How to Get Google to Index My Site: 6 Free Steps, in Order

How to get Google to index my site, in order

Indexing is not ranking. Google indexes a page in three stages: it discovers the address, crawls the page, and decides whether to store it. Only a stored page can appear in a search result, so a page that is not indexed cannot rank for anything. Every step below is free; the first five take between ten minutes and an afternoon, and the sixth is a monthly habit.

In Theo's audit of 5,974 small business websites, 48 percent of the 4,027 sites with search data got an estimated zero visitors a month from Google search; The State of Small Business Websites 2026 breaks that figure down by platform and industry. Some of those sites are indexed and rank for nothing; some are not indexed at all. The check below tells the two apart.

What does Google index mean

Google's index is the store of pages Google has fetched, read and decided to keep. A search looks through that store, not the live web. So a page is indexed when Google holds a stored copy of it, and not indexed when it holds none, whatever the reason.

The three stages fail differently. Discovery is Google learning that the address exists, from a link or a sitemap. Crawling is Googlebot fetching the page. Indexing is the decision to keep it. Ranking comes after all three: an indexed page can sit on page nine for years.

How to check whether Google has indexed a page, and when

Check first, because the answer decides which step matters. If nothing is indexed and Search Console has never been set up, start at step 1. If pages show a reason such as Discovered, currently not indexed, read the status table below; resubmitting will not change it. If the pages are indexed and simply rank low, indexing is not the problem: the guide to a website not showing up on Google covers the other reasons.

On a brand-new site, expect the clean case: URL Inspection says URL is not on Google and the reason reads URL is unknown to Google. Nothing is blocking the page; steps 1 to 3 are the entire fix. The checks, in order:

1. Type site: followed by the domain into Google, no space, as in site:yourdomain.com. Google returns the pages it has stored from that domain; add a path to check one page. The count is approximate.
2. In Search Console, paste the full page address into the URL Inspection box at the top. A green check beside URL is on Google means the page is indexed. Anything else names the reason, and the reason is the diagnosis.
3. To find out when a page was indexed, open the Page indexing panel in that same result. Google does not print an indexed-on date; the closest thing is Last crawl, the date and time Googlebot last fetched the page.
4. Open the Page indexing report in the left menu for the whole site: the count of indexed pages at the top, and below it every page that is not, grouped by reason.

Step 1: Add the domain to Search Console as a Domain property (15 minutes)

Search Console is Google's own report on the site: which pages it has indexed, which it has seen and skipped, and why. The sitemap submission and the indexing requests both happen inside it, so nothing else on this list works until this step is done. It is free and needs only a Google account. If the site has no domain of its own yet, the guide to getting a website on Google covers that first.

1. Open Search Console, choose Add property, and pick Domain. A Domain property covers the root, the www version, http and https in one place.
2. Verify the real domain, not a builder's preview address, because a property set up on a preview has to be redone once the domain is live. On a Theo site the owner refines the preview in chat, publishes it, then connects their own domain; Analytics and Search Console connect only after that domain is active.
3. Prove ownership. The screen shows a record to add at the registrar and names the record type; add it, give the registrar's change time to reach Google (usually within a few hours), and click Verify. If it fails, the change has not reached Google yet; try again later that day or the next.
4. Then leave it alone for a few days. The Page indexing report stays empty for a new property until Google has looked.

Step 2: Submit a sitemap, and keep its dates honest (10 minutes)

A sitemap is a plain list of every page address on the site, with the date each page last changed. Google's own page on asking it to recrawl a site names the sitemap as the method for many pages at once and says it is useful after a launch or a site move. Most builders generate the file; the owner's job is to submit it and keep it truthful.

1. Find the sitemap address, usually /sitemap.xml on the domain, and open it in a browser. It should list real page addresses, each one the version the site actually uses, with no test pages and no deleted pages.
2. In Search Console, open Sitemaps in the left menu, paste the address, and click Submit. The report shows the status (Success, Has errors, or Couldn't fetch), the date Google last read it, and how many addresses it found.
3. Keep the last-modified dates true. Google uses them only while they are consistently accurate, so the same date on every page teaches it to ignore them, and it ignores the priority and change-frequency fields altogether.
4. Keep one address per page. If a page also loads with a trailing slash, with .html, or under www, redirect every extra version to the one in the sitemap; the post on www versus non-www explains how to pick one.
5. Resubmit only after real changes: a batch of new pages, a set of rewritten ones, a move to a new domain.

Step 3: Request indexing once per page in URL Inspection (20 minutes a day)

URL Inspection has a Request Indexing button that asks Google to put one page in the crawl queue. Google describes it as the method for a few pages, available to an owner or full user of the property, and says there is a quota for submitting individual URLs; it does not publish the number. A request is a nudge, not an order.

Take a plumbing company site laid out like the demo below: click-to-call in the header, service pages for repair, drains and water heaters, service-area pages by town, and reviews and license details.

1. List the pages that earn money or answer a customer's question, in priority order: the homepage, each service page, the service-area pages, contact, about, then the strongest posts.
2. Inspect the first one, wait for the verdict, then click Request Indexing. Work down the list until Search Console reports the quota is reached, and pick up the next day where it stopped.
3. Request each page once. Google's guidance is that requesting a recrawl multiple times for the same URL will not get it crawled any faster, and that a request does not mean the page will be included in search results at all.
4. Write down the date. The crawl can take weeks, so read the report a week later, not an hour later. Last crawl is the first thing that changes.

Step 4: Link every page from a page already in the index (an afternoon)

Google decides what to crawl by predicted value, and on a small site the strongest everyday signal of value is a link from a page Google already visits. A page with no links pointing at it, or only a footer link, reads as unimportant. Most owners skip this step, and it is usually what moves the report.

• Link from the homepage copy, in a sentence, to each service and location page: the line about water heaters links to the water heater page. A footer list of every page is boilerplate; a link inside the text of a related page carries more meaning.
• Keep every page within three clicks of the homepage, and link each new page from two or three related pages already in the index, with a link back.
• Use the page's own words in the link. Water heater repair in Mesa tells Google what the page is before it fetches it; click here tells it nothing.
• Earn a few real links from outside: a supplier's dealer page, a trade association directory, the chamber of commerce, a local news story. One or two a month is a realistic pace.
• Make it a monthly job. The SEO maintenance checklist puts an internal-link pass and an indexing check on the monthly list, so a page added in March is not still orphaned in June.

Step 5: Clear the settings that tell Google to stay out (an hour, if any turn up)

URL Inspection names the block when there is one, and the status table below shows how each reads. These are settings a site picks up by accident, usually left over from a builder's staging mode, a redesign or a domain change. Each needs one Request Indexing after the fix.

1. Open yourdomain.com/robots.txt in a browser. A line reading Disallow: / beneath User-agent: * tells every crawler to stay out of the whole site; remove it, or narrow it to the one folder that should stay private.
2. Look for a noindex tag. Builders set it on hidden and draft pages, and on the whole site while it is unpublished, and a redesign often carries it over. It lives in the page's settings or in the page's head.
3. Check each page's canonical, the address that tells Google which version of a page is the real one. It should point at the page itself. A canonical left pointing at an old domain after a move sends every page's credit away.
4. Check the domain itself. A parked or forwarding domain sends Googlebot to the registrar's page instead of the site; turn off parking and forwarding at the registrar and point the domain at the site. The registrar's page keeps showing until the records update, usually within a few hours.

Step 6: Speed, fresh pages and a reason to come back (ongoing)

A crawl allowance is earned. Google spends more of it on a domain whose pages respond quickly and get used: linked, visited, and updated with something new. Page speed and fresh content are often listed as indexing steps; both help, though a fast site with no links still waits.

• Keep pages fast. A server that answers slowly gets fewer fetches per visit, so Googlebot sees less of the site each time. Compressed images, no unused scripts, and a host that serves static pages over HTTPS cover most of it; page speed is one of the technical items the ongoing search work on Starter and Pro improves.
• Publish fewer, more distinct pages. Twenty service-area pages that differ only by the town name look, from outside, like one page copied twenty times. Five town pages with real detail (the neighborhoods served, the jobs done there, a review from each) earn crawls.
• Add something worth fetching on a steady cadence: a page that answers a question customers ask on the phone, a post about a job, an updated prices page. Each new page linked from an indexed page gives Google a reason to return.

Sharing the site, Google Business Profile and analytics: what each does for indexing

Three more things owners are often told to do: share the site, set up a Google Business Profile, connect analytics. None of them puts a page in the index on its own.

• Sharing the site on social media brings people, not Googlebot. Links in social posts are usually marked so that search engines do not count them. What sharing does is put the site in front of someone who might link to it from a site of their own, and that link is what counts.
• A Google Business Profile does not index the website either, but it puts the business on the map and links to the site from a page Google runs. Theo's Boost checklist walks owners through Google Business Profile, Apple Business Connect, Bing Places, Yelp and the other listings that matter, and each listing is one more real link to the homepage.
• Google Analytics tells the owner when visitors begin; it does not affect whether Google indexes anything. Connect it once the domain is active, and read the Search Console Performance report, not Analytics, for impressions, which arrive weeks before clicks.

What Discovered, currently not indexed and the other Page indexing statuses mean

Google's own description of Discovered, currently not indexed: the page was found but not crawled yet, because Google expected the crawl to overload the site and rescheduled it. In practice it means Google knows the address and has decided, for now, that fetching it is not worth the trip. When many pages sit in Discovered at once, treat it as a verdict on the site: Google's spam policies describe doorway pages as pages created to rank for specific, similar searches, and give pages targeted at particular towns that funnel visitors to one page as an example; twenty service pages that differ only by town look like that from outside.

The first three statuses are the three stages of indexing; the rest are the blocks from step 5 and the leftovers of a move.

The two shortcuts that do not work: the Indexing API and paid indexing services

The Google Indexing API sounds like the answer: a way to tell Google about a page directly, with no queue. Google's documentation limits it to pages that carry job posting or live-stream video markup and says that attempts to abuse it may result in access being revoked, with its spam policies applying to anything submitted through it. A plumber's service page, a restaurant's menu, a law firm's homepage: none of these qualify, and the plugins that push every page through it rely on a loophole, not a feature.

Paid indexing services promise any page in Google within days, for a fee. The routes into Google's index are Search Console requests, a sitemap and links, and an owner already has all three for free. IndexNow, the protocol these services tend to cite, is accepted by Bing and some other search engines; Google does not take part in it. What is left is the way Googlebot has always found pages: following links from pages it already crawls, so the service places a link to the page on a network of pages Googlebot visits often. That can get the page fetched. It cannot get the page kept, which is Google's quality decision.

How long it takes Google to index a site, and when to stop waiting

Google's published range for a crawl after a request is a few days to a few weeks, with no promise that the page will be included at all. Budget six weeks before reading the Page indexing report as a verdict on the whole site. On a Theo site the submissions and the technical fixes are part of the ongoing search work on Starter and Pro; on any site the checks below take ten minutes a week.

A simple test for what kind of problem the site has: request indexing for the most important pages over a few days, then look at Last crawl a week later. If most were crawled, the queue was the problem and the rest will follow. If only a few were, Google is declining even on request, and the answer is the site itself: fewer pages, each more distinct, better linked.

What indexing looks like when it works: impressions first, then clicks

Impressions are the first thing to move. A page that has just been indexed starts appearing far down the results for a few searches, counted as impressions long before anyone clicks. So the order of good news in the Performance report is impressions, then average position, then clicks, and an owner watching only the click count gives up a month early.

Buy My Lunch, which takes school lunch orders online, came to Theo in June 2026. Over the next 90 days, as the growth work published pages, search impressions rose from 7,500 to 61,700 against the 90 days before, average position moved from 13 to 8, and clicks from Google search rose from 131 to 929. Those are the project's own Search Console numbers through September 27, 2026, shown with the owner's permission.

Mistakes that keep a site out of Google's index

Each of these shows up in Search Console once Google next crawls the page.

• Leaving each page's canonical pointing at the old domain after a move. The credit for every page flows to an address that no longer serves it; set every canonical to the page itself, then request the homepage once.
• Submitting a sitemap that still lists redirecting or deleted addresses. Google spends the visit collecting Page with redirect and Not found (404) verdicts instead of fetching live pages.
• Publishing a homepage that says only Coming soon. Google fetches it, files it as a Soft 404, and is slow to return for the finished version.
• Moving to a new domain or builder and dropping the old addresses. The pages Google trusted return 404 and the new ones start from zero; redirect the old addresses that still get traffic to the matching new pages, and read what a move does to rankings before the switch.

Frequently asked questions