How to Do SEO for a New Website from Scratch

A new site has no links, no track record and no audience that searches for it by name, so the workable sequence is narrow and disciplined: lay a solid technical foundation, target specific long-tail searches instead of head terms, earn a handful of real links, and plan for a timeline that Google itself says can run to several months. The mistake to avoid is chasing competitive head terms a brand-new domain isn’t yet equipped to win, however good the content. Get the sequence right and the site has a realistic path; aim too high and it can produce excellent pages nobody finds.

The new-site reality

Search a competitive head term and look at who ranks. Where the results are held by sites with long link histories and a record on the subject, one thorough page on a new domain doesn’t yet carry what those pages carry. Accepting that separates a realistic plan from one that stalls. The early job isn’t to win the biggest terms; it is to win small, specific ones and build the record that makes bigger terms reachable later.

Technical foundation from day one

The technical baseline doesn’t set you apart, but missing any of it can cap what is built on it. At launch:

• Verify the site in Search Console and submit an XML sitemap. Google’s sitemap overview lists “Your site is new and has few external links to it” among the cases where you might need one.
• Serve everything over HTTPS.
• Make the site work well on phones. Google announced in October 2023 that the move to mobile-first indexing was complete. Google crawls and indexes as a smartphone, so the mobile page needs to be as complete as the desktop version.
• Use clean, readable URLs and a logical structure.
• Give each page a clear main heading and a unique, descriptive title and meta description. One H1 per page is our house rule for clarity, not a rule we have found in Google’s documentation.

None of this earns rankings by itself. It removes the technical reasons a page might fail to be crawled, rendered or indexed. Treat it as the floor, then put attention on content and links.

Indexing on a new site

New pages can sit in the Page Indexing report under “Discovered – currently not indexed” or “Crawled – currently not indexed” for a while. Google’s page indexing documentation defines the first as “found by Google, but not crawled yet,” and says that “typically” Google wanted to crawl the URL but expected the crawl to overload the site, so it rescheduled. The second means crawled but not indexed, which may or may not be indexed in the future. Neither is a penalty.

Crawl budget is a narrower question for a small new site than its reputation suggests. Google’s crawl budget guide is intended primarily for three kinds of sites: large sites (1 million+ unique pages) with content that changes weekly, sites of 10,000+ pages whose content changes daily, and sites with a large portion of their URLs classified as “Discovered – currently not indexed.” Google calls the page counts “a rough estimate,” not exact thresholds. A small new site enters that audience only through the third door: if a large share of your pages sits in Discovered, read the guide; if not, fix the discovery paths first.

What helps is giving Google clear paths to your pages. The same indexing documentation says that for Google to learn about a page, you must submit a sitemap or a crawl request, or Google must find a link to it somewhere. So:

• Link every important page from the homepage and main navigation, so nothing is orphaned. The sitemap overview defines proper linking as “all pages that you deem important can be reached through some form of navigation.”
• Keep the sitemap current.
• Earn a handful of real links from other sites, which give Google more routes to your pages.

Repeatedly requesting indexing for the same URL doesn’t speed things up. Google’s recrawl guidance says there’s a quota for submitting individual URLs and that repeated requests for the same URL “won’t get it crawled any faster.”

Keyword strategy: read the results, target the long tail

For a new site, target specific, long-tail searches where the current results are weak. The test needs no paid tool: search the exact query and read the first page.

• Major brands and established publishers fill the page: a new domain doesn’t yet carry what those pages carry. Set that term aside for now.
• Forum threads, thin posts and outdated pages fill the page: the search may be winnable, and a better page can rank.

That habit drives the content plan. Turn broad article ideas into specific angles where competition is thinner. A broad guide to a competitive topic can be out of reach for a new site for now; a precise answer to a narrow version of it, where the results are weak, is within range. Build the first content list from those, then widen toward broader terms as the site’s record grows.

Early links and a known name

Early links are about discovery and credibility, not volume. Pursue real relationships and placements, never bought or spammy ones:

• Relationships with relevant sites, partners and communities you are part of.
• Legitimate directory and profile listings where they fit your business.
• A useful resource others can cite.

Searches for your name are the searches you are best placed to win early, provided the name is distinctive: a person typing your business name is looking for you, and they arrive already intending to find you. A name that is also a dictionary word faces a different problem. Give your early audience reasons to look you up.

Topical focus and a realistic timeline

Concentrating on one tight topic gives a new site depth where it can be judged, while scattering across unrelated subjects spreads the same effort thin. A new site that publishes deeply in one well-defined area shows what it is about. Pick a focus and go deep before broadening.

On the “sandbox” question: we haven’t found Google documentation describing a filter that holds new sites back for a fixed period, so don’t plan around one. What is real is that a new site lacks the links and record that support steady rankings, which can produce slow early traction whatever you call it.

Google’s SEO Starter Guide says every change takes time to be reflected on Google’s end: “Some changes might take effect in a few hours, others could take several months.” Plan in months, expect the first rankings on strong, specific pages before meaningful traffic, and commit to the long arc.

Measure the right early signals

Much of a new site’s early disappointment comes from the wrong scoreboard, not only from slow rankings. Organic traffic can stay near zero for a long stretch on a new domain, so the traffic line says little early on. Watch the leading signals instead:

• Pages moving into the indexed count in the Page Indexing report.
• Impressions appearing in the Search Console Performance report, before clicks do.
• Average position improving on your specific long-tail targets.

Treat clicks and traffic as lagging outcomes that can follow once those move. A page that earns impressions but no position gains may have an intent or relevance problem to fix before you write the next one.

Frequently asked questions

Why are my new pages stuck as “Discovered – currently not indexed”?

Google defines it as found but not crawled yet, and says that “typically” the crawl was rescheduled because it was expected to overload the site. Make sure every page is linked from the homepage or navigation, keep the sitemap current, and earn a handful of real links; repeated indexing requests for the same URL don’t speed it up. If a large share of your URLs stays in that status, Google’s crawl budget guide lists your site among the ones it is written for.

Should I target the keywords with the highest search volume?

Not on a new site. Where head terms are held by established sites, a new domain doesn’t carry their links and record yet. Read the results for each target, pursue searches where the top results are weak, and leave the brand-dominated terms for later.

Is the Google sandbox real?

We haven’t found Google documentation describing one. New sites lack links and a record, which can produce slow early traction that looks like a sandbox. The response is the same either way: build a focused topic, earn real links and give it time.