How New Brands Get Recommended by ChatGPT and Perplexity: 5 Practical Steps

Hi everyone, this is Neo.

A lot of independent site owners share the same misconception: they assume a brand that just launched — or a site that just went live — can’t possibly get recommended by AI search engines like ChatGPT, Gemini, or Perplexity. The common refrain: “We’re too new. AI has no idea we exist.”

But the latest experiment data says: you’re wrong.

When AI “ignores” you, it’s usually not because you’re “too new” — it’s because your structure, repetition, and trust signals aren’t clear enough. In other words, AI doesn’t think you’re “bad” — it just thinks citing you is “risky.”

The deep-dive report we’re talking about today followed a brand-new B2B company that, with zero history, reached a 16.5% visibility rate in AI search results after just 6 weeks of work.

If you’re a new brand — or frustrated that AI search can’t see you — you need to read this one to the end.

Why Does AI “Ignore” New Brands?

When LLMs answer user questions, their top priority is avoiding risk. They gravitate toward brands that have existed for years with a big digital footprint.

New brands usually face these 4 problems, which make AI hesitate to cite them:

  1. Missing historical signals: there’s almost nothing about you on the web.
  2. No trust endorsements: no backlinks, no reviews, no press coverage.
  3. Brand name confusion: the name is too generic and gets mixed up with other terms.
  4. Vague positioning: your site’s descriptions are cryptic, or appear only once — AI can’t remember you.

Neo’s take:

In the generative search era, visibility is no longer about ranking — it’s about reasoning. AI has to reason its way to confirming you’re a trustworthy entity before it will recommend you to users.

The Experiment: From 0 to 16.5% Visibility in 6 Weeks

To test whether a new brand can break into AI search, a team overseas ran a 12-week experiment on a brand-new B2B company: 0 history, 0 backlinks, 0 press coverage.

Results (first 6 weeks):

  • The brand appeared in 39 of 150 buyer-related questions.
  • It racked up 74 mentions, 42 of which were used as cited sources.
  • Citation accuracy hit an impressive 61.6%.

How did they pull it off? The experiment surfaced 6 key patterns.

Pattern 1: Structure Beats Topic

AI loves structured content. Rambling, endless articles are hard for it to extract key points from. But if your article answers the question directly, has clear steps, and makes crisp points, AI can grab it far more easily.

Pattern 2: Social Media Is an “Amplifier”

In the first two weeks, most AI citations didn’t come from the brand’s own site — they came from LinkedIn and Medium articles. Strategy: publish your core ideas on high-authority platforms like LinkedIn first. AI discovers them there faster than it indexes your own site.

Pattern 3: AI “Hallucinates” When Your Signals Are Too Weak

When your pages load too slowly (5-15+ seconds) or crawlers get blocked, AI can’t retrieve solid information — so it starts “making things up” or citing your competitors instead. Neo’s take: technical SEO is still critical in the AI era — arguably more so.

Pattern 4: The 3-Week Indexing Window

New domains typically need 3-4 weeks before AI crawls their content and uses it in answers. In the experiment, the first citation from the brand’s own site didn’t appear until day 27. After that, the cycle speeds up to around 8 days.

Pattern 5: Win the “Explanation” Before Fighting for “Rankings”

A new brand shouldn’t go straight for cutthroat terms like “Best CRM software.” Instead, claim the definitional and educational questions first — “What is X?” “How does X work?” Get AI to treat you as a knowledge source and build trust.

Pattern 6: Close the “Trust Gap” (The Most Important One)

A great site and great content aren’t enough — you need third-party validation. If you’re the only one singing your praises, AI won’t believe you. You need industry press coverage and third-party reviews to tell AI: “See? Other people think I’m legit too.”

5 Steps: Make Your New Brand “Visible” to AI

Now that we know the patterns, what do we do? Here’s the battle-tested 5-step playbook:

Step 1: Map Your Brand Entity

Before you even build the site, define a “brand identity” machines can understand.

  • Use semantic triples: [subject] -> [predicate] -> [object]. Example: “NeoShop -> provides -> cross-border payment services.”
  • Use common language: prefer terms already on Wikipedia or Wikidata — don’t invent obscure words.
  • Establish authority: list 3-5 factual claims that prove why you deserve trust.

Step 2: Design Your Benchmark Prompt Set

Don’t rely only on traditional keyword tools — their data is unreliable in the AI era.

  • Reverse-engineer buyer questions: figure out what your users actually ask in ChatGPT.
  • Build a test set: compile 150 real questions covering brand, category, pain points, comparisons, and more.
  • Test weekly: run those questions through ChatGPT, Gemini, and Perplexity every week and see whether your brand shows up.

Step 3: Make Your Brand “Machine-Readable”

AI doesn’t care if your site is pretty — only if it’s readable.

  • Deploy Schema markup: use Organization, Service, FAQ, and other Schema types to tell AI exactly who you are.
  • Add an llms.txt file: drop an llms.txt in your site root — it’s meant specifically for AI crawlers, outlining your site structure and core pages. (Very new standard — worth watching.)
  • Remove crawl obstacles: make sure robots.txt isn’t blocking AI crawlers, and speed up your site.

Step 4: Publish “Easy-to-Retrieve” Content

Write for the “impatient AI analyst.”

  • Answer first: open each section with a direct, factual answer.
  • Semantic chunking: break content into self-contained logical blocks so AI can extract any single passage.
  • Stay fresh: AI favors content updated in the last 60-90 days. For fast-moving industries like SaaS or finance, refreshing core pages quarterly is a must.

Step 5: Get External Validation

This is the final step — and the one that builds your moat.

  • Complete your business profiles: fill out Crunchbase, G2, LinkedIn, Yelp, and keep the info consistent. Inconsistent information is a leading cause of AI hallucinations.
  • Earn authoritative mentions: get coverage from industry media.
  • Backlinks: aim for at least 3 authoritative backlinks to every core page.

Neo’s Summary

For new brands, the core of AI search optimization isn’t even optimization — it’s building authority.

Don’t start out trying to beat the giants in “best X” recommendation lists. Your first goal is reducing uncertainty.

  1. Build a clean site structure that AI can easily parse.
  2. Write content that’s clear and answers questions directly.
  3. Show up on third-party platforms (LinkedIn, Medium, industry media) to build trust signals.

When you can clearly explain to AI “who you are” and “why you’re trustworthy,” the recommendations follow naturally.

Hope this gives you some inspiration if you’re building an independent site or taking a brand global.

Reference: Search Engine Journal - 5 Ways Emerging Businesses Can Show up in ChatGPT, Gemini & Perplexity