Original research

The Founder Authority Engine: Methodology and Results

AA

Adev Aarons · Founder, Specularis
August 30, 2026 · Updated August 30, 2026 · 5 min read

By Adev Aarons, Founder of Specularis · Updated August 2026

When a buyer asks ChatGPT “who is the best structural steel contractor in Denver” or “which agency should I hire for AI visibility,” the model names one to three companies. The Founder Authority Engine is the system for making sure one of those names is yours. It is not about ranking on Google. It is about being the answer an AI gives when nobody typed your name.

What is the Founder Authority Engine?

The Founder Authority Engine is a repeatable method for getting a founder and their business cited by AI answer engines like ChatGPT, Claude, and Perplexity. It works across six signals the models actually use to decide who to recommend. Fix the signals in the right order, and you move from invisible to cited on the questions your buyers ask.

The name has two parts on purpose. Authority is what AI is really measuring: does the wider web treat you as a credible source. Founder is the fastest lever, because a founder with a clear, machine-readable story becomes an entity the model recognizes, and that recognition pulls citations toward every company they touch. A competitor can copy your homepage. They cannot copy the fact that you have run the business for a decade.

How is it different from traditional SEO?

SEO earns a click by ranking a page in a list of ten. The Founder Authority Engine earns a citation by making you the answer in a list of one to three. That difference changes everything about what you optimize.

On Google, being the tenth result still gets traffic. In an AI answer, being the tenth-most-relevant company means you are never mentioned. The models do not show a page of options. They pick a few and explain why. So the work shifts from “rank this page for this keyword” to “make the whole web agree that this founder and this company are the credible choice for this need.”

The signals overlap with SEO. The goal does not. You are optimizing to get quoted, not just ranked.

The methodology: six signals AI uses to decide who to cite

Every AI answer engine is doing the same thing under the hood. It reads your site, works out who you are, checks whether your content is easy to quote, looks for other sources that back you up, checks whether you are current, and confirms the technical basics. We score each of those as a weighted pillar out of 100.

The weighting is the important part. Off-site corroboration carries the most because it is the only signal that measures whether the rest of the web already vouches for you. A perfect website with no one else talking about it still loses. This is why so many businesses do the technical work, pass the crawler and schema checks, and stay invisible anyway. Being readable is not the same as being recommended.

The branded versus unbranded gap

The clearest evidence for why off-site matters comes from outside our own data. Foundation Marketing and AirOps analyzed 57.2 million AI citations in 2026 and found that a brand’s own website is cited about 78 percent of the time when someone asks about that brand by name, and only about 2 percent of the time on unbranded category questions. That is a roughly 14x gap.

In plain terms: AI already knows you when a buyer types your name. It does not surface you when a buyer asks who is best. And unbranded discovery is where new customers come from. Closing that gap is the whole job, and it lives almost entirely in the off-site pillar: category-authority content on your own site, plus earned citations across the sources the AI already trusts.

What the data shows

The pattern holds across everything we test. In a study of 561 Florida businesses, we checked each site the way ChatGPT reads it. Close to half were effectively invisible to AI on the questions their buyers ask, and a large share were actively blocking the AI crawler in their robots file without realizing it. These were not failing businesses. Many had strong reviews and steady Google traffic. They were simply absent from the layer where more and more buying decisions now start.

Research backs the fix. The Princeton GEO study found that adding statistics, quotations, and source citations to a page lifts its inclusion in generative-engine answers by 30 to 40 percent.

“Adding statistics, quotations, and source citations lifts generative-engine inclusion by 30 to 40 percent.” — Aggarwal et al., 2023

We see the same thing in placement. Roughly 44 percent of the sources AI cites come from the first third of a page, which is why the Founder Authority Engine leads every page with a clear, quotable answer instead of a slow warm-up.

How to apply it

The order matters more than the effort. Start where the points are.

1. Make sure AI can read you. Check your robots file is not blocking GPTBot, ClaudeBot, or PerplexityBot. This is a five-minute fix that quietly kills visibility.
2. Make your identity unmistakable. Add Organization and Person schema so the model knows exactly who the company and the founder are, and resolve any name collisions.
3. Lead with quotable content. One clear H1, an answer-first opening, statistics, a data table, and outbound citations to authoritative sources.
4. Build off-site corroboration. Get named in the directories, roundups, and communities the AI already pulls from. This is the slowest and the most valuable.
5. Stay current. Dated content and updated schema tell the models the site is alive.

You cannot track visibility you do not have yet. The fastest way to see where you stand is to measure it.

Adev Aarons is the founder of Specularis, an AI visibility company that gets founders and businesses cited by ChatGPT, Claude, and Perplexity. Run a free audit.