Search or filter. Each term gets a plain definition first, and the part that actually costs money underneath.
How much of your site a search engine will fetch in a given period. It matters on sites with tens of thousands of URLs and almost never on sites with fifty.
Google combines crawl capacity (how hard it can hit your server without slowing it) with crawl demand (how much it wants your pages). Faceted navigation and infinite parameter URLs burn it fastest.
Read the lessonStoring a fetched page so it can be returned in results. Crawled and indexed are two different states, and a page can be one without the other.
In Search Console, «Crawled - currently not indexed» means the page was fetched and judged not worth keeping. That is a quality and linking signal, not a bug.
A directive that tells search engines not to keep this page in their index. Set in a meta tag or an X-Robots-Tag header.
A page disallowed in robots.txt can never be read, so a noindex inside it is never seen. Blocking and de-indexing are opposite instructions and combining them backfires.
A hint saying which URL is the master version when several show the same content.
It is a hint, not an order. If your internal links, sitemap and canonical disagree, the search engine picks for you, and it often picks the wrong one.
A file at the root of the domain listing paths crawlers should not fetch.
It controls fetching, not indexing. A blocked URL can still appear in results with no description if enough pages link to it.
A machine readable list of URLs you want found, with last modified dates.
It speeds up discovery on large or poorly linked sites. It does not make a page rank and it does not replace internal links.
Links from one page of your site to another. The ones inside body text carry information; the ones in header and footer are navigation.
We measured our own network: an average page had 8.2 inbound internal links, but only 2.6 came from inside text. The other 68% was template.
Read the lessonA page nothing links to. The stricter and more useful definition: a page with no contextual inbound links.
A page reachable only through the footer is treated as furniture. On our own site the blog index had fifteen inbound links and not one from a sentence.
The clickable words of a link. They describe the destination to both people and search engines.
When one page links twice to the same target, normally only the first link and its anchor count. Your navigation link usually wins over the good one in your paragraph.
A broad page that links to and from several specific pages on the same subject.
The linking is the mechanism, not the word count. A four thousand word page with nothing linking to it is a long page, not a pillar.
A group of pages answering neighbouring questions, linked to each other and to a hub.
Publish the cluster together rather than one page a month. The interlinking is most of the effect and it only works once the pages exist.
Two or more of your own pages competing for the same query, so the search engine picks one and usually the weaker one.
Search site:yourdomain.com plus the term. If more than one of your pages appears, decide which wins and point the others at it.
Everything on a results page that is not a plain organic link: featured snippets, People Also Ask, videos, maps, shopping.
They take space above organic results. You can be first in organic and seventh on the screen, which is why traffic falls while rank stays flat.
Read the lessonAn answer lifted from a page and shown above the results.
It can raise visibility and remove the click at the same time, because the answer is already on the screen. Worth winning for terms where the next step needs your page.
What the person actually wanted: to learn, to compare, to buy, or to reach a specific site.
The results page tells you the accepted answer format. If every result is a comparison table, an essay will not win however good it is.
Experience, Expertise, Authoritativeness and Trust. Concepts from Google's quality rater guidelines.
Not a score in the algorithm. It describes what raters are asked to look for, which is why named authors, sources and dates matter.
A broad change to ranking systems, announced by Google, usually rolling out over one to three weeks.
Recovery is rarely a single fix. Comparing your own traffic to the rollout dates is the only way to tell an update from a seasonal dip.
How many times your page appeared in results, whether or not anybody clicked.
Impressions on terms you did not previously appear for is the earliest honest signal that new content worked. It moves before traffic does.
Clicks divided by impressions. On a results page it measures whether your title and description earn the click.
Compare branded and non branded separately. Mixing them produces an average that describes nothing.
The mean position of your page across the impressions it received.
It bounces daily and it averages across queries and countries. Reacting to weekly movement is how good pages get deleted.
Machine readable markup, usually JSON-LD, describing what a page is: an article, a product, a FAQ.
It can earn a richer result. It does not directly raise rankings, and marking up content the reader cannot see is a policy violation.
Google's measured set for loading, interactivity and layout stability: LCP, INP and CLS.
They are a small ranking input and a large usability one. The honest reason to fix them is that slow pages lose people before content matters.
Markup telling search engines which language and region version of a page to serve.
Every version must point at every other version including itself. One missing return link and the whole set is ignored.
A link attribute asking search engines not to pass ranking signals.
Since March 2020 Google treats it as a hint rather than a rule, so its value is unknown and greater than zero. Paid links should be marked sponsored.
Read the lessonA link from another site to yours.
Judge a placement by whether real people read the page. A link on a page with no readers sends you nothing, whatever its attribute says.
A permanent instruction that a URL has moved, sending both people and crawlers to the new address.
Chains of redirects lose time and clarity. Redirect straight to the final URL rather than through three intermediate ones.
Nothing matches that. Try a shorter word.
Name the term and your page. We reply with what to check first and why. No cost, no call.