Maria Dykstra
AI Visibility Architect · Creator of the Algorithmic Authority Stack™
13 years inside Microsoft working on the systems that decided what reached buyers at scale. Now diagnosing why B2B companies disappear from AI-generated shortlists.
Maria Dykstra diagnoses why B2B companies are invisible to AI systems. 25+ years studying how information systems decide what gets seen, cited, and skipped. First in advertising infrastructure. Now in AI.
At Microsoft, Dykstra was in the rooms where probabilistic ad delivery and search-as-buying-signal got built. Across 13 years and four roles, she worked at the intersection of advertising infrastructure, search, and scale before any of it had a category name.
The systems are different. The structural logic is not. Both depend on signals, classification, relevance, and confidence under uncertainty. The systems that decided which ad reached which buyer now decide which company AI cites.
Three threads from the Microsoft years map directly to today's AI visibility work:
- Search advertising as it emerged as a buyer-intent signal
- Product roadmap for media selling tools driving $2 billion in annual ad revenue across 36 global markets
- IAB cross-industry workgroup on digital attribution standards
After Microsoft, Dykstra founded TreDigital in 2012, a digital agency she ran for 13 years scaling Fortune 500 companies and early-stage startups across digital strategy, growth systems, and visibility architecture.
Today, Dykstra operates as an independent AI Visibility Architect. She created the Algorithmic Authority Stack™, a 7-layer diagnostic framework for identifying why B2B companies are structurally invisible to AI systems like ChatGPT, Perplexity, and Google AI Overviews. She has applied the framework to 50+ B2B companies across healthcare, SaaS, financial services, professional services, and technology.
What Pattern Recognition at Scale Looks Like
Dykstra has diagnosed the same five structural failure patterns across every AI visibility audit she has run. They appear in funded SaaS companies. They appear in established professional services firms. They appear in companies that have been publishing content for years.
Every one of them is an architecture failure, sitting underneath the content.
The company uses three to five different descriptions of itself across surfaces. AI systems cannot classify an entity that contradicts itself. They default to competitors who do not.
AI reads the range as contradiction.
Core concepts get described with inconsistent terminology across pages, teams, and time. Each variation registers as a separate topic. Authority diffuses instead of compounds.
Your authority does not compound. AI cannot connect the evidence.
Content volume without underlying structural architecture. Publishing more content into a broken architecture produces more invisible content. Structure decides whether any of it registers.
More content can make the problem worse.
Expertise exists. Proof signals do not. AI systems require verifiable, cross-referenced evidence of claimed authority. A structured entity node supplies that. A bio paragraph reads as an assertion.
AI reads your claims without enough reason to repeat them.
Language intentionally broad enough to appeal to human buyers registers as unclassifiable to AI systems. "Intelligent solutions for modern businesses" is semantically empty. AI skips it.
A buyer may understand the vibe. AI cannot build a recommendation from a vibe.
Career
The Algorithmic Authority Index
In 2025 and 2026, Dykstra conducted Wave 1 of the Algorithmic Authority Index, a systematic study of AI visibility performance across B2B companies. Distinct from the 50+ client audits, the Index is the empirical research foundation behind the framework.
Twelve of the 20 companies showed inconsistent or incomplete AI descriptions when tested against category and buyer-intent prompts on ChatGPT, Perplexity, and Gemini.
The findings form the empirical foundation of the Algorithmic Authority Stack™ and the Algorithmic Authority Audit™. Wave 2 initial findings published July 2026, with additional industry reports to follow. Read the Index findings.
Four anonymized client diagnostics. Four different broken chains.
Real company examples showing how the same Algorithmic Authority Stack diagnostic finds different weakest layers across industries. Additional case studies are being added as engagements complete.
- Industrial B2B · Industry Leader Displaced by a Newcomer · L1 Identity Fragmentation →
- SaaS Hybrid · 47-point B2B vs B2C citation gap · L1 Entity foundation →
- B2C Consumer Goods · The Polyglot Penalty · L1 →
- B2C Regulated Retail · Editorial authority gap · L4 Training-Ready Content →
Core IP & Frameworks
The Algorithmic Authority Stack™
7-layer diagnostic framework for identifying and fixing AI visibility failures. The structural model behind every audit.
The Algorithmic Authority Audit™
Diagnostic service applied to 50+ B2B companies. Layer-by-layer visibility analysis across ChatGPT, Perplexity, and Google AI Overviews.
The Algorithmic Authority Index
Proprietary research study. Wave 1: 20 companies, 5 industries, 3 AI platforms, 420 assessments. The empirical foundation of the framework.
The Invisible Audience
The book. How AI decides who gets seen and who gets skipped. Forthcoming October 2026.
Previous Press & Recognition
Earlier writing and commentary, predating current AI visibility work. New press coverage forthcoming with the October 2026 release of The Invisible Audience.
LinkedIn Top Voice in Digital Strategy (August 2024). LinkedIn Top Voice in Content Marketing (June 2024).
Board Member, Women in Wireless. Director, Seattle Chapter, Founder Institute.
By the time a buyer visits your site, AI may already have decided whether you belong in the shortlist.
The Algorithmic Authority Stack shows where AI fails to classify, verify, or distinguish you. Both tracks run the same three steps at the same prices. The difference is the subject: the company's entity, or yours.
If you are a B2B company that needs AI to recognize the corporation.
- AI Visibility Snapshot · 48 hours$497
- Algorithmic Authority Audit · 3 weeks$4,500
- Implementation · 90-day initial term$4,000/mo
If you are a founder or fractional executive who needs AI to recognize you personally.
- AI Visibility Snapshot · 48 hours$497
- Algorithmic Authority Audit · 3 weeks$4,500
- Implementation · 90-day initial term$4,000/mo
The Snapshot credits in full toward the Audit if you upgrade within 30 days. Implementation continues month to month after the initial term, with 30 days notice to stop.
Prefer to talk first? Book a 15-minute call.