How long does it take a new SaaS site to show up on Google?

A new domain with no backlinks usually takes three to four weeks to get indexed on its own. You can shortcut the important pages: submit your sitemap in Search Console, then use the URL Inspection tool to request indexing on each priority URL, which typically triggers a crawl within 24 hours. Indexing is not ranking. Getting crawled is step one; earning a position on a query takes content and time on top of a clean technical floor.

Does my SaaS need an llms.txt file?

It is worth adding, with low expectations. llms.txt is cheap to write and does no harm, but as of early 2026 no major AI company has confirmed its production systems read it, and real-world request volume for the file is close to zero. The work that actually earns AI citations is server-rendered HTML, clean structured data, and answers placed near the top of the page. Add llms.txt as a small bet, not as the thing that gets you cited.

What is the most common technical SEO mistake at a SaaS launch?

A site-wide noindex tag that shipped from staging and was never removed in production. It tells Google to drop every page, and teams often miss it for weeks because the site looks normal to humans. Check the meta robots tag on your live pages before anything else. The second most common is a canonical tag pointing every URL at the homepage, which collapses the whole site into one indexed page.

Does FAQ schema still do anything after Google removed FAQ rich results?

Google stopped showing FAQ rich results in search on 7 May 2026, so FAQPage markup no longer paints an expandable accordion under your listing. It is still a valid schema.org type and it will not harm your rankings. It also still helps AI engines parse question-and-answer content, which is the reason to keep it. The SERP feature is gone, but the markup keeps a job.

Technical SEO checklist for a B2B SaaS launch in 2026

graphs of performance analytics on a laptop screen

The technical SEO floor is the set of checks a B2B SaaS site has to pass before launch so search engines and AI engines can find it, render it, and cite it. It sits below the growth work: content, links, positioning. Get the floor wrong and none of that growth work registers, because Google never sees the pages cleanly in the first place.

Most SaaS teams reach this checklist late. The product ships, marketing writes the launch post, and three weeks later someone notices the site has zero impressions in Search Console. We run these eight checks before a domain goes public, because each one is cheap to fix pre-launch and expensive to retrofit after a bad first crawl. Here is the floor, in the order that matters.

What the floor measures

Four things decide whether a new SaaS site competes at all. A crawler can reach the page. It can read the content without running your JavaScript. It can understand the structure. It trusts that this is the canonical page. Every check below maps to one of those four. None of them is about keywords or backlinks. That work comes after the floor holds.

1. Serve content as crawlable HTML

Googlebot renders JavaScript, but it does so slowly and on a second pass that can lag days behind the first crawl. AI crawlers mostly do not render JavaScript at all. If your marketing pages ship as an empty div that React fills in the browser, an AI engine reads a blank page. The bar: every word you want indexed is present in the raw HTML response, before any client-side script runs. View source on each key route and check the text is there. For a Next.js SaaS this means server components or static generation for marketing and blog routes, not client-only rendering.

2. Pass Core Web Vitals at the 75th percentile

Core Web Vitals are three field metrics Google measures from real Chrome users, at the 75th percentile. The 2026 thresholds: Largest Contentful Paint under 2.5 seconds, Interaction to Next Paint under 200 milliseconds, Cumulative Layout Shift under 0.1 (web.dev). INP replaced First Input Delay in March 2024 and it is now the most commonly failed of the three: roughly 43% of sites miss the 200 ms bar. B2B SaaS sites fail it for a specific reason. Marketing stacks pile weight onto the main thread: HubSpot tracking, an Intercom or Drift widget, Segment, a tag manager carrying 40-plus tags, an A/B testing script that blocks render. Each one adds latency to the first click. The average B2B mobile LCP sits near 7 seconds, almost three times the threshold. Audit the third-party scripts before launch, defer what you can, and measure on a mid-range phone, not your laptop. We wrote the full Next.js playbook in how to ship Core Web Vitals in green.

3. Ship an XML sitemap and submit it

A sitemap does not guarantee indexing and it does not lift rankings. It does tell Google which URLs you consider canonical and when they last changed, which speeds discovery on a domain with no backlinks yet. Generate it from your published content, not by hand, so it never drifts out of sync. Submit it in Google Search Console and Bing Webmaster Tools on day one. For the pages you care about most, use the URL Inspection tool and request indexing directly: that usually triggers a crawl within 24 hours instead of the three to four weeks a new site waits otherwise.

4. Set canonical tags and kill the accidental noindex

The single most common launch disaster: the staging site shipped with a site-wide noindex meta tag, and nobody removed it in production. Check it first. Every indexable page needs a self-referencing canonical tag, no stray noindex, HTTPS enforced, and the staging environment locked behind auth so Google never indexes a duplicate. One mislabeled canonical pointing every page at the homepage can erase a whole site from search. These are five-minute checks that save months.

5. Emit structured data as JSON-LD

Structured data is the machine-readable layer that tells search and AI engines what your pages are. In 2026 it does more than unlock rich results: it is a primary signal for whether AI Overviews, Perplexity, ChatGPT, and Claude cite you or a competitor. Use JSON-LD, the only format Google recommends. For a SaaS site the useful types are Organization (site-wide, with your name, logo, and social profiles), SoftwareApplication (for the product, with category and pricing), and Article or BlogPosting on every post. One caveat worth knowing: Google removed FAQ rich results on 7 May 2026, so FAQPage markup no longer paints an expandable result, though it stays a valid type and still feeds AI engines. We break down the types in the 7 schema markup types every blog needs.

6. Get hreflang right if you ship more than one language

A multi-locale SaaS site that gets hreflang wrong shows the Spanish page to English searchers and splits its own ranking signal across duplicates. Each localized URL needs an hreflang cluster listing every language variant, including a self-reference and an x-default. Keep the canonical tag pointing at the page itself, not at the English version. If you run one language, skip this check entirely: a single wrong hreflang tag does more damage than none. This is the floor most international SaaS sites trip on after launch.

7. Open robots.txt to the crawlers that matter

robots.txt governs access. Get it wrong and nothing else here matters, because the crawler never arrives. Two failures are common. The first: a Cloudflare or host default that blocks AI crawlers wholesale, so ChatGPT and Claude cannot read pages you want cited. The second: a leftover Disallow from staging that blocks Googlebot from a whole directory. Training crawlers (GPTBot, ClaudeBot) run separately from AI-search crawlers (OAI-SearchBot, Claude-SearchBot), so you can opt out of model training while staying eligible for AI-search citations. For a SaaS site chasing visibility, the usual call is to leave search crawlers open. We covered the Cloudflare default in Cloudflare blocks AI crawlers by default.

8. Add llms.txt, with realistic expectations

llms.txt is a Markdown file at your site root that lists your key pages and what they cover, the way an XML sitemap does for search crawlers. It costs an hour to write and it does no harm. Be honest about the payoff: as of early 2026, no major AI company has confirmed its production systems read llms.txt, and one study of 500 million AI bot visits over 90 days found only 408 requests for the file. Treat it as a cheap bet, not a lever. The thing that actually earns AI citations is the rest of this floor plus content an engine can extract: clean HTML, real structure, the answer near the top. We weighed it up in llms.txt: what it is and if you need one.

The order to run these in

If you have one afternoon before launch, spend it like this. Check for the accidental noindex and the canonical tags first; they are the fastest way to erase a site and the fastest to fix. Confirm the content renders in raw HTML second, because a blank page beats nothing else in reverse. Open robots.txt third. Then sitemap, structured data, Core Web Vitals, hreflang, and llms.txt. The first three are pass-or-fail gates. The rest are graded: you want them strong, but a site can launch and improve them in week one.

The payoff is measurable. A SaaS site that clears this floor gets crawled within a day of submitting URLs, shows up in Search Console with impressions inside a week, and stays eligible for the AI citations that, by current data, convert several times better than ordinary organic clicks. A site that skips it can sit invisible for months while the team wonders why the launch post got no traffic. The floor is boring. It is also the cheapest SEO work you will ever do, because every item is faster to fix before launch than after.

Sources

Photo by Luke Chesser ↗ on Unsplash ↗

Frequently asked questions

How long does it take a new SaaS site to show up on Google?: A new domain with no backlinks usually takes three to four weeks to get indexed on its own. You can shortcut the important pages: submit your sitemap in Search Console, then use the URL Inspection tool to request indexing on each priority URL, which typically triggers a crawl within 24 hours. Indexing is not ranking. Getting crawled is step one; earning a position on a query takes content and time on top of a clean technical floor.
Does my SaaS need an llms.txt file?: It is worth adding, with low expectations. llms.txt is cheap to write and does no harm, but as of early 2026 no major AI company has confirmed its production systems read it, and real-world request volume for the file is close to zero. The work that actually earns AI citations is server-rendered HTML, clean structured data, and answers placed near the top of the page. Add llms.txt as a small bet, not as the thing that gets you cited.
What is the most common technical SEO mistake at a SaaS launch?: A site-wide noindex tag that shipped from staging and was never removed in production. It tells Google to drop every page, and teams often miss it for weeks because the site looks normal to humans. Check the meta robots tag on your live pages before anything else. The second most common is a canonical tag pointing every URL at the homepage, which collapses the whole site into one indexed page.
Does FAQ schema still do anything after Google removed FAQ rich results?: Google stopped showing FAQ rich results in search on 7 May 2026, so FAQPage markup no longer paints an expandable accordion under your listing. It is still a valid schema.org type and it will not harm your rankings. It also still helps AI engines parse question-and-answer content, which is the reason to keep it. The SERP feature is gone, but the markup keeps a job.