Where I write

The method lives on LinkedIn. This page just points you there.

This site updates slowly, on purpose — the actual writing happens on the channels below, on a real cadence. Follow there for anything current; the two posts below are permanent, not a stream.

Follow along

Where the writing actually happens

Posts, threads, and comments applying the method to whatever's current — this is where new entries in the case log show up first.

LinkedIn

~3x / week

The primary channel. Full posts applying the method to current marketing, AI, and comms stories — plus comments and discussion on other people's posts.

Follow on LinkedIn →

EM Digest

LinkedIn · Weekly

Before I trust a marketing trend, I ask who benefits. Weekly notes on what I find — delivered straight to your LinkedIn feed and inbox.

Subscribe on LinkedIn →

X / Twitter

~2x / week

Shorter threads distilling the same arguments — good for a faster read or for following the discussion as it happens.

Follow on X →
Start here

Two posts, kept permanent

Everything else lives on the channels above and will move on. These two stay — one predictive call, one self-critique — because together they're the clearest proof of how the method actually works.

Case 01 · Self-critique

I was confident paid social would convert. It didn't — and here's why I believed it anyway.

A few years ago I was confident paid social would drive conversions for a flagship enterprise event I was running marketing for. It's the standard modern-marketing move — every platform, every case study, every account rep tells you social converts. So we invested.

The engagement numbers looked great. Awareness climbed. What didn't show up was proof of actual conversions — nothing I could trace from an ad to a closed deal.

The moment that actually forced the question wasn't a dashboard. It was my boss, in a meeting, asking me directly to show the connection between the spend and the pipeline — and realizing I couldn't. That's a worse feeling than any chart. I'd been confident in a channel because it's what "good digital marketing" is supposed to look like, not because I'd tested whether it fit how our actual buyers were deciding.

We shifted the approach — dynamic ads and customized landing pages built around specific target-account signals instead of broad reach. Conversions became visible almost immediately, because I could finally trace them: the accounts we were deliberately targeting were the ones converting.

I still don't think paid social was the wrong channel in the abstract. I think I borrowed a tactic built for short, impulse-adjacent decisions and applied it to a long, multi-stakeholder enterprise buying process, because that's what the playbook says to do — and platforms selling ad inventory have no reason to correct you.

I don't have a clean rule to hand you out of this. What I have is a habit now: before I commit budget to a channel because "it's working for everyone," I ask who actually benefits if I believe that — and whether it's the buyer I'm trying to reach, or someone selling me the channel.

Case 03 · Predictive call

GEO leaderboards are a snapshot, not a strategy.

A few months ago I published an analysis questioning why marketers were building GEO strategy around Reddit's citation share in AI search results, while most GEO advice was still saying "double down, it's working right now."

It wasn't a prediction. It was arithmetic. Citation share isn't earned authority — it's the output of one company's retrieval decisions, which can and do change without warning, and without any obligation to the marketers who built a strategy on top of them. Treating a citation-share snapshot as a stable hierarchy was always going to be a bet on someone else's engineering roadmap.

Since then, Reddit's presence in AI search citations has fallen sharply, tracking a retrieval-method change on the platform side — exactly the kind of shift that "go where it's citing today" advice can't protect you from, because it's optimizing for a photo of a moving target.

Worth asking honestly: who benefits from marketers believing GEO leaderboards are stable? Not the marketer building on one. The advice sells easily because a leaderboard sounds authoritative — but in this system, "authoritative" just means "whichever retrieval decision a lab shipped last quarter."

The actual strategy was never "chase this month's winner." It's building a presence that doesn't depend on any single platform's retrieval logic staying the same — because it won't.

Stay in the loop

Everything current lives on LinkedIn and the newsletter — this page won't.