AIFree

The AI Citation Outreach Kit: Get Recommended by ChatGPT & Perplexity

How to get your app, website, or agency cited and recommended by ChatGPT and Perplexity. Complete 5-step automation workflow, copy-paste prompts, and outreach templates.

By Vikas YadavPublished 20 Nov 2026Updated 20 Nov 20268 min read

Get the PDF

Free. No email or account needed. PDF · 511 KB

Download PDF

The preview only loads when you tap "Read PDF here", which saves mobile data. Download gives you the original PDF file.

Key takeaways

  • AI search engines rarely cite your homepage directly; they repeat brands listed on third-party directories, comparison lists, and community forums.
  • Narrow queries combining use case + location + year trigger recent lists that are significantly easier to get placed on.
  • Always offer legitimate value (affiliate rev-share, product access, benchmark data) instead of mass automated pitch emails.

The AI Citation Outreach Kit: Complete Playbook

How to Get Your App, Website, or Agency Recommended by ChatGPT & Perplexity

Curated by VikasBuilds (vikasbuilds.com) · Adapted from the Otterly.AI "Agency X" Citation Study

When a buyer asks ChatGPT or Perplexity for a recommendation, the AI rarely relies on a company's own homepage. Instead, it searches the web, reads a short list of third-party "Best of" articles, directories, and community threads, and repeats the brands named on those pages. Get your product or service added to those exact source pages, and AI search engines begin recommending you.


01. The Real Experiment Behind This Playbook

AI search tracker Otterly.AI tested this mechanism by launching a brand-new agency website that had never existed before—just seven basic pages, zero blog posts, and zero SEO backlinks. Instead of building links to the website, they focused exclusively on getting the brand listed on third-party pages already cited for narrow local queries.

Documented Results (After 14 Days)

  • Third-Party Mentions Earned: 16 placements across directories, listicles, and community posts
  • Total AI Answers Naming the Brand: 90 AI responses across 9 tracked high-intent buyer questions
  • ChatGPT Ranking: Top 3 among all tracked brands in the target category
  • Google AI Overviews Ranking: #2 among all tracked brands
  • Average List Position: 1.71 (consistently placed near the top of AI shortlists)

Practical Reality Check: Narrow queries combining a use case + location + year (such as "Top app developers in California 2026") trigger recently updated lists that are much easier to join. Start with free industry directories before emailing blog editors, and offer real value—such as free product access, benchmark data, or an affiliate partnership—rather than asking for undisclosed paid rankings.


02. The 5-Step Automation Workflow

Step Action Tools Used Time Needed
01 Generate high-intent buyer questions Claude (with web search) + Prompt 1 10 minutes
02 Extract repeating cited URLs ChatGPT / Perplexity + Prompt 2 25 minutes
03 Find page owners & listing paths Claude (with web search) + Prompt 3 20 minutes
04 Send personalized outreach Prompt 4 + your email inbox 30 minutes
05 Track weekly AI visibility Fresh AI chats + Prompt 5 15 mins / week

03. Step 01: Find Your Buyers' High-Intent Questions

Open a new Claude chat with web search enabled, fill in the brackets inside Prompt 1, and run it. Specificity is the key: prompts that include a buyer type, city/country, and current year generate short brand lists that are far easier to enter than broad global keywords.

Prompt 1 / The Buyer Question Generator

You are my buyer research analyst. Find the exact questions my buyers ask ChatGPT before choosing a product or service like mine. About my business: - What I sell, in one sentence: [describe your app, website, or service] - Who buys it: [target customer or job role] - Where I sell: [city, country, or global/online] - Price range: [starting price or pricing model] - Top competitors: [3 to 5 competitor names] Search the web first. Check competitor pages, review sites, and Reddit threads in my niche to capture the exact language buyers use. Then write 100 realistic questions a buyer would type into ChatGPT across every stage: - Problem: "Why does my...", "What causes..." - Solution: "How do I...", "What is the best way to..." - Shortlist: "Best [category] for [use case] in [location] 2026" - Compare: "[Competitor] vs [competitor]", "Alternatives to [competitor]" - Decide: "Is [competitor] worth it?", "[Category] pricing 2026" Rules: - Include a location, current year, or specific buyer persona in at least half of the questions. - Do not mention my brand name and do not invent search volume numbers. - Rate each question's likelihood of triggering specific brand names: High, Medium, or Low. Return one table sorted with High ratings first: # | Question | Stage | Names Brands? Then list the top 25 High-intent questions (including at least 5 Compare questions) as plain lines I can copy.

04. Step 02: List the Pages ChatGPT & Perplexity Cite

Run your 25 priority questions in fresh, logged-out or temporary ChatGPT and Perplexity sessions so your personal history does not skew the answers. Open the cited sources under each response and log every link into a spreadsheet. Instead of waiting an entire month before acting, prioritize any URL that gets cited across 3 or more prompts right away, and continue weekly logging to catch rotating sources.

Tracking Template Structure

  • Date: YYYY-MM-DD
  • Tracked Question: Best [category] in [location] 2026?
  • Named? (Y/N): Y / N
  • Rank: #
  • Cited URLs: Paste all source URLs

Prompt 2 / The AI Citation Analyst

You are my AI citation analyst. Below is my log of questions asked to AI search engines and the source URLs cited in each answer: [paste your log rows here] Build my outreach target list: 1. Strip tracking parameters (such as ?utm_source=chatgpt.com) and merge duplicate URLs pointing to the same page. 2. Count how many total times each page was cited and across how many unique questions. 3. Categorize each page: Listicle / Ranking Article, Review Platform, Directory, Forum Thread, News Article, Affiliate Roundup, Competitor Site, or Other. 4. Mark whether my brand can realistically be added: Yes (lists, directories, roundups), Maybe, or No (competitor homepages or pages naming no brands). 5. Move competitor homepages and "No" pages to a separate list at the bottom. Return a sorted table (most-cited pages first): # | Page Title | Clean URL | Domain | Page Type | Times Cited | Unique Questions | Can Be Added? Summarize in three lines which domains and page formats dominate my category.

05. Step 03: Find the Owner or Listing Path for Each Page

Copy the rows marked Yes and Maybe from Step 2 and run them through Claude in batches of 20. Always claim free directories and self-serve listings first before pitching editorial writers.

Prompt 3 / The Page Owner & Listing Finder

You are my outreach research agent. For each URL below, find who controls the content and the legitimate way to get listed: [paste your "Yes" and "Maybe" rows: Page Title and URL] For each page: 1. Open the URL and record the author's name, role, and the article's last updated date. 2. Identify the best contact in order: byline author, section editor, or site publisher. 3. Extract only publicly listed contact channels: a published email address, an official contact/editorial submission form URL, or a public LinkedIn/X profile. Never guess email addresses. 4. Identify the listing mechanism: Editorial Pitch, Self-Serve Directory Submission (include link), Affiliate Program, or Official Advertising/Sponsorship Page. 5. Note helpful context (e.g., outdated year in title, open submission guidelines, or missing category options). Return a table: # | Page Title | Contact Name | Role | Email or Form URL | Social Profile | Listing Path | Source URL | Notes

06. Step 04: Send Personalized, Non-Spammy Outreach

Editors ignore generic mass templates. Keep every message under 110 words, prove you actually read their article with one specific detail, and offer genuine value—such as free lifetime access, a revenue-share affiliate link, verified user metrics, or a clearly labeled sponsored placement.

Prompt 4 / The Personalized Outreach Writer

You are my outreach writer. Write one concise, personalized email for each contact row below: [paste the contact table from Prompt 3] My details: - Brand & Website: [Name and URL] - One-sentence description: [What your app, website, or service does] - Concrete proof: [Real user count, delivery speed, pricing edge, or case result] - My offer: [e.g., "Free pro access + 30% affiliate commission" or "Interested in your sponsored listing rates"] - Sender name & role: [Your Name, Role] Email structure: Subject: Quick addition for your "[Page Title]" guide Hi [First Name or Site Team], I was reading your "[Page Title]" roundup and noticed [insert one true, specific detail from their article]. I build [Brand]—[one-sentence description]. [One line of concrete proof.] Given your focus on [topic], we would be a strong fit for your readers. [State your offer clearly, or ask for their editorial/sponsored inclusion process.] Best, [Your Name] [Role], [Brand] Rules: - Keep every email under 110 words. No fluff or fake compliments. - If you cannot verify a detail on the live page, insert [VERIFY DETAIL] so I can check it manually. - Separate self-serve directory links into a checklist at the end instead of writing emails for them. - Add a two-sentence follow-up message for each contact to send 5 days later. Output each email and follow-up clearly, followed by a CSV block with columns: Contact, First Name, Subject, Body, Follow-Up.

07. Step 05: Track Weekly Mentions & Community Threads

AI models also pull heavily from Reddit discussions and YouTube reviews. Alongside your weekly prompt check, run targeted Google and YouTube searches for your top 25 buyer questions to find active threads where you can leave transparent, helpful replies disclosing your brand.

Prompt 5 / The Weekly AI Visibility Tracker

You are my AI visibility analyst. Below is my weekly tracking log across my 25 buyer questions, plus the dates my new placements went live: [paste weekly log: Date | Question | Named? (Y/N) | Position | Cited URLs] [paste live placements: Page URL | Date Published] Analyze my progress: 1. Calculate my weekly mention rate (% of questions naming my brand) and raw counts. 2. Calculate my average list position whenever my brand is named. 3. Identify which newly added pages are now being cited by AI and how many days after going live they appeared. 4. List the top questions where competitors appear without me, along with the exact source URLs driving those competitor mentions. 5. Give a one-sentence summary of the overall trend.
All resources
Security & PrivacyFeatured

Application Security & Anti-Reverse Engineering Playbook

Complete roadmap to protect your mobile app and web application's source code, core logic, proprietary assets, and brand name from reverse engineering, APK repackaging, and copycats.

Updated 20 Nov 2026
App Development

35 UI & UX States Visual Guide (Checklist & Mockups)

A complete visual guide and practical checklist of 35 essential UI and UX states every website and app must design before shipping. Includes mobile mockups, rules, and quick tests.

Updated 19 Nov 2026