
Answers explained
- The first becomes a landing page; the second stays out of the crawl. The first has demand, enough holidays and a distinct purpose, so it earns a clean URL, its own introduction and a sitemap entry. The deep combination has no demand and two results; its filter links should not be crawlable, as the faceted navigation guide sets out.
- The noindex is never seen. Robots rules stop crawling, so the engine cannot read the noindex, and blocked URLs linked from elsewhere may still be indexed without content. Choose per tier: crawl control for deep combinations, noindex or a Canonical TagTells search engines the preferred URL. for views the crawler will meet.
- Missing return links and no x-default. Annotations must be reciprocal: every edition lists every edition, itself included. Without return tags the pairs may be ignored. Add one x-default per group, pointing at a language chooser or the main edition, which is itself in the group.
- Both break the group. hreflangAnnotation for language/region versions. values must be final, indexable addresses, so old URLs that redirect are not valid targets; regenerate the sets from the live URL scheme. A canonical pointing at another edition tells the engine the page is a duplicate, which removes it from the group; every edition must be canonical to itself. The hreflang guide recommends generating and validating sets in the build.
- Not reliably. The annotation routes searchers; it does not make near-duplicates distinct. Give each edition what matters in its market: prices, departure ports, contact details, terms.
- Add a local layer, starting with the most valuable offices. Name-swapped pages help neither people nor search. Each page needs its own directions, services, team, opening times and reviews, a self-referencing canonical, and plain links from a crawlable office finder.
- One master record with owners. Hours, addresses and phone numbers are changed only in the master record, which feeds the website, the business profile (GBPLocal listing on Google Search and Maps.) and the booking platform. Add a change procedure for seasonal hours with a check afterwards, as the location pages and listings guide describes.
A learner with six or seven sound answers can decide index treatment by rule rather than URL by URL. Fewer means rereading the guides behind the misses.
Where to go next
If questions one or two caught you out, redo the fabric shop filter indexing exercise. Otherwise start the final module with the recipe publisher rankings drop exercise, and use the intermediate SEO level to refresh canonical and robots basics.
