Skip to content

Level 04 Lesson 01 of 3

Crawling and Indexing

Follow the search-bot delivery route from URL discovery through crawling and processing to indexing and ranking eligibility.

Learning objectiveFollow the search-bot delivery route from URL discovery through crawling and processing to indexing and ranking eligibility.

Estimated time9 min read

0 lessons completed0%

Current lesson · not completed

Your learning progress is stored only in this browser and is not sent anywhere.

The search-bot delivery route

A search system can discover and crawl a page without placing it in an index. Imagine a library intake desk: finding a book, requesting it, reading it and cataloguing it are separate jobs. Only catalogued candidates can be considered for a particular search; being catalogued does not guarantee a ranking.

URL intake pipeline
DiscoverFind the URL
RequestFetch if allowed
ProcessInterpret content
IndexStore as a candidate
ConsiderAssess for a search

Text equivalent: a system discovers a URL, may request it, processes a successful response, may index it, then may consider that indexed page for a specific query.

Three different checkpoints

Crawled

The bot requested the URL. The response might still be an error, duplicate or blocked from indexing.

Indexed

The system chose a version for its searchable catalogue. This is not a visibility promise.

Ranked

The indexed candidate was selected and ordered for a particular search, place, time and context.

How discovery and controls interact

Search engines can find pages through a descriptive internal link or an XML sitemap. A sitemap is a discovery aid, not an indexing guarantee. Robots.txt guides crawler access; it is not a dependable way to remove a URL already indexed. A noindex directive generally must remain crawlable so a crawler can see it. A canonical URL is a preferred-version hint, not a guaranteed command.

Blocked crawl + noindex

The crawler cannot fetch the page, so it may not see the noindex instruction.

200 response + thin duplicate

Fetch succeeded, but the system may choose a different canonical or not index it.

Orphan page

No internal route points to it; a sitemap can aid discovery, but cannot guarantee inclusion.

404 response

The address reports that the resource is missing and should not serve an indexable page.

Google-specific reference: compare the official guidance for robots.txt, noindex and canonicalisation. Other search engines can implement their own behaviour.

Related dictionary terms

Level checkpoint / Mission review

Check what you learned

Answer all three questions correctly, then complete the lesson. Retry any answer whenever you need.

Question 1 of 3What does a successful crawl prove?

Question 2 of 3Why should a page with noindex generally remain crawlable?

Question 3 of 3What is a canonical URL?

Complete all three questions to unlock lesson completion.

Your learning progress is stored only in this browser and is not sent anywhere.