Want your brand’s content to get cited by AI search tools?
The first step is to get LLMs (large language models) to trust your website, which means building online authority.
What’s that?
Online authority refers to the amount of earned trust a brand has by consistently demonstrating expertise, reliability, and credibility across the internet.
High-quality original content, strong user reviews, and backlinks on trusted websites are all ways brands can establish online authority.
Moreover, authority is the cornerstone of AI visibility, and it matters even more than it does for traditional SEO.
On classic search engines like Bing and Google (pre-AI Overviews), authority helped you rank higher, but your content could still appear in the search results without a ton of it.
For example, if your site only had mid-tier authority, you may rank in position #6 instead of #1, but you’d still see some traffic.
However, on AI search tools like ChatGPT, authority is what determines if your content gets cited at all.
In other words, it’s impossible to get cited by AI tools without building authority first.
This is an extremely urgent topic for marketers, which is why we developed this guide.
You’ll learn:
- How authority on AI tools differs from traditional SEO
- The type of online authority LLMs prefer
- Ways to start building authority right now
| Want to start building authority while you read? Check out our Digital PR service to start earning the kind of authority AI trusts. |
How Authority Worked in Legacy SEO
You can think of authority on classic search engines as a spectrum or gradient.
Trust could be earned incrementally, momentum could be built over time, and visibility was shared across the spectrum of organic results.
Even if you didn’t achieve position #1 rankings for all your targeted keywords, ranking in the top 10 could still net decent amounts of traffic and notoriety.
Also, links were the main proxy for authority.
Google’s PageRank connected authority signals to backlinks. The more backlinks you had from ‘authoritative’ domains (which themselves needed backlinks to prove authority), the more trust you built with Google.
At the time, the PageRank system was brilliant because it leveraged existing infrastructure (the link graph) to weigh authority.
Complex semantic understanding and natural language processing (NLP) were decades away, so Google’s search algorithm needed an easy and reliable way to determine the trustworthiness of online content.
Hyperlinks were already something every website used, so it was the perfect infrastructure.
In this type of system, some domains carry more clout than others, which is where the gradient analogy comes in.
It’s possible to amass authority by building backlinks from mid-tier blogs and online directories. Every positive signal helped you climb the rankings, which is why targeting link quantity in addition to quality was a viable tactic on Google for a very long time.
As Google’s algorithm began to incorporate machine learning and NLP elements, things like relevance and topical authority began to matter more. This meant:
- Sites that produced truly helpful content that exhibited real-world expertise began to dominate top search rankings.
- Producing large volumes of mid-tier blogs and backlinks was still somewhat effective at improving rankings, especially when done to supplement higher-quality content.
So, even when Google had sophisticated algorithm components like BERT and MUM, the authority spectrum was still in effect.
It wasn’t until Google’s LLM-powered AI Overviews began to dominate results pages for all types of search intent that the role of online authority shifted:
How Authority Works in AI Search
The transition from classic search algorithms to LLMs marked the end of authority being a spectrum.
Instead, a certain type of authority (largely editorial) became a necessity for getting cited in the first place. If your authority doesn’t meet LLMs’ standards, your content won’t get cited at all, so it’s become a lot more stringent.
Today, it’s more appropriate to think of authority as a binary system instead of a spectrum or gradient. Either your brand is authoritative enough to warrant citations, or it isn’t.
There’s no ‘in between’ like there was before, where you could still rank in the top 10 with ‘so-so’ authority levels.
AI Overviews only cite a handful of trusted sources, and they dominate results pages. Not only that, but according to research by Xponent21, they appear for over 50% of all searches across all types of intent, not just informative.
They’ve also caused zero-click searches to skyrocket, with research uncovering that 60% of all searches in 2024 ended in no clicks (and that was before AIOs really took off in early 2025, so the number is likely much higher now).
TL;DR?
Search visibility has undergone a transformation where most users don’t click on any results and get the information they need from AI Overviews.
The popularity of LLM-powered search tools like Perplexity, ChatGPT, and Claude has only added fuel to the fire.
Now that the organic results have become increasingly irrelevant, brands have to establish authority with LLMs if they want to remain visible online.
Here’s how LLMs weigh authority.
Entity-level recognition
LLMs use entity understanding to identify and disambiguate real-world concepts like people, places, things, and organizations.
Named Entity Recognition (NER) and Entity Linking (EL) are two pivotal parts of this process, and they’re key to establishing relevancy and authority for your brand.
In order to appear as a cited source for relevant prompts, LLMs first need to identify your brand and link it to an associated industry or niche.
Here’s how they work:
- NER detects and classifies entities in text.
- EL matches these entities to a knowledge base like Wikidata or Google’s Knowledge Graph.
When it comes to entity understanding, consistency is crucial, especially for how you present your brand name.
Always refer to your brand the same way in your content, schema markup, and citations on other websites. Also, ensure your entry in Wikidata is accurate and consistent with your other branded information.
If you don’t have a Wikidata entry, create an account and make one. In your entry, link to your Wikipedia page or other verifiable branded sources online.
This will make the entity recognition process nice and straightforward for LLMs.
Brand mentions in trusted publications
Both NER and EL are about properly recognizing your brand as a relevant entity within your industry. Neither actually determines whether you’re a credible source or not.
That comes later during the entity ranking process. It also goes by the names authority weighting, authority scoring, and entity salience, but they all refer to the same thing.
It’s where the LLM evaluates your trustworthiness, and the most significant way it does so is by evaluating your branded mentions online.
These are your backlinks, unlinked brand mentions, reviews, product recommendations, and random shout-outs on social media and Reddit forums.
Entity ranking involves:
- Checking source reputation – How trusted and respected are the websites mentioning your brand?
- Verifying relevance – Are the websites mentioning you contextually or topically related to what you do?
- Evaluating brand sentiment – Is the general discussion surrounding your brand mostly positive?
These processes help determine whether your content should appear in things like AI Overviews and ChatGPT responses.
User trust signals
LLMs lean on user reviews and community-level discussion on platforms like Reddit to build trust with brands.
If you have lots of positive reviews and a strong brand sentiment, LLMs are far more likely to surface your content in their responses.
That means brands need strong reputation management if they want to improve their AI visibility. You should always be A) finding ways to generate more positive reviews and B) immediately responding (and trying to resolve) any negative criticism.
The Top Ways to Build AI Authority Today
Now that you know what LLMs are looking for, let’s learn some candid ways you can start building authority.
Digital PR and earned media
Digital PR and AI SEO are a match made in heaven. Virtually every practice involved with digital PR will help build the kind of authority you need to earn AI citations.
A major aspect of digital PR is generating brand mentions, backlinks, and media placements on trusted outlets, which are exactly the kind of trust signals LLMs use.
Digital PR is also about attracting attention through expert interviews, podcast appearances, newsjacking trending stories, and posting on social media. This type of channel diversity also helps strengthen online authority and can help establish your brand as a major player in your field.
Producing long-form, expert content (that’s structured for AI visibility)
Topical authority matters to LLMs, and they love to cite brands that have large libraries of original, insightful content covering the same topics.
That’s why you should invest in producing thought leadership content that covers the same topics, answers common questions, and contains insights that readers (and LLMs) can’t find anywhere else. Developing original research is a great way to go about this, as long as the data is relevant.
Online content creators link to original studies and surveys all the time, so producing research is a surefire way to attract lots of backlinks and brand mentions (further strengthening your authority).
However, you need to structure your content so that it’s LLM-friendly and machine-readable. That means using semantic HTML, schema markup, and clear formatting (H1s, H2s, H3s, bulleted lists, short paragraphs, etc.).
Also, answer each question immediately after posing it. That’s because LLMs prefer clear, self-contained Q&A pairs. This makes it easy for an LLM to find and cite juicy snippets for its generated responses.
Speaking engagements
If you have the ability, participating in speaking engagements can work wonders for your online authority. That’s because they fire on multiple authoritative cylinders, including:
- Beefing up author credibility – LLMs analyze author bios to establish credibility, so being able to mention your participation in relevant speaking engagements is a huge win.
- Generating a buzz and brand mentions online – Speaking events attract lots of online attention via social media posts, videos, blog recaps, and brand mentions.
- Aiding entity recognition – A great perk of engaging in speaking events is that they will tie your brand name to respected conferences in knowledge bases.
- Validating your brand’s expertise – Being featured at prestigious conferences reflects well on your brand’s experience and expertise (core components of E-E-A-T).
Case Study Example: MedTech
Does improving your authority really lead to improved AI search visibility?
Let’s examine a real-world instance where this was exactly the case.
Healthcare SaaS company MedTech Solutions decided to optimize for AI search over a 5-month period. Before the campaign, they were barely visible in AI search, as they only appeared for 2% of AI Overviews related to the healthcare SaaS space.
This was despite the fact that they had extremely strong traditional SEO, which goes to show that strong organic SEO doesn’t always translate to strong AI visibility.
After improving their content, strengthening their entity profile, and fine-tuning their link profile for relevance, they started to see impressive results.
They achieved a 289% increase in AI Overview citations and saw a 523% increase in qualified leads from AI traffic (AI Mode Boost):
That’s a huge improvement from being practically invisible in the AI search space. Once they improved their authority, better AI visibility and lead generation followed.
How The HOTH Builds Authority on AI Search Tools
To summarize, online authority is a must-have for AI search, as it’s increasingly becoming a zero-sum game (either you have it or you don’t).
We’ve also noticed firsthand that authority improves AI search visibility. One of our clients saw a major increase in AI Overview visibility (an over 950% increase in citations) after one of our managed SEO campaigns.
This inspired us to compile our AI SEO best practices into a single product: AI Discover.
It combines Digital PR staples with expert blogs to deliver the authority boost you need to start earning valuable citations and direct brand recommendations.
AI Discover is the perfect way to make the transition to LLM-powered search without having to do any heavy lifting, so don’t wait to sign up!
