Why does Google still insist on anchor tags instead of JavaScript for your links?

In practice? A link like <div onclick="location.href='/page'"> has no guarantee of being crawled. Google must execute the JavaScript, interpret the event handler, and guess the navigation intent. Nothing is explicit.

The <a href="/page"> tag exposes the URL directly in the initial HTML DOM. Google sees it immediately, without code execution. The href attribute is a clear and universal signal: "this is a link to this destination."

An onclick requires full JavaScript execution. Google must load scripts, execute them in its rendering engine, then analyze triggered events. This is resource-intensive and subject to silent failures: timeout, blocked script, execution error.

Does this directive impact all types of sites?

Modern frameworks generally generate correct <a href> tags (Next.js Link, Nuxt NuxtLink, Gatsby Link). The problem arises with custom implementations or UI components that simulate links without actually being ones.

Is this recommendation consistent with what Google claims elsewhere?

Because real-world reality contradicts marketing claims. Tests regularly show JavaScript links not discovered, content not indexed, rendering timeouts. Rendering is expensive, slow, and Google prioritizes it. Not all JavaScript links pass the threshold. [To verify] on sites with large page volumes.

In what specific cases does a JavaScript link pose a problem?

First case: links loaded after interaction. A dropdown menu that generates <div onclick> on hover will never be crawled — Googlebot doesn't simulate complex user interactions. Links must be in the initial DOM or loaded automatically.

Second case: complex event handlers. An onclick that triggers a function that calls another function that modifies window.location... Google may abandon mid-execution. The longer the execution chain, the higher the failure risk.

Third case: client-side conditional links. Code that displays a link only if a localStorage variable exists, or if an API responds — Google doesn't have that context. It sees the "disconnected" or "default" version.

Are there situations where onclick is acceptable?

Edge case: a <a href="/page" onclick="track()"> where onclick does tracking then lets the browser follow the href. This is acceptable — the href remains present and functional. The JavaScript is an additional layer, not the primary mechanism.

How do I audit my site's links to detect problems?

Use Screaming Frog in JavaScript mode: compare links discovered in pure HTML crawl vs JavaScript rendering. If URLs appear only in JS mode, they depend on rendering — risk identified.

Test with Search Console: take 10 key URLs with JavaScript navigation, pass them through the URL inspection tool, check the "Additional information" tab > "Detected outbound links." If your menus or pagination don't appear, Google doesn't see them.

What concrete code changes should I make?

For SPAs: verify that your navigation components (Link, RouterLink, etc.) properly generate <a href> in the final HTML. Inspect the rendered DOM, not just source code — some frameworks conditionally render the href.

If your navigation relies on history.pushState() without a corresponding href, add fallback links: a visible <a href="/page"> that triggers modern JavaScript, but remains functional if JS fails.

How do I validate that fixes work?

• Crawl the site with Screaming Frog in HTML-only mode: all critical links must be detected
• Verify that each navigation link, menu, pagination uses <a href="URL">
• Test 5-10 URLs in Search Console URL inspection tool, outbound links tab: confirm discovery
• Disable JavaScript in Chrome DevTools, navigate the site: links must remain clickable
• Audit onclick/addEventListener events: none should be the only mechanism for navigating to an indexable page
• Validate with Lighthouse or PageSpeed Insights: no warnings about non-crawlable links