The Tech Founder and The Invisible Audience

Your Next Buyer Already Asked AI About You. Did You Make the List?

A new book by Maria Dykstra, AI Visibility Architect, argues that AI doesn’t so much overlook B2B companies as misread them, and that much of what it recommends comes from vendors grading their own market.

Picture a CFO shortlisting fraud-detection vendors. Before she visits a single website or takes a sales call, she types a question into ChatGPT: which companies should I consider? Within seconds, she has a tidy list of five names. If yours isn’t one of them, you may never learn she was looking.

That moment is the subject of The Invisible Audience: Why AI Leaves You Off the List, and What Actually Matters, a new book from digital marketing expert and author Maria Dykstra, AI Visibility Architect. Dykstrahas spent years examining why some strong B2B companies vanish from AI-generated recommendations while their rivals appear.

Her conclusion runs against the usual advice. “Most companies assume their problem is getting found,” she said. “More often, the machine finds them and gets them wrong.”

Who’s really behind the recommendation

The book’s sharpest finding comes from Dykstra’s ongoing study, the Algorithmic Authority Index. She traced 1,855 citations across 160 AI answers to buyer questions about which vendors to choose, then examined 40 of the source pages in depth.

Thirty of those 40 pages, or 75 percent, were published by companies selling into the category they were writing about. Twenty-two ranked or included the publisher’s own products. The neutral-sounding “best tools for X” roundup, in other words, is frequently written by one of the contestants.

“Before your buyer ever sees you, an AI system has already decided whether you belong on the list,” Dykstra said.

For founders, that cuts two ways. AI shortlists are less objective than they look. But the companies shaping those lists are doing something that can be learned.

Seven places it breaks

Dykstra’s answer is the Algorithmic Authority Stack, a seven-layer diagnostic she has applied to more than 50 B2B companies in industrial, healthcare, fintech and cybersecurity markets. The layers ask whether AI can tell what a company is through renames and parent brands, verify what it knows, match the buyer’s vocabulary to the company’s, lift a clear answer from its pages, find it on the platforms buyers actually use, weigh outside proof about it, and measure any of this accurately.

Across those companies, she found, size never decided the outcome.

The book also separates being ranked, mentioned, cited and recommended, arguing that only the last earns a place on the shortlist. It closes with a first-30-days plan for teams that don’t yet know what’s broken, and a look ahead to AI agents that build vendor lists with no human in the loop.

Dykstra comes to the problem from the inside. She spent 13 years at Microsoft on advertising and search systems, including a role overseeing the roadmap for media selling tools behind $2 billion in annual ad revenue, before founding and running TreDigital for another 13 years.

The Invisible Audience is available in paperback ($14.99) and on Kindle ($9.99) at Amazon. More at mariadykstra.com.