Pre-rendering paywalled content — Flexible Sampling without cloaking

Introduction

1. 1

   Pick a Flexible Sampling model: metering or lead-in

   Metering gives users a fixed number of free articles per month. Lead-in gives all users the first paragraphs free. Either is acceptable to Google; choose based on user experience goals, not SEO.

2. 2

   Serve full text to Googlebot and metered / lead-in to users

   Pre-rendered snapshots contain the full article. Human responses implement metering or lead-in via the same middleware layer that does bot detection.

3. 3

   Mark CreativeWork.isAccessibleForFree = false

   Add isAccessibleForFree: false to the Article JSON-LD. Include hasPart with cssSelector pointing to the paywalled section. This is the "no cloaking" insurance policy.

   jsonCopy

   1{
   2  "@context": "https://schema.org",
   3  "@type": "NewsArticle",
   4  "isAccessibleForFree": false,
   5  "hasPart": {
   6    "@type": "WebPageElement",
   7    "isAccessibleForFree": false,
   8    "cssSelector": ".paywalled"
   9  }
   10}

4. 4

   Apply the .paywalled class to locked content

   The cssSelector referenced in hasPart must match real DOM in the pre-rendered snapshot. If you omit the class or strip it from the snapshot, Google treats the article as fully free and you drift into cloaking territory.

5. 5

   Monitor Search Console for cloaking warnings

   If Google suspects cloaking, a manual action appears in Search Console. Check monthly. If a warning appears, audit the diff between the snapshot and the human response on 5 sample URLs.

• Guide — hydration failures

  Paywall state often lives in hydration — mismatch risk.

• Guide — cache headers

  Vary on User-Agent for bot vs user responses.

• Guide — crawl frequency signals

  Freshness on subscription content.

• Use case — SaaS

  Public surface vs gated content boundary.

• Compare — vs Prerender.io

  Vendor comparison.

• Technology — how pre-rendering works

  Pipeline primer.