Why your website isn't showing up on Google

Two completely different problems feel identical from where you are sitting: Google does not have your pages, or Google has them and nobody sees them. They need opposite responses, and most advice on this question never separates them.

This guide separates them first, then works through the actual causes in the order that costs you the least time. Every technical claim here is checked against Google’s own documentation, because the pages currently ranking for this question repeat at least four things Google has never said.

Before anything: two things that change the answer

You need a verified Search Console property. Every check below happens there, and if you have never set one up, that is step zero — add the site, verify it by DNS record or by uploading the file it gives you, and wait a day or two for data to appear. Search Console shows nothing retroactively, so the sooner it exists the better. Verifying it does not affect your rankings either way; it just means you can see what is happening.

Is this one page, or the whole site? They are different problems. A single new page missing is normal and usually resolves itself. An established site disappearing all at once points at something site-wide — a robots.txt change, a migration, an accidental setting. Check a few URLs before concluding anything.

Not indexed, or not ranking?

Check whether Google has your pages at all

Search Console → Indexing → Pages. A non-zero Indexed count means your site is in Google. Whatever else is wrong, it is not discovery.

For a single URL, use URL Inspection. It tells you whether the URL is on Google, when it was last crawled and by which crawler, its robots.txt and canonical status, and which pages link to it. It also offers a live test and a Request indexing button — that button is quota-limited, and pressing it repeatedly does not make anything happen faster.

Check whether you rank and simply get no clicks

Search Console → Performance → Queries, three-month window. Then read the shape of it:

• Impressions but almost no clicks → you rank. Your title and description are the problem, not your visibility.
• Impressions falling over time → you had something and lost it. Now look for a cause with a date attached.
• No impressions at all, ever → you never ranked for that term. Nothing broke.

That last case is the one people misdiagnose most expensively. A worked example: a two-month-old site with 158 indexed pages, no manual action and clean technical health had 233 queries with impression data over three months — and its own brand name appeared in none of them. There was no decline to investigate. The name had simply never been something Google associated with the site.

Start in Google Search Console

In this order, because each step makes the next one cheaper.

Manual actions and Google penalties

Security & Manual Actions → both should read “No issues detected”.

This is first because a manual penalty makes everything else irrelevant until it is resolved: a human at Google has made a decision about your site, and no amount of markup or content work overrides it. The common triggers are spam links you bought or swapped for, thin pages generated at scale, and hacked content you have not noticed.

It is also rare. Most people who suspect a penalty do not have one, and checking takes ten seconds — which is exactly why it goes first.

(If you still call it Google Webmaster Tools, this is the same product. It was renamed to Search Console in 2015.)

The Page indexing report

You want a non-zero indexed count and a curve that is flat or rising rather than falling off a shelf.

Read the “Why pages aren’t indexed” table underneath calmly. Redirects, noindex and canonical exclusions are usually things you asked for. The ones worth attention are Not found (404), Blocked by robots.txt, and the status below.

“Crawled — currently not indexed”

This one deserves its own explanation, because it is the most common status people misread and the only one that is a judgement rather than a fault.

It means Google came, read the page, and decided not to index it. Nothing is broken. No setting is wrong. Google’s systems prioritise “the fast inclusion of high quality, useful content”, and indexing is explicitly not guaranteed — Gary Illyes has said content still has to pass quality checks to get selected for indexing at all.

There is no button that fixes this and no word count that satisfies it. If a batch of your pages sits in this status, the useful questions are whether each page answers something a real person searches for, whether it says anything the pages already ranking do not, and whether anything on your own site links to it. Pages that exist only because a template generated them are the usual occupants.

A related status, “Discovered — currently not indexed”, means Google knows the URL exists but has not crawled it yet. That one is usually about crawl budget and internal linking rather than quality, and it often resolves on its own.

The URL Inspection tool

Use it on the specific page you expected to see. It distinguishes “Google has never seen this URL” from “Google has it and chose not to index it” — again, two different problems.

Your query data, the decisive test

Covered above, and it is the step people skip. Searching Google yourself is a sample size of one, personalised and localised. Your query data is what Google actually recorded.

What blocks search engines

noindex tags you did not mean to ship

Check the raw HTML and the rendered page. Google executes JavaScript, so a robots meta tag injected at runtime counts — and it will not appear in “view source”.

If you noindex your tag and author archives, which is reasonable, read the condition carefully:

if (/^\/(tag|author)\//.test(location.pathname)) { … }

That is anchored with ^, so it only matches paths that begin with /tag/ or /author/. Remove the anchor and it matches those letters anywhere in the URL — a post at /blog/steam-tags/ contains “tag”. One character separates correct hygiene from quietly removing part of your site.

Also check for an X-Robots-Tag in the HTTP headers. Same effect, invisible in the page source.

robots.txt and the discourage-search-engines switch

Confirm robots.txt does not disallow the paths you care about, and that it declares your sitemap.

If you are on WordPress, check Settings → Reading for “Discourage search engines from indexing this site”. It gets switched on during development and forgotten. Most site builders have an equivalent.

Duplicate content and canonical tags

There is no duplicate content penalty. Google has said so since 2008 and still says it. What actually happens is that Google clusters near-identical pages, picks one as canonical, and the others simply are not shown separately.

So duplicates do not get you punished — they get you consolidated, which can look identical from Search Console if the page you were watching was not the one Google picked. Set self-referencing canonicals so you choose, rather than letting Google guess.

You rank, just not where anyone looks

Search Console counts an impression when your page appears anywhere in the results — including page six. So “I have impressions but I cannot find myself” is not a contradiction. It is the most common version of this problem, and none of the technical checks above will touch it.

The keyword may simply be too competitive

Check your average position for the term. If it is 40, you are not invisible — you are on page four, behind sites with years of links and coverage. No technical fix moves that, because nothing is broken.

For a new site the honest response is to stop competing for the term and go after ones nobody is defending. A page on a low-competition question that reaches position 8 is worth more than a page on a flagship keyword that reaches position 45.

Your page may not match search intent

Google ranks pages that answer the query in the format the searcher expects. If everything on page one is a comparison table and you wrote an essay, you will rank below all of them regardless of how good the essay is.

The cheap diagnostic: search the term, look at the top five results, and ask what kind of page they are. Guide, list, tool, product page, forum thread. If yours is a different kind of thing, the mismatch is the problem, not your on-page SEO.

Title tags and meta descriptions decide the click

If your query data shows impressions and almost no clicks, you have already won the visibility fight and lost the next one. That is a title and description problem.

Google rewrites titles when it thinks its version serves the query better, so this is not fully in your control — but a title that names the thing the searcher typed, and a description that says what they will get, is the difference between a 0.3% and a 3% click-through. Fixing it takes minutes and does not require ranking any higher.

Indexed but invisible

If your pages are indexed and your query data shows nothing, the issue is that Google does not yet connect your name to your site as an entity — a thing in the world, rather than a string that happens to appear on a page.

What Google needs to connect a name to a site

Three ingredients: you state the name in a machine-readable way, someone else corroborates it, and time passes. You control the first completely, the second partly, the third not at all.

The markup that states your name

On your homepage, an Organization node. The property that earns its place is alternateName. If your brand is a compound word, your site says “Crit Pick” and people type “critpick”. Nothing else tells Google those are the same entity. Include the one-word form, the two-word form and the bare domain.

sameAs is the corroboration signal — links to profiles you control on platforms Google already trusts. It is how a name stops being a string and becomes something that exists in more than one place.

Then make everything agree: your <title>, your og:site_name, your profile bios. Inconsistent spelling dilutes the signal for free.

Verify the markup actually landed

Do not trust that it saved. On the live page, confirm the JSON-LD is in the delivered HTML, that every block parses — one trailing comma kills a whole block silently — and that there is exactly one WebSite node. Many platforms already output one, and a duplicate is worse than adding nothing.

Validate at validator.schema.org rather than the Rich Results Test. The latter only reports on features it can display, so it stays quiet about things it has no rich result for.

How Google finds your pages

Internal links do the discovery

Google finds pages by following links. A page with nothing pointing at it is a page Google has to discover some other way. This is documented, and it is the main reason a well-linked small site needs very little else.

One caveat on vocabulary: “internal links distribute authority” is SEO shorthand, not language Google uses in its current documentation. The discovery half is documented. The equity-flow half is a reasonable inference — treat it as such.

Broken internal links mean the path behind them discovers nothing. Worth fixing, but keep it in proportion: John Mueller stated in January 2026 that 404s and 410s are not a negative quality signal, and that it is how the web is supposed to work. The real cost is user experience, not punishment.

Why you probably do not need a sitemap

This contradicts most advice on this question, and it is Google’s own position: if your site is roughly 500 pages or fewer and internally well-linked, you likely do not need a sitemap at all. Google’s documentation says a sitemap “helps search engines discover URLs but doesn’t guarantee that all items will be crawled and indexed.”

Sitemaps genuinely earn their keep on large sites, brand-new sites with few external links, and media or news content. On a small, well-linked blog, submitting one is not the fix people expect it to be.

What does not cause this

The pages currently ranking for this question repeat several things Google has never said. If you are troubleshooting, these will waste your time.

The numbers that do exist

Core Web Vitals, at the 75th percentile of real user data.

And speed is a weak signal. Google’s own documentation says there is “no single page experience signal”, that passing Core Web Vitals does not guarantee ranking, and that it will show the most relevant content even when page experience is sub-par. If your site is invisible, speed is not why.

The related stat — “over 50% of users leave if a page takes more than 3 seconds” — is real but usually mis-cited. It is 53% of mobile visits, from a Google/DoubleClick study published in 2016 using data from early that year. Ten years old, mobile-only, and about abandonment during loading.

How long this takes

Google’s only published statement is that crawling “can take anywhere from a few days to a few weeks”, and that indexing is not guaranteed even after crawling. There is no official timeline beyond that.

The widely repeated “4 days to 6 months” figure has no primary source — it traces back through SEO blogs citing each other. Similarly, “only 5.7% of top-ranking pages were published within the year” is a misstatement of a 2017 Ahrefs study, which found that 5.7% of newly published pages reached the top 10 for at least one keyword within a year. Different claim, and nine years old.

The honest version: days to weeks for crawling, no guarantee of indexing, and for a genuinely new domain with no external links, months before search contributes anything meaningful.

Common questions

Sources: Google Search Central documentation — crawling and indexing, sitemaps, canonicalization, spam policies, page experience · web.dev, Core Web Vitals thresholds · Google Search Console Help, URL Inspection · Google/DoubleClick, “The Need for Mobile Speed”, 2016 · HubSpot, Marketing Benchmarks from 7,000+ Businesses, 2013 · Ahrefs, how long it takes to rank, 2017 · statements by John Mueller and Gary Illyes as reported by Search Engine Roundtable and Search Engine Journal.