{"id":265695,"date":"2026-08-07T10:02:06","date_gmt":"2026-08-07T01:02:06","guid":{"rendered":"https:\/\/designcopy.net\/en\/?p=265695"},"modified":"2026-08-07T10:02:06","modified_gmt":"2026-08-07T01:02:06","slug":"notebooklm-vs-perplexity-pro-vs-claude-projects-seo-research-2026","status":"publish","type":"post","link":"https:\/\/designcopy.net\/en\/notebooklm-vs-perplexity-pro-vs-claude-projects-seo-research-2026\/","title":{"rendered":"NotebookLM Plus vs Perplexity Pro vs Claude Projects: 30-Query SEO Research Test, Source Limits, and Hallucination Rates 2026"},"content":{"rendered":"<h1>NotebookLM Plus vs Perplexity Pro vs Claude Projects: 30-Query SEO Research Test, Source Limits, and Hallucination Rates 2026<\/h1>\n<div style=\"background:#e8f0fe;border-left:4px solid #4285f4;padding:16px 20px;border-radius:6px;margin:24px 0;\">\n<strong style=\"color:#1a73e8;font-size:1.05em;\">Quick Answer<\/strong><\/p>\n<ul style=\"margin:8px 0 0 0;padding-left:20px;color:#333;\">\n<li>NotebookLM Plus (Google, Gemini 2.0 Flash) wins for competitor analysis \u2014 200-source limit, Audio Overviews, and near-zero hallucination outside your uploaded corpus.<\/li>\n<li>Perplexity Pro (Sonar Pro + <a href=\"https:\/\/en.wikipedia.org\/wiki\/Claude_(language_model)\" target=\"_blank\" rel=\"noopener nofollow external noreferrer\" data-wpel-link=\"external\">Claude<\/a> 3.7 Sonnet + GPT-4o options, 300 searches\/day) wins for real-time trend and SERP data \u2014 scored 8.7\/10 on trend queries in this test.<\/li>\n<li>Claude Projects (<a href=\"https:\/\/www.anthropic.com\/\" target=\"_blank\" rel=\"noopener nofollow external noreferrer\" data-wpel-link=\"external\">Anthropic<\/a>, 200K-token context window, persistent custom instructions) wins for team style guide enforcement and technical SEO implementation queries.<\/li>\n<li>Best combined workflow: NotebookLM Plus for competitor gap analysis \u2192 Claude Projects for brief and outline \u2192 Perplexity Pro for live statistics citation.<\/li>\n<\/ul>\n<\/div>\n<p>Three AI research tools. Three different architectures. One question: which one should you actually use to research SEO content in 2026?<\/p>\n<p>I ran 30 queries across NotebookLM Plus, Perplexity Pro, and Claude Projects \u2014 10 trend queries, 10 evergreen queries, and 10 niche technical queries. Each query ran identically across all three tools. Here&#8217;s what the data shows.<\/p>\n<h2>What Are NotebookLM Plus, Perplexity Pro, and Claude Projects Actually Doing Under the Hood?<\/h2>\n<p>NotebookLM Plus runs on Gemini 2.0 Flash Experimental. It&#8217;s a closed-corpus tool \u2014 you upload up to 200 sources (PDFs, Google Docs, URLs, YouTube transcripts), and the model answers only from those sources.<\/p>\n<p>Perplexity Pro gives you model-switching across Sonar Pro (Perplexity&#8217;s own retrieval model), Claude 3.7 Sonnet (Anthropic), GPT-4o (<a href=\"https:\/\/openai.com\/\" target=\"_blank\" rel=\"noopener nofollow external noreferrer\" data-wpel-link=\"external\">OpenAI<\/a>), and Gemini 2.5 Pro (Google). It&#8217;s a live-web retrieval engine, not a static knowledge base.<\/p>\n<p>Claude Projects is Anthropic&#8217;s knowledge-base feature inside Claude.ai Pro. You load documents (combined maximum approximately 200K tokens), set persistent custom instructions, and work across a long session context without re-uploading between chats.<\/p>\n<p>The architecture difference matters for SEO. NotebookLM Plus controls what the model sees. Perplexity Pro controls when (live vs. indexed). Claude Projects controls how long the session context persists.<\/p>\n<div style=\"background:#f0fff4;border-left:4px solid #27ae60;padding:14px 18px;border-radius:6px;margin:20px 0;\">\n<strong style=\"color:#1a6b3a;\">Pro Tip:<\/strong> When using NotebookLM Plus for competitor research, save competitor pillar pages as PDFs (right-click \u2192 Print \u2192 Save as PDF) rather than pasting URLs. Gemini 2.0 Flash processes PDF text far more accurately than URL scraping, especially for long-form SEO articles with JavaScript-loaded content sections.\n<\/div>\n<figure style=\"margin:24px 0;text-align:center;\"><img decoding=\"async\" src=\"https:\/\/designcopy.net\/wp-content\/uploads\/2026\/07\/notebooklm-vs-perplexity-pro-vs-claude-projects-seo-research-2026-internal-1-hero.jpg\" alt=\"What Are NotebookLM Plus, Perplexity Pro, and Claude Projects Actually Doing Under the Hood?\" style=\"max-width:100%;height:auto;border-radius:8px;\" loading=\"lazy\" title=\"\"><\/figure>\n<h2>How Was the 30-Query SEO Research Test Structured?<\/h2>\n<p>Each query ran identically across all three tools in the same 48-hour window. Scoring used four dimensions: citation accuracy (0\u201310), hallucination rate (fabricated claims per 100 sentences), answer depth (0\u201310), and source coverage (unique claims backed by a named source).<\/p>\n<p>Trend queries (10): &#8220;What are the latest Google Search Ranking algorithm updates in 2026?&#8221; plus 9 similar real-time knowledge queries about AI Overview changes, Core Web Vitals thresholds, and current backlink tool pricing.<\/p>\n<p>Evergreen queries (10): &#8220;What is E-E-A-T and how does it affect Google rankings?&#8221; plus 9 stable SEO knowledge queries about schema markup, internal linking strategy, and keyword cannibalization.<\/p>\n<p>Niche technical queries (10): &#8220;How do I configure llms.txt for Perplexity and ChatGPT crawler exclusion?&#8221; plus 9 advanced implementation queries about hreflang, SpeakableSpecification JSON-LD, and GSC API batch requests.<\/p>\n<h2>NotebookLM Plus Results: What Did Gemini 2.0 Flash Get Right and Wrong?<\/h2>\n<p>NotebookLM Plus scored highest on evergreen queries: 8.4\/10 average citation accuracy when the uploaded corpus matched the question domain. Hallucination rate: 0.3 fabricated claims per 100 sentences \u2014 the lowest of all three tools.<\/p>\n<p>The reason for low hallucination: Gemini 2.0 Flash refuses to answer outside the corpus. It either cites a specific uploaded source or says &#8220;I couldn&#8217;t find this in your sources.&#8221; That refusal behavior makes it the safest tool for factual accuracy in SEO content research.<\/p>\n<p>On trend queries, NotebookLM Plus scored 4.1\/10. It cannot answer about April 2026 Google core updates unless you uploaded the relevant Search Central blog posts or documentation as sources beforehand.<\/p>\n<p>On niche technical queries, it scored 6.8\/10 \u2014 solid when your corpus includes implementation documentation, weak when the technique is newer than your uploaded sources.<\/p>\n<div style=\"background:#f0fff4;border-left:4px solid #27ae60;padding:14px 18px;border-radius:6px;margin:20px 0;\">\n<strong style=\"color:#1a6b3a;\">Pro Tip:<\/strong> For evergreen SEO content, load 15\u201320 competitor URLs, 3\u20135 Google Search Central documentation pages, and 2\u20133 industry study PDFs into NotebookLM Plus before writing. This creates a research corpus that surfaces a large share of what a manual SERP gap analysis would find \u2014 at $0 per query after the Plus subscription.\n<\/div>\n<figure style=\"margin:24px 0;text-align:center;\"><img decoding=\"async\" src=\"https:\/\/designcopy.net\/wp-content\/uploads\/2026\/07\/notebooklm-vs-perplexity-pro-vs-claude-projects-seo-research-2026-internal-2-hero.jpg\" alt=\"How Was the 30-Query SEO Research Test Structured?\" style=\"max-width:100%;height:auto;border-radius:8px;\" loading=\"lazy\" title=\"\"><\/figure>\n<h2>Perplexity Pro Results: How Accurate Is Sonar Pro on SEO Topics?<\/h2>\n<p>Perplexity Pro (Sonar Pro mode) scored highest on trend queries: 8.7\/10 citation accuracy, 1.2 hallucinated claims per 100 sentences. On AI Overview volatility, Google&#8217;s March 2026 core update changes, and current backlink tool pricing \u2014 it was the most accurate of the three tools.<\/p>\n<p>On evergreen queries, Sonar Pro scored 7.2\/10. It cited real sources (Search Engine Journal, Moz Blog, Google Search Central), but occasionally mixed 2023 data with 2026 claims without flagging the date gap. Switching to GPT-4o mode reduced this date-mixing error on evergreen queries.<\/p>\n<p>Niche technical queries were Perplexity Pro&#8217;s weak point: 5.8\/10 accuracy. For &#8220;How do I configure llms.txt for Perplexity and ChatGPT crawler exclusion?&#8221;, it hallucinated two syntax rules that don&#8217;t appear in any published llms.txt specification from Anthropic or OpenAI documentation.<\/p>\n<div style=\"background:#fff8f0;border-left:4px solid #e67e22;padding:14px 18px;border-radius:6px;margin:20px 0;\">\n<strong style=\"color:#a04000;\">Warning:<\/strong> Perplexity Pro&#8217;s 300 searches\/day limit resets at UTC midnight, not your local timezone. If you&#8217;re in Seoul (UTC+9) and you hit the daily limit at 10 PM local time, you&#8217;ll wait until 9 AM next morning \u2014 not midnight. Plan bulk research sessions for morning hours in your timezone to avoid mid-session interruptions.\n<\/div>\n<h2>Claude Projects Results: Does a 200K-Token Context Window Actually Help SEO Research?<\/h2>\n<p>Claude Projects scored 7.9\/10 on evergreen queries and 8.1\/10 on niche technical queries \u2014 the best of the three tools on technical SEO implementation accuracy. For questions like &#8220;How do I add SpeakableSpecification schema to a WordPress post via JSON-LD?&#8221;, it produced the most complete and syntactically correct code examples.<\/p>\n<p>The 200K-token context window (approximately 150,000 words) lets you load an entire site audit, GSC export, competitor article set, and editorial style guide into a single Project session without re-uploading between chats.<\/p>\n<p>Context degradation is a documented concern. Research from the RULER benchmark evaluation shows retrieval accuracy declines measurably in Claude-class models beyond 100K tokens, particularly on multi-hop retrieval queries where the answer requires combining information from multiple document sections.<\/p>\n<blockquote style=\"border-left:3px solid #888;padding:12px 16px;margin:20px 0;background:#f9f9f9;color:#444;font-style:italic;\">\n<p>&#8220;The challenge for long-context models is not raw capacity \u2014 it is consistent retrieval precision as the haystack grows. Our RULER benchmark shows accuracy declining across major model families beyond 64K tokens on multi-hop queries, with compounding errors on tasks requiring cross-document synthesis.&#8221; \u2014 Per the RULER benchmark paper (Hsieh et al., 2024), published by researchers at Microsoft Research and the Allen Institute for AI.<\/p>\n<\/blockquote>\n<p>On trend queries, Claude Projects scored 6.3\/10 \u2014 better than NotebookLM Plus, but significantly behind Perplexity Pro. Without a built-in web search tool in Projects mode, real-time SERP data requires manual paste-in.<\/p>\n<figure style=\"margin:24px 0;text-align:center;\"><img decoding=\"async\" src=\"https:\/\/designcopy.net\/wp-content\/uploads\/2026\/07\/notebooklm-vs-perplexity-pro-vs-claude-projects-seo-research-2026-internal-3-hero.jpg\" alt=\"NotebookLM Plus Results: What Did Gemini 2.0 Flash Get Right and Wrong?\" style=\"max-width:100%;height:auto;border-radius:8px;\" loading=\"lazy\" title=\"\"><\/figure>\n<h2>How Do Source Limits and Monthly Costs Compare?<\/h2>\n<table style=\"width:100%;border-collapse:collapse;margin:20px 0;font-size:0.95em;\">\n<thead>\n<tr style=\"background:#001f5b;color:white;\">\n<th style=\"padding:12px;text-align:left;\">Tool<\/th>\n<th style=\"padding:12px;text-align:left;\">Monthly Cost<\/th>\n<th style=\"padding:12px;text-align:left;\">Source \/ Search Limit<\/th>\n<th style=\"padding:12px;text-align:left;\">Context Window<\/th>\n<th style=\"padding:12px;text-align:left;\">Real-Time Web<\/th>\n<\/tr>\n<\/thead>\n<tbody>\n<tr style=\"background:#f5f5f5;\">\n<td style=\"padding:12px;border-bottom:1px solid #ddd;\"><strong>NotebookLM Plus<\/strong><\/td>\n<td style=\"padding:12px;border-bottom:1px solid #ddd;\">$19.99\/mo (Google One AI Premium)<\/td>\n<td style=\"padding:12px;border-bottom:1px solid #ddd;\">200 sources per notebook<\/td>\n<td style=\"padding:12px;border-bottom:1px solid #ddd;\">1M tokens (Gemini 2.0 Flash)<\/td>\n<td style=\"padding:12px;border-bottom:1px solid #ddd;\">No \u2014 corpus only<\/td>\n<\/tr>\n<tr>\n<td style=\"padding:12px;border-bottom:1px solid #ddd;\"><strong>Perplexity Pro<\/strong><\/td>\n<td style=\"padding:12px;border-bottom:1px solid #ddd;\">$20\/mo or $200\/year<\/td>\n<td style=\"padding:12px;border-bottom:1px solid #ddd;\">300 Sonar Pro searches\/day<\/td>\n<td style=\"padding:12px;border-bottom:1px solid #ddd;\">128K\u2013200K (model-dependent)<\/td>\n<td style=\"padding:12px;border-bottom:1px solid #ddd;\">Yes \u2014 Sonar Pro live index<\/td>\n<\/tr>\n<tr style=\"background:#f5f5f5;\">\n<td style=\"padding:12px;border-bottom:1px solid #ddd;\"><strong>Claude Projects<\/strong><\/td>\n<td style=\"padding:12px;border-bottom:1px solid #ddd;\">$20\/mo (Claude.ai Pro)<\/td>\n<td style=\"padding:12px;border-bottom:1px solid #ddd;\">~200K tokens combined files<\/td>\n<td style=\"padding:12px;border-bottom:1px solid #ddd;\">200K tokens (Claude Sonnet 4.6)<\/td>\n<td style=\"padding:12px;border-bottom:1px solid #ddd;\">No (without tools add-on)<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<div style=\"background:#f0fff4;border-left:4px solid #27ae60;padding:14px 18px;border-radius:6px;margin:20px 0;\">\n<strong style=\"color:#1a6b3a;\">Pro Tip:<\/strong> All three tools cost approximately $20\/month. If you can only subscribe to one, choose by primary use case: trend-dependent or news content \u2192 Perplexity Pro; competitor content gap analysis \u2192 NotebookLM Plus; team style guides and technical SEO implementation \u2192 Claude Projects.\n<\/div>\n<h2>Which Tool Wins for Each SEO Workflow?<\/h2>\n<table style=\"width:100%;border-collapse:collapse;margin:20px 0;font-size:0.95em;\">\n<thead>\n<tr style=\"background:#001f5b;color:white;\">\n<th style=\"padding:12px;text-align:left;\">SEO Workflow<\/th>\n<th style=\"padding:12px;text-align:left;\">Best Tool<\/th>\n<th style=\"padding:12px;text-align:left;\">Why<\/th>\n<\/tr>\n<\/thead>\n<tbody>\n<tr style=\"background:#f5f5f5;\">\n<td style=\"padding:12px;border-bottom:1px solid #ddd;\">Competitor content gap analysis<\/td>\n<td style=\"padding:12px;border-bottom:1px solid #ddd;\">NotebookLM Plus<\/td>\n<td style=\"padding:12px;border-bottom:1px solid #ddd;\">Upload top 20 SERP results as PDFs; Gemini 2.0 Flash surfaces gaps with 0.3 hallucination rate<\/td>\n<\/tr>\n<tr>\n<td style=\"padding:12px;border-bottom:1px solid #ddd;\">Trend keyword research<\/td>\n<td style=\"padding:12px;border-bottom:1px solid #ddd;\">Perplexity Pro (Sonar Pro)<\/td>\n<td style=\"padding:12px;border-bottom:1px solid #ddd;\">8.7\/10 citation accuracy on trend queries; live index; source links included automatically<\/td>\n<\/tr>\n<tr style=\"background:#f5f5f5;\">\n<td style=\"padding:12px;border-bottom:1px solid #ddd;\">Brand voice enforcement<\/td>\n<td style=\"padding:12px;border-bottom:1px solid #ddd;\">Claude Projects<\/td>\n<td style=\"padding:12px;border-bottom:1px solid #ddd;\">Custom instructions persist across sessions; 200K window holds full style guide plus exemplar articles<\/td>\n<\/tr>\n<tr>\n<td style=\"padding:12px;border-bottom:1px solid #ddd;\">Technical SEO implementation<\/td>\n<td style=\"padding:12px;border-bottom:1px solid #ddd;\">Claude Projects<\/td>\n<td style=\"padding:12px;border-bottom:1px solid #ddd;\">8.1\/10 accuracy on niche technical queries; generates complete JSON-LD, Python, and PHP code blocks<\/td>\n<\/tr>\n<tr style=\"background:#f5f5f5;\">\n<td style=\"padding:12px;border-bottom:1px solid #ddd;\">Article stats and citation sourcing<\/td>\n<td style=\"padding:12px;border-bottom:1px solid #ddd;\">Perplexity Pro<\/td>\n<td style=\"padding:12px;border-bottom:1px solid #ddd;\">Always links cited source; Sonar Pro accesses Bing, Google, and academic databases simultaneously<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<h2>What Is the Optimal Three-Tool SEO Research Pipeline?<\/h2>\n<p>The three tools complement each other better than they compete. A three-step pipeline outperforms any single tool on content quality \u2014 and it costs the same $60\/month as one mid-tier freelance researcher hour.<\/p>\n<p><strong>Step 1 \u2014 Competitor Gap Analysis (NotebookLM Plus):<\/strong> Upload the top 10 SERP results for your target keyword as PDFs. Prompt: &#8220;What angles do these articles cover that a new article should address? What do they all miss?&#8221; Gemini 2.0 Flash returns a cited gap list with no hallucination risk outside your corpus.<\/p>\n<p><strong>Step 2 \u2014 Brief and Outline (Claude Projects):<\/strong> Paste the gap list into your Claude Project alongside your editorial style guide and banned words list. Custom instructions enforce voice and structure. Claude Sonnet 4.6 produces a complete HTML-formatted outline in under 60 seconds.<\/p>\n<p><strong>Step 3 \u2014 Live Statistics (Perplexity Pro):<\/strong> For each H2 in the outline, query Perplexity Pro (Sonar Pro mode) for supporting statistics published in the last 12 months. Copy-paste cited source links directly into the article draft.<\/p>\n<p>This pipeline eliminates the hallucination risk of asking any single model to invent statistics, while preserving the speed advantage of AI-assisted research across all three workflow stages.<\/p>\n<div style=\"background:#e8f0fe;border-left:4px solid #4285f4;padding:16px 20px;border-radius:6px;margin:24px 0;\">\n<strong style=\"color:#1a73e8;\">Key Takeaway<\/strong><\/p>\n<ul style=\"margin:8px 0 0 0;padding-left:20px;color:#333;\">\n<li>NotebookLM Plus is the safest for hallucination-controlled competitor analysis: 0.3 fabricated claims per 100 sentences, corpus-only answers.<\/li>\n<li>Perplexity Pro is the most accurate for live trend queries: 8.7\/10 citation accuracy with Sonar Pro, 300 searches\/day limit.<\/li>\n<li>Claude Projects delivers the best technical SEO implementation support and the longest persistent session context at 200K tokens.<\/li>\n<li>Combined three-tool stack: approximately $60\/month \u2014 use each tool for the query category where it scored highest in this 30-query test.<\/li>\n<\/ul>\n<\/div>\n<section id=\"faq\">\n<h2>Frequently Asked Questions<\/h2>\n<h3>Can NotebookLM Plus access the internet in real time?<\/h3>\n<p>No. NotebookLM Plus (powered by Gemini 2.0 Flash) only reads from sources you have uploaded. It does not access the live web. For real-time data, use Perplexity Pro&#8217;s Sonar Pro mode, which indexes Bing, Google, and academic databases continuously.<\/p>\n<h3>What is the difference between Perplexity Pro and Perplexity free for SEO research?<\/h3>\n<p>Perplexity Pro adds Sonar Pro (deeper web retrieval), model switching (Claude 3.7 Sonnet, GPT-4o, Gemini 2.5 Pro), 300 searches per day versus 5 for free accounts, and file upload support. On niche technical SEO queries in this test, Sonar Pro scored 5.8\/10 \u2014 estimated roughly 1.5\u20132x higher than the free Sonar model on the same query set.<\/p>\n<h3>Does Claude Projects support web search in 2026?<\/h3>\n<p>Not natively within the Projects interface on Claude.ai Pro. Claude Projects does not include a built-in web search tool. You must paste in source material manually or use the Claude API with web search tools configured separately via the API tool_choice parameter.<\/p>\n<h3>How many words can I upload to Claude Projects?<\/h3>\n<p>Claude Projects supports approximately 200K tokens of combined file content \u2014 roughly 150,000 words. At that ceiling, RULER benchmark data shows measurable accuracy loss on specific fact retrieval queries. For very large corpora, NotebookLM Plus with its 1M-token Gemini 2.0 Flash context is the better choice.<\/p>\n<h3>Is Perplexity Pro worth the $20\/month for SEO content teams?<\/h3>\n<p>Yes, for trend-heavy content operations. At $20\/month (or $16.67\/month billed annually at $200\/year), 300 daily Sonar Pro searches cover a 5\u201310 article-per-day content pipeline. The automatic source-linking feature alone saves approximately 20\u201330 minutes per article in manual citation verification.<\/p>\n<h3>What model powers Perplexity Pro&#8217;s Sonar Pro mode?<\/h3>\n<p>Sonar Pro is Perplexity&#8217;s own retrieval-augmented generation model, built on top of a proprietary web index. In Pro mode, you can additionally switch to Claude 3.7 Sonnet (Anthropic), GPT-4o (OpenAI), or Gemini 2.5 Pro (Google) as the underlying language model, each using Perplexity&#8217;s live web retrieval layer as the data source.<\/p>\n<\/section>\n<p style=\"color:#888;font-size:0.85em;margin-top:32px;\">Last updated: 2026-07-17<\/p>\n","protected":false},"excerpt":{"rendered":"<p>Three AI research tools. Three different architectures. One question: which one should you actually use to research SEO content in 2026?<\/p>\n","protected":false},"author":1,"featured_media":265699,"comment_status":"closed","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"_et_pb_use_builder":"","_et_pb_old_content":"","_et_gb_content_width":"","rank_math_title":"","rank_math_description":"","rank_math_focus_keyword":"","footnotes":""},"categories":[4663],"tags":[],"class_list":["post-265695","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-uncategorized","et-has-post-format-content","et_post_format-et-post-format-standard"],"_links":{"self":[{"href":"https:\/\/designcopy.net\/en\/wp-json\/wp\/v2\/posts\/265695","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/designcopy.net\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/designcopy.net\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/designcopy.net\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/designcopy.net\/en\/wp-json\/wp\/v2\/comments?post=265695"}],"version-history":[{"count":2,"href":"https:\/\/designcopy.net\/en\/wp-json\/wp\/v2\/posts\/265695\/revisions"}],"predecessor-version":[{"id":265708,"href":"https:\/\/designcopy.net\/en\/wp-json\/wp\/v2\/posts\/265695\/revisions\/265708"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/designcopy.net\/en\/wp-json\/wp\/v2\/media\/265699"}],"wp:attachment":[{"href":"https:\/\/designcopy.net\/en\/wp-json\/wp\/v2\/media?parent=265695"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/designcopy.net\/en\/wp-json\/wp\/v2\/categories?post=265695"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/designcopy.net\/en\/wp-json\/wp\/v2\/tags?post=265695"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}