WordPress Content Behind a Login: What Should You Make Public for Visibility in Google and AI Search?

TL;DR Content behind a login is invisible to AI crawlers, because they don't log in. For Google it's visible only when Googlebot can reach it. The key decision: keep blog posts, docs, and case studies public; you can partially gate pricing and product internals. The method: publish a public subset, add clear summaries and specs, and mark paid content correctly with isAccessibleForFree:false plus hasPart. Control it: check robots.txt, bot access, indexing, and citations. Track "mention vs citation" separately. Semly helps you monitor how your brand shows up in AI answers. It measures visibility, it doesn't guarantee results.

If you run a WordPress site with a membership plugin or a paywall, you've probably asked yourself a simple question: how much should I actually hide? The honest answer is that hiding everything is the fastest way to disappear from AI answers. Here's how to decide what stays public, what gets gated, and how to set it up without breaking your SEO.

WordPress content behind a login: the visibility problem in one minute

Gated content is any page a visitor can't read without logging in, registering, or paying. A classic example: a detailed buying guide that only opens after you create a free account.

Here's the catch. Googlebot can index that page only if it can reach the content. AI crawlers are a different story. Bots like OAI-SearchBot, GPTBot, ChatGPT-User, PerplexityBot, and ClaudeBot don't log in. They don't fill out forms. They don't have accounts. So if your best content sits behind a login wall, it's effectively invisible to AI answers.

One more thing worth knowing: Google-Extended is a token that controls whether your content is used for Gemini training and grounding. It does not affect your inclusion in Google Search. Blocking it won't remove you from Search, and allowing it won't boost your rankings.

Watch out for this: don't assume an AI bot will log in like a regular user. It won't. If the content needs credentials, the bot sees nothing.

Common questions at this stage:

• Will Google index my login page? Usually no. Login and registration pages are typically set to noindex by default.
• Do AI bots log in? No.
• If I block one bot, do I vanish from AI? Not necessarily, but blocking a search bot like OAI-SearchBot removes your site from ChatGPT search answers.

Step 1: Decide what stays public (content-type map)

Before you touch a single setting, map your content. The rule is simple: make public whatever has citation value. That's the stuff an AI model would want to quote or summarize.

Start with your blog and docs. Those are usually the easiest wins. Pricing is the trickiest call, because it's both sensitive and highly searched.

This is also where measurement helps. Semly analyzes prompts, sources, competitors, and generated responses to show which topics actually earn citations for your brand. That tells you where to open the doors first, instead of guessing.

Step 2: Prepare your inputs and access before you change anything

Don't start editing until you have these ready:

• Access to your WordPress admin panel
• Your SEO plugin (Yoast or Rank Math)
• Access to your robots.txt file
• A list of your content types (posts, pages, docs, custom post types)
• A full backup of your site
• A staging environment for testing
• Tools for checks: Google Search Console, a JSON-LD validator, and a bot tester

Warning: always take a backup and test on staging first. A wrong robots.txt line or a stray noindex tag can quietly remove pages from Google, and you won't notice for weeks.

To check whether Googlebot has access, use the robots.txt tester in Search Console and the URL Inspection tool on a few key pages.

Step 3: Open the right doors in WordPress (indexing settings)

Now configure WordPress so public content is indexable and logins are not.

1. In Yoast or Rank Math, check the default noindex settings for author, date, and tag archives. These are usually thin or duplicate pages, so leaving them noindexed is fine.
2. Confirm that your posts and documentation pages are set to index.
3. Make sure login and registration pages stay noindex.
4. Hunt down accidental noindex tags on important content. This is a common mistake after a site migration or a plugin update.

Checkpoint: pick three random public URLs and inspect them in Google Search Console. They should be indexable and not blocked.

Step 4: Configure robots.txt for Google and AI crawlers

This is where you decide, on purpose, who gets in.

The golden rule: don't block public content by accident. A single misplaced Disallow: / can take your whole site out of the picture.

Checkpoint: test your robots.txt after every change. Load it in a browser and run it through a tester to confirm the rules do what you think they do.

Step 5: Mark paywalled content correctly (so you don't look like cloaking)

If part of a page is gated, tell Google. Use isAccessibleForFree:false together with hasPart and a cssSelector pointing to the gated section. This is how you avoid looking like you're showing different content to crawlers than to users, which is cloaking.

Here's a short example:

{
  "@context": "https://schema.org",
  "@type": "Article",
  "isAccessibleForFree": false,
  "hasPart": {
    "@type": "WebPageElement",
    "isAccessibleForFree": false,
    "cssSelector": ".paywall"
  }
}

This tells search engines that the article is partly paid and which part is behind the wall.

Flexible Sampling is another option. Instead of a hard login, you let readers see a few free articles per month (Google's guidance points to roughly 6 to 10), or you show a lead-in before the paywall. Google notes that satisfaction drops when a paywall is shown more than about 10% of the time.

Watch out for this: don't mark fully public content as paid. That confuses crawlers and can hurt how your pages are treated.

Step 6: Publish a public subset that AI can actually cite

Now the practical part. What do you actually put out in the open?

• Public summaries of gated guides
• FAQ sections that answer real questions
• Specifications and technical details
• Comparison pages
• Definitions and glossary entries

Keep the content in HTML. If your key information lives only in a PDF or an image, AI systems have a much harder time reading and citing it. HTML is the safe bet.

Hybrid gating works well here: put the form or the signup prompt on the page, but keep the core content in HTML. That way you capture leads and stay crawlable at the same time. Add FAQ or Article schema to make the structure obvious.

Checkpoint: view the page source. If the text isn't there in HTML, it isn't reliably citable.

Step 7: Check your work (checkpoints and warning signs)

Run through this checklist after your changes:

• Can Googlebot and your chosen AI bots access your public pages?
• Are your key pages indexed in Search Console?
• Do your pages show up in AI answers for relevant prompts?
• Are you tracking "mention vs citation"? A mention is your brand being named. A citation is your content being linked as a source. They're not the same thing.

Warning signs that something is off:

• Public content isn't getting indexed
• AI bots aren't visiting at all
• Citations are dropping over time

Semly is built for exactly this kind of monitoring. It tracks how your brand appears in answers from ChatGPT, Gemini, Claude, Grok, and Google AI Mode, and reports visibility, citation rate, and share of voice. In one Semly case, a brand's visibility moved from 11.8% to 22.6% over 90 days, and its citation rate went from 2.7% to 6.8%. Those are measured outcomes, not promises.

Common mistakes and how to fix them

• Blocking all AI bots "just in case." Fix: decide bot by bot. Block training bots if you want, but keep search bots like OAI-SearchBot open if you want to appear in AI answers.
• No paywall markup. Fix: add isAccessibleForFree:false and hasPart to partly gated pages.
• PDF instead of HTML. Fix: publish the same content as an HTML page.
• Noindex on key content. Fix: audit your SEO plugin settings and remove noindex from pages you want found.
• No measurement. Fix: set up tracking for visibility and citations so you can see what's working.

What to do next

Pick your path based on how much you're willing to open up:

• Full public: best for blogs, docs, and case studies. Maximum crawlability and citation potential.
• Hybrid gating: best when you want leads and visibility. Form on the page, content in HTML.
• Metering: best for publishers with lots of content. A few free reads per month, then a wall.

Then monitor and iterate. Visibility in AI answers isn't a one-time setup. It's a loop: publish, measure, adjust. Semly supports that loop with AI visibility monitoring, AI-readable catalogs, and AI Agent Leon, so you can see what's changing and act on it. No guarantees, just clear data.

FAQ

Does Google index login pages? Usually not. Login and registration pages are typically set to noindex by default, and they rarely have content worth ranking.

Do AI crawlers log in? No. Bots like OAI-SearchBot, GPTBot, and ClaudeBot don't authenticate. Content behind a login is invisible to them.

Does blocking one AI bot remove me from all AI answers? No. Each bot feeds a different system. Blocking OAI-SearchBot affects ChatGPT search, while blocking GPTBot affects training, not search visibility.

Does Google-Extended affect Google Search? No. Google-Extended controls Gemini training and grounding only. It doesn't impact your inclusion in Google Search.

Is metering better than a hard login? It depends on your goal. Metering keeps some content crawlable and citable, while a hard login hides everything. For visibility, metering usually wins.