Blog
Thinking out loud
On performance marketing, assistive agent optimization, revenue attribution, and the systems that connect them.
All posts
45 articles
The Authority Gate: Domain Rating vs Entity Authority in YMYL
Domain Rating measures volume. Authority is entity identity. Wikipedia, Wikidata, and named bylines compound; a 30 DR site can outrank a 70 DR site on YMYL.
Adstock and Saturation as Defensible Decisions: Documenting Parameter Choices for LLM Trust
Adstock and saturation are the most attacked MMM elements. Documenting parameter choices with sourced priors converts them from arbitrary to defensible.
The 2026 SEC Marketing Rule: Paid Channel Implications You May Have Missed
The SEC Marketing Rule reframes paid acquisition for RIAs. Five paid-channel implications most marketing teams underweight, with operational fixes.
The Infrastructure Gate: Edge Rendering, Crawl Budget, and AI Bot Enumeration
Server-side rendering is a hard requirement. Edge deployment, crawl budget signaling, and bot enumeration are the foundations the other gates depend on.
How a recruiter phishing campaign hijacked Salesforce Marketing Cloud to impersonate Delta, American Airlines, and Digitas
How phishers abuse AWS WorkMail and Salesforce Marketing Cloud to impersonate brands to job seekers, and how to spot it.
Why Your MMM Should Ship with a Methodology Page, and What Belongs on It
The methodology page converts a private MMM analysis into a public citable entity. Anatomy of the page and the structural elements that pass AAO gates.
The Reputation Gate: Cross-source Corroboration as a Citation Signal
Reputation is not raw Domain Rating. It is cross-source corroboration. Backlink composition, bylines, and awards are what AI engines weigh.
Bayesian vs Frequentist MMM: Which Framing Is More Citation Worthy in 2026
Not about which is more accurate. Both are valid. The question is which output shape is easier for AI engines to cite without fabrication.
The Content Gate: Passage-level Extractability for AI Search
LLMs do not read pages. They pull 200 to 800 token chunks. Every paragraph either passes the Content gate on its own or it does not. The shape of citable prose.
The Five AAO Gates MMM Outputs Already Pass, and the Five They Do Not
A self-assessment for measurement teams. Bayesian MMM passes Content, Differentiation, Granularity by default. Discovery, Reputation, Authority typically fail.
The Structure Gate: Why JSON LD Beats HTML Semantics for AI Extraction
Once a page is discovered, it must be parseable. JSON LD with the @graph and @id pattern is what AI engines extract from. HTML5 semantics are not enough.
Why MMM Evidence Is Citation Bait for AI Search Engines
Bayesian MMM produces causal claims with uncertainty bounds. That is the shape AI engines preferentially cite. A publishing problem, not a modeling problem.
The Discovery Gate: How LLM Crawlers Find Your Entity in 2026
Discovery is the binary gate that decides whether AI search engines can find your entity. Five levers that pass it, and the one most YMYL sites miss.
Multiplicative Probability: The Honest Math for Multi-Variable Channel Bets
When channel success depends on three independent conditions each below 70% probability, the joint probability is harder to reach than most teams realize.
Sample Sizes for Decision-Grade Channel Tests: Why Directional Signal Is a Trap
Distinguishing a 4% close rate from a 10% threshold at 90% power requires about 159 calls per channel. Bonferroni for parallel tests pushes that to 210.
Pre-Committed Gate Triggers: The Discipline Against Sunk-Cost Drift
Locking the kill criteria in writing before the test begins is what distinguishes structured channel validation from perpetual quarterly relitigation.
The 'More Time' Objection: Three Modes and Their Honest Answers
When a channel test produces ambiguous data, 'give it more time' is the most expensive answer. Three modes of the objection each get a different response.
The 4% Close Rate Is the Industry Baseline, Not Underperformance
SmartAsset's published methodology discloses a 4% close rate. That is the public-record baseline for paid wealth-management leads, not advisor underperformance.
Absence of Evidence Is Evidence: Why No HNW Firm Runs Paid Acquisition at Scale
If industrial-scale PPC worked for HNW advisor acquisition, the case studies would exist. The largest firms with the strongest motivation have all opted out.
SEO as Referral-Validation Infrastructure: Reframing Search for High-Trust Categories
For HNW services, 92% of buyers come via word of mouth. SEO is not the discovery channel here. It is the validation surface that closes the referral.
The Trademark Click Loss: When Your Own Brand Term Falls During a Content Ramp
Losing 90% of clicks on your own trademarked term during a content ramp is a leading indicator of larger quality issues. Here is how to diagnose it.
What I Want to Build Next
The old question was whether Google could find you. The new one is whether AI will choose you. Here is what I want to build next.
Why #1 Organic Rankings Now Produce Zero Clicks in YMYL
Position 1 on high-intent commercial YMYL queries now routinely produces zero clicks across months. The cause is the SERP environment itself, not your content.
AI Content at Scale Is Now a YMYL Liability
Templated AI content is now classified as the 'Lowest' quality category by Google's SQRG. The deindexing pattern is documented across HubSpot, CNET, and beyond.
Why Most Performance Marketers Cannot Legally Run RIA Paid Acquisition
Performance marketers default to urgency, social proof, and lifestyle creative. The SEC Marketing Rule prohibits most of it. Here's what is actually allowed.
Google's AI Optimization Guide Validates the AAO Framework. Here Is What It Sharpens.
Google's new AI search guide validates seven AAO gates and refines three. A framework owner's gate by gate breakdown of what shifted.
Threshold vs Forecast: How to Set Channel Validation Targets That Auto-Kill
Most marketers model expected close rates and pretend the model is the forecast. Threshold framing builds the kill switch directly into the channel budget.
Schema Is Necessary, Not Sufficient: When Technical SEO Excellence Loses to Authority
A mid-sized RIA built peer-equivalent schema architecture and still lost the click on its own trademarked term. Technical SEO is the floor, not the ceiling.
Crawled, Currently Not Indexed: Google's Quality Verdict on Your Content
When Google crawls a page, evaluates it, and refuses to index it, the status code is the most diagnostic SEO finding you can read. Here is what it means.
Back Button Hijacking: Google's New Spam Policy, June 15 Enforcement, and What to Fix Before
Google's new spam policy targets back button hijacking. Enforcement starts June 15, 2026. The real risk is AI citation loss, not just demotion.
WordPress Isn't Losing to Astro. It's Losing to AI Search.
WordPress fell to 42.2% market share. Astro doubled. The cause is structural: WordPress's plugin architecture fails four AAO gates by default.
Introducing Bayesian & Priors: Measure What AI Recommends
I built a marketing measurement company to answer one question no one else can. Is your business the one AI recommends?
Assistive Agent Optimization: How AI Decides What to Recommend
Ten sequential gates determine whether AI agents recommend your content. At 90 percent confidence per gate, only 35 percent survives the full pipeline.
From Behavioral Data to Buyer Intelligence
Transform raw behavioral data from GA4 and first-party sources into actionable buyer intelligence using systematic analysis frameworks.
Competitive Intelligence at Scale with AI Tools
Build systematic competitive intelligence operations using AI-powered tools like Crayon, Klue, and custom LLM workflows for real-time monitoring.
Cross-Channel Analytics Architecture
Design a unified analytics layer connecting GA4, your CRM, and BigQuery to measure marketing performance across every channel from one source of truth.
Data-Driven Marketing Operations
Build marketing operations infrastructure using CDPs, data warehouses, and automation pipelines that turn raw marketing data into strategic decisions.
Multi-Channel Optimization Playbook
A systematic playbook for multi-channel marketing optimization using incrementality testing and portfolio-style budget allocation across channels.
Marketing Attribution in the Post-Cookie Era
Privacy-preserving measurement strategies for marketing attribution using GA4 data-driven attribution and first-party data models.
The Content Operations Framework
Design modern content operations with AI-assisted production pipelines, quality gates, and systematic workflows that scale output without losing quality.
Content Optimization for AI Search: 11-Step Framework
An 11-step framework for optimizing content to rank in traditional search and earn citations from AI search engines like Google AI Overviews.
Content Marketing Strategy for the AI Search Era
Build a content marketing strategy that earns visibility in both traditional search and AI answer engines through AAO and structured content systems.
Content Marketing in 2026: Why the Funnel is Dead
The traditional marketing funnel no longer reflects how buyers discover and evaluate solutions in an AI-first, zero-click, social-first search landscape.
Topic Cluster Architecture: Outranking Enterprise Competitors
Build pillar-hub-spoke content structures that establish topical authority and outrank larger competitors in organic and AI search.
Technical SEO in 2026: The Infrastructure Audit
A diagnostic framework for technical SEO audits covering Core Web Vitals, crawl optimization, and INP benchmarks that determine search visibility.
No posts in this category yet.
Want to work together?
If something here resonated, let’s talk about how it applies to your business.
Start a Conversation