Submitting a Website to Search Engines: What Still Works in 2026

The top result for this query is usually a page offering to submit your site to fifty search engines, and roughly forty-six of those no longer exist as independent indexes. There are two search indexes that matter for almost everyone, and both take about ten minutes to set up properly.

Submission was a meaningful activity in 1998, when directories were how discovery worked. It has not been since. What replaced it is a small set of concrete steps: make the site crawlable, publish a sitemap, verify ownership in two consoles, and optionally ping an instant-indexing API. That is the whole list, and everything else on those fifty-engine pages is noise.

Table of contents

• Two indexes, not fifty
• Step one: be crawlable at all
• Step two: sitemap and the two consoles
• Step three: IndexNow, for when new pages should be found today
• What actually determines whether you get indexed
• How this fits the rest of the stack
• FAQ

Two indexes, not fifty

Almost every search interface people actually use is served by one of two crawls. Google runs its own. Microsoft runs Bing, which also supplies results to a number of other front ends including several privacy-focused ones. There are genuine independent indexes beyond these, and they represent a rounding error in most traffic reports.

So the practical target is: be properly indexed in Google and Bing. Do that and you have covered the overwhelming majority of the search traffic available to you. Everything else is optional and most of it is a waste of an afternoon.

The submission-directory pages persist because they rank for this query, not because they work. Submitting to a directory that no human visits produces a link from a page with no authority, which does nothing, and occasionally produces one from a link farm, which is mildly negative.

Ignore them. The real steps follow.

Step one: be crawlable at all

Before submitting anything, verify the site can actually be crawled. A meaningful share of indexing problems are self-inflicted and were shipped along with the site.

1. Check robots.txt. A staging Disallow: / that survived the launch is the classic. Fetch the file and read it rather than assuming.
2. Check for noindex. Both the meta tag and the X-Robots-Tag header. A header-level noindex is invisible in page source and catches people out repeatedly — request the page with a tool that shows response headers.
3. Check that content renders without JavaScript. Crawlers do execute JavaScript, but rendering is deferred and less reliable than HTML. If the page is empty with scripting disabled, indexing will be slower and patchier than it needs to be.
4. Check canonical tags. A canonical pointing at the wrong URL — or at a staging domain — tells search engines to index something else instead.
5. Check the site returns 200 and is reachable. Including from outside your own network, and on both the apex and www.

Fix anything here before submitting, because submitting a page that says do not index me achieves nothing and consumes the crawl budget you were trying to attract.

Step two: sitemap and the two consoles

A sitemap is an XML file listing your canonical URLs with last-modified dates. It does not guarantee indexing; it tells crawlers what exists and when it changed, which materially speeds up discovery on a new site with few inbound links.

• Generate it from the build rather than by hand, so it cannot go stale. Every static site generator and CMS does this, often with a single plugin or integration.
• Include only canonical, indexable URLs that return 200. Redirects, 404s and noindex pages in a sitemap are a quality signal against you.
• Reference it from robots.txt with a Sitemap: line, which is how crawlers find it without being told.
• Split it if you exceed 50,000 URLs or 50MB, using a sitemap index.

Then verify ownership in Google Search Console and Bing Webmaster Tools. Verification is a DNS record, an HTML file, or a meta tag — DNS is the most durable because it survives redesigns and host changes. Submit the sitemap URL in each.

Bing Webmaster Tools will offer to import your Google Search Console configuration, which is worth accepting: it copies verification and sitemaps in one step and saves doing everything twice.

The consoles are worth more after this than during it. They are where you find out that a page you thought was indexed is not, and why — which is information you cannot get any other way.

Step three: IndexNow, for when new pages should be found today

IndexNow is a simple protocol for telling participating search engines that a URL has changed, rather than waiting for the next crawl. Bing implements it, along with Yandex, Seznam and Naver. Google does not participate.

Setup is genuinely small:

1. Generate a key — a hex string of reasonable length.
2. Host it at https://yourdomain.com/<key>.txt, with the file’s contents being the key itself. This proves you control the domain.
3. POST a JSON body containing your host, the key, the key location and a list of changed URLs to an IndexNow endpoint.
4. Wire that POST into your publish process so it happens automatically.

A single submission can carry up to ten thousand URLs, and the participating engines share submissions with each other, so one POST reaches all of them. The response is immediate and tells you whether the payload was accepted.

For Google there is an Indexing API, but it is officially scoped to job postings and livestream structured data. Using it for ordinary pages is common and is a hint rather than a supported guarantee. The reliable Google path remains a good sitemap and pages worth crawling.

One caution on expectations: accepted is not indexed. Every one of these mechanisms tells a search engine that a URL exists. Whether it gets crawled, and whether it gets indexed after crawling, is the engine’s decision based on whether the page appears worth having.

What actually determines whether you get indexed

Submission solves discovery. It does not solve the question search engines are really asking, which is whether this page is worth storing and showing.

The things that move that decision:

• At least a few inbound links from pages that are themselves indexed. This is still the strongest discovery and quality signal there is. A site with no inbound links at all is a hard case regardless of how well you submitted it.
• Content that is not a near-duplicate of something already indexed. Thin or templated pages get crawled and dropped, and a site with many of them gets crawled less over time.
• Internal linking. Orphan pages reachable only from the sitemap are discovered late and treated as less important. Every page should be reachable by clicking from the homepage.
• A site that responds quickly and consistently. Crawl rate adapts to what your server can handle. A slow or erratic origin gets crawled less.

For a brand-new site, expect days to weeks for meaningful indexing even with everything correct. If a page is still missing after a few weeks, the console will usually tell you the specific reason — and the reason is almost never that you failed to submit it to enough directories.

How this fits the rest of the stack

Crawl rate adapts to how quickly and reliably your origin answers, which makes indexing partly a hosting question — a slow or flaky server gets crawled less, and pages that are not crawled are not ranked. The RunxBuild hosting calculator covers what a fast setup costs across the service, database, storage and bandwidth. A static build served from the edge answers quickly and consistently by construction; RunxBuild builds one from your repository, with headers and redirects as configuration so canonical handling is part of the deploy.

Useful related references:

• n8n + Qdrant: A Vector Search Node for Real Workflows
• The Jira REST API: Authentication, Search, and the Pagination Trap
• Vercel Database: The Marketplace, the Built-In KV, and the Question the Search Results Do Not Answer
• Services on RunxBuild

FAQ

Do I still need to submit my website to search engines?

Not to directories — those are a relic. What you do need is a sitemap, ownership verification in Google Search Console and Bing Webmaster Tools, and a crawlable site. That covers effectively all the search traffic available, and takes about ten minutes.

How long does it take for a new website to get indexed?

Days to weeks with everything configured correctly, and longer for a site with no inbound links. Submission speeds up discovery but does not guarantee indexing — search engines still decide whether a page is worth storing after they have crawled it.

What is IndexNow and should I use it?

A protocol for notifying search engines that URLs have changed, implemented by Bing, Yandex, Seznam and Naver but not Google. Setup is a key file hosted on your domain and a JSON POST per publish. It is worth wiring into your publish process — one submission reaches all participating engines.

Why is my page not getting indexed?

Check the obvious blockers first: a robots.txt Disallow, a noindex meta tag or X-Robots-Tag header, or a canonical pointing elsewhere. If none apply, it is usually a quality or discovery issue — no inbound links, thin content, or the page being reachable only from the sitemap. Search Console will normally state the specific reason.

Does submitting a sitemap improve rankings?

No. A sitemap helps discovery, which matters most for new sites and large sites with weak internal linking. It has no direct effect on ranking. Including redirects, 404s or noindex pages in it actively works against you as a quality signal.