How discovery works
Before a search engine can crawl a page it has to learn that the address exists. That first step is discovery, and it happens in a handful of ways: a link from a page the crawler already knows, an entry in an XML sitemap it has been given, a redirect pointing at the new address, a submission through Search Console or a protocol such as IndexNow, and occasionally a mention picked up somewhere else on the web.
Discovery is not the same as crawling, and neither is the same as indexing. A URL can be known and never fetched, fetched and never stored, or stored and never shown for anything anyone types. Each stage fails for its own reasons, which is why Search Console reports them separately rather than as one status.
The word is also used for two unrelated things — Google Discover, the personalised feed on mobile, and the retired Discovery ad campaign type — so check which one a report means before drawing any conclusion from it.
Why discovery matters
It is the cheapest problem in SEO to fix and the most damaging to ignore. A page nobody has told the search engine about cannot rank at any price, and no amount of content work changes that. New sites, newly launched sections and pages buried behind filters are the usual casualties.
For an ecommerce catalogue it is a structural question rather than a one-off task. Products reachable only through a faceted filter, or only from a paginated list several clicks deep, are found slowly if at all, while the same products linked from a category page are found quickly.
Where discovery goes wrong
The most common cause is a page nothing links to — an orphan page that exists in the content management system and nowhere in the navigation. Publishing it and waiting is not a plan.
Close behind are sitemaps generated once and never updated, sitemaps full of addresses that redirect or return errors, and navigation built only in JavaScript so that no real link ever appears in the HTML. Blocking a section in robots.txt also stops discovery of everything that would have been reached through it, including pages you wanted found.
How to act on it
Link to a new page from somewhere already crawled regularly — a category page, the blog index, a relevant older article — on the day it goes live. Internal links do more for discovery than any submission tool, because they also tell the crawler roughly how important the page is meant to be.
Keep an XML sitemap that is generated automatically, lists only indexable pages that return a success status, and is submitted in Search Console. Then use the URL Inspection tool to request indexing for genuinely important new pages rather than for everything you publish. Where a whole section is being missed, the cause is usually structural, and that is technical SEO work rather than a content fix.