SEO
How Does SEO Work?
SEO works because search engines discover pages, interpret what they are about, and rank them for a query against other pages that might answer the same job. Your work is to make that process easier for a URL that actually deserves to be found — not to hide a weak offer behind a denser keyword list. If you only need the definition, start with what SEO is; this page is the mechanism.
Discovery
Nothing ranks until a crawler finds the URL. Discovery happens when a bot follows an internal link, an external link, or a listing in an XML sitemap, or when you submit a URL in Search Console. Orphan pages — URLs linked from nowhere and omitted from the sitemap — are easy for Google to miss. If the money page is three clicks from home and never appears in the sitemap, you are hoping, not publishing.
Robots.txt can block a crawler from fetching a path. A noindex tag can allow a fetch and still keep the page out of the index. Those are different failures. A 404 or a soft 404 (a “not found” message served as a 200) tells the engine the URL is not a real page. Redirect chains, especially through HTTP then a second hop, waste crawl attention and confuse which URL is canonical. The practical checks live in the technical SEO guide: can Google fetch it, which URL is the one you want stored, and does the sitemap agree with the live site?
- Link every page you care about from a crawlable HTML link, not only from a JavaScript click handler.
- Keep the sitemap to indexable, canonical URLs you still want found.
- Inspect the live URL in Search Console before you rewrite the copy for the fifth time.
- Do not block CSS or JavaScript in robots.txt; the crawler needs those files to understand the page.
Understanding the page
After a fetch, the engine tries to understand the document: the visible text, the title, the headings, the links, images and their alt text, and structured data if you provided it honestly. It also renders the page well enough to see content that appears only after scripts run. If the offer lives only inside a canvas or a slider that never appears in the HTML, you are asking a crawler to guess.
Understanding is not the same as stuffing a phrase into every heading. The page should read as an answer to a job. A query like “emergency plumber in Memphis after hours” wants hours, a phone number, and whether you actually take that call — not a 2,000-word history of plumbing. Keyword research is how you list those jobs before you write. Duplicate or near-duplicate URLs make understanding worse: the engine has to pick a version, and you may not like the pick. Canonical tags, consistent internal links, and one URL per job reduce that mess.
Ranking for a query
Ranking is a comparison for one query at one moment. The engine estimates which indexed pages are likely to satisfy the searcher. Relevance, prominence, and usefulness all play a part. A page can be easy to crawl and still lose because another site answers the job faster, covers the location the searcher cares about, or is simply more trusted for that topic. You do not submit a “rank request.” You publish a page and the systems compare it.
Query interpretation moves. The same words can mean a how-to one month and a local map pack the next if enough searchers behave that way. Your page did not “lose SEO” overnight because a title character count changed. It may have lost because the results became more local, more commercial, or more documentary than your URL. Read the live results for your primary queries before you change the site. If the first page is a map and three directories, a national essay will not displace them.
| What you observe | What it usually means | What to check first |
|---|---|---|
| URL unknown to Search Console | Discovery or robots problem | Internal links, sitemap, robots.txt, inspection |
| Crawled but not indexed | Quality, duplication, or a noindex signal | Canonical, uniqueness, and whether the page has a job |
| Indexed with impressions, few clicks | Title and snippet fail to match the query | Title, description, and the first heading |
| Clicks but no enquiries | The page answers search and fails the sale | Offer, proof, and the next step on the page |
Why results move
Results move because the web moves. Competitors publish, you change a template, a CMS starts serving a different canonical, or Google adjusts how it interprets a class of queries. One-day swings are common. Look at weeks of Search Console data, not a screenshot from Tuesday night. A ranking change is a prompt to inspect, not a reason to buy a link package.
Technical accidents cause sudden drops more often than “the algorithm decided to hate you.” A sitewide noindex left in staging, an accidental block in robots.txt, a canonical that points every page to the home page, or a migration that 404s the old service URLs will look like an overnight disaster. Before you rewrite content, fetch the live headers and run URL inspection. Google’s spam policies also act on sites that scale thin or AI-generated pages, or that buy links meant to manipulate rankings. Those are avoidable problems, not mysteries.
What you actually control
You control the pages you publish, the links between them, whether the server returns the document, how honest the titles are, and whether a visitor can finish the job. You do not control other people’s sites, Google’s ranking systems, or a competitor who decides to do the work you postponed. An SEO strategy is a short list of those controllable items tied to a business outcome — booked calls, quote requests — not a fantasy calendar of fifty articles.
- Name the queries that already produce impressions or that sales already hears.
- Give each query one URL that can honestly serve it; merge leftovers.
- Make those URLs crawlable, indexable, and fast enough to use on a phone.
- Measure in Search Console and on the form or phone, then fix the weakest step.
That is the whole loop. When you want someone else to run the inspection and the page list, Scriplit’s SEO services are built around those steps. Send the domain and the two or three URLs that should earn work through the SEO contact form.
Frequently asked questions
What is crawling and indexing?
Crawling is discovering URLs. Indexing is storing a usable version of the page. A page that cannot be crawled or is marked noindex will not rank, no matter how good the copy is.
Do backlinks still matter?
Links from relevant, trusted sites can still help search engines understand prominence. Buying links to manipulate rankings is against Google’s spam policies and is not a substitute for a page that answers the query.
Why did my ranking change overnight?
Results shift because of competition, technical issues, query interpretation, and Google updates. One-day swings are common; look at weeks, not hours.