Skip to content
Book a Growth Audit
GrowthFlex · Growth Notes

Why Google Sees an Empty Page on Your Vue Marketing Site and What to Check First

For SaaS teams whose Vue marketing pages render in the browser but show as an empty shell to Google: how to check initial HTML, choose prerendering vs SSR vs CSR, and what crawl evidence to prepare before asking for help.

By GrowthFlex
SEO & Digital Marketing

Brand promotion: this article introduces Guangsuan's SEO and website services; it is not an independent review.

If your Vue marketing pages look complete in a browser but Google sees an empty shell, the problem is not Vue itself. It is that the text, links, titles or structured data that matter for ranking are missing from the HTML the server returns and only appear after client-side JavaScript runs. When that happens, check the server response first, then choose a rendering delivery that puts core content in that response, then collect crawl evidence before contacting hosting or SEO support.

First check: does the initial HTML contain the body text?

Do not judge crawlability from your own browser view. A browser runs scripts, keeps login state, and remembers prior selections. Google fetches the URL, checks crawl permission, then queues the page for rendering as a separate step. That sequence means an empty first response plus a delayed render leaves room for missing content, blocked resources, or conflicting signals.

For each important marketing URL, check the same three items in the same order:

  1. Open the delivered HTML, not the rendered view. Use View Source, save the response with a simple request, or reload with JavaScript disabled. Search that HTML for one product sentence, the H1 text, and one internal link you expect visitors to follow. If they are absent there, the page depends on client rendering for its core.
  2. Check the four Vue failure points. Hash-only paths such as #/page are hard to treat as separate URLs, so use real paths for any page you want ranked. Body text loaded by an API call after mount may leave only a shell in the first response, so keep primary content in the server response. Titles and meta set only by client navigation may leave the crawled page with the default value, so set per-route title and meta at render time. Structured data added after an API response may be missed, so emit it in the rendered HTML.
  3. Check discovery and permission signals on the same URL. Confirm the page has a real unique URL, returns a normal public response, is not blocked for that path by robots.txt, does not carry a noindex that was left from a preview build, and declares the intended canonical. Internal links to the page should use normal a href elements; links present in the initial HTML are found earlier and more reliably than links injected only by script.

The following teaching example is hypothetical and is not a report of any client site. Suppose /pricing/ shows full plans after clicking from the homepage, but View Source shows only a header, a loading message, and script tags. That pattern points to content fetched after mount. The fix to investigate is server delivery of that pricing text, not rewriting the pricing copy.

Prerendering, SSR, or CSR: which fits your release rhythm and hosting?

Pick by content type, update frequency, and what your hosting can operate, not by framework slogan.

StrategyWhat the crawler first receivesWhen it fits
Client-side rendering (CSR)A near-empty shell; content appears only after JS runsLogged-in app screens that truly do not need ranking
Prerender / static generation (SSG)Static HTML built at build time for known routesFixed marketing or content pages with infrequent changes
Server-side rendering (SSR)HTML generated per request with real content and linksPages that must be indexed reliably or vary per request
HydrationSSR or prerendered HTML, then JS attaches interactivityAny SSR or prerendered page that also needs to be interactive

Apply this rule before choosing: if a page’s ranking matters, its core text and internal links should exist in the HTML the server returns, before any client script runs. Hydration is a useful addition to SSR or prerendering because it makes a page interactive, but the content must already be in the first response.

For many SaaS marketing sites, the workable split is: prerender the stable marketing and documentation pages so each has a build-time HTML file, use SSR for pages that change often or depend on request conditions, and keep CSR for private product screens where search discovery is not the goal. Server-side or prerendered delivery is still recommended because it makes the site faster for users and crawlers, and not all bots can run JavaScript. Google can render JavaScript, so a CSR page is not automatically invisible, but rendering is a second step with its own timing, so SSR or prerendering removes that uncertainty.

Keep titles, canonical, and structured data in the same decision. Do not set canonical to a different address in client code than the address given in the original HTML, and do not start an indexable page with noindex and rely on JavaScript to remove it later. If you use hash routing for historical reasons, migrate indexable content to path-based URLs and keep server support for direct visits and refresh.

What evidence to collect before asking hosting or SEO support

Support can act faster when each URL has its own row with the exact observation. Prepare one row per affected URL with these fields:

  • Final URL, response status after redirects, and date checked
  • robots.txt result for that path, meta robots or X-Robots-Tag value, and declared canonical
  • Whether the main text, H1, internal links, and expected title appear in the initial HTML
  • Whether the same items appear after rendering, plus any blocked scripts, failed API calls, console exceptions, or loading placeholders left at render time
  • Search Console URL Inspection result for the exact URL: indexed version, last crawl date, canonical selected by Google, and a live test showing the HTML Google fetched now

Run the live test from URL Inspection and compare it with your saved HTML. Check three things: is the main text there, are the internal links there, and is the title the one you expect. If the crawled HTML is an empty shell, further copy editing will not fix indexing; the rendering delivery will. Record both the indexed version and the live test with dates, because they answer different questions about what Google knows versus what it can fetch now.

Do not treat a single Search Console label as a proven cause. A page reported as discovered but not yet crawled points to discovery paths, server availability, and how many URLs are exposed. A page reported as crawled but not indexed points to the page itself, duplicates, canonical, or Google’s judgment about inclusion. The needed next step is different in each case, so keep the label wording, the URL, and the check date together.

Where Guangsuan fits

If the evidence above shows an empty first response, conflicting titles, or links missing from delivered HTML, the next task is front-end delivery plus technical SEO review rather than more content. Guangsuan's Vue SEO guide compares CSR, prerendering, SSR, and hydration and lists the Vue decisions that decide whether content and links stay visible. For wider repair, Guangsuan's Google SEO service describes technical checks, content work, and link resources as separate scopes. How Google processes script-dependent pages is also described in Google's JavaScript SEO basics, including crawling, rendering, and indexing stages.

FAQ

Will switching every page to SSR fix indexing? No. SSR helps when the missing part is server delivery of public text, links, titles, or structured data. Blocked resources, wrong canonical, noindex left from staging, login requirements, or duplicate pages still need their own fixes and a retest of the same URLs.

Should we check only the homepage? No. Test representative templates and states: one marketing page, one article or docs page, one product or pricing page, one filtered or paginated view if it should be indexable, and one missing-page address. A homepage pass does not establish that deeper routes work on direct visit.

How long should we wait after a fix? Allow time for recrawl and reprocessing after the repair is live, and compare the same URL, country, device, and date range before and after. Passing a render test means Google could fetch the content now; it is not a promise that the page is indexed or will rank.