Case Study Zero
I run the method on myself first.
The fair test of a method is whether the person selling it has pointed it at their own business first. So before I quote a price, my own site is the first subject — the Citation Engine, run on patrickrobinson.consulting, in public, from a measured start.
Real before-and-after only. No sample data, no numbers a chatbot made up for a slide. The baseline below is a real zero, and it can't be quietly edited later.
Citation rate today
0/7
category buyer queries where an AI engine currently cites my work, across Perplexity, Claude, ChatGPT and Gemini.
Baseline measured . This number is the experiment — when it moves, you'll see it here first.
The series
Episode 01
The honest baseline
Why I run the Citation Engine on my own site before selling it to anyone — and the unflattering place I'm starting from: cited in zero of seven category buyer queries.
Read the transcript
I'm Pat, and welcome to this explainer on Case Study Zero: proving the Citation Engine. I have a strictly held core philosophy that guides absolutely everything I do in this space, and it's this: I absolutely refuse to sell a method without running it on my own business first. You see, in the world of AI search engines, transparently showing your work is the literal only way to prove a method actually does what it claims.
And that brings me to the principle of proof. My own site, patrickrobinson.consulting, is acting as the very first subject for this methodology. Long before I ever framed an offer or quoted a price to a single prospect, I made my own business the test case. Because like I said, I refuse to sell a method I haven't run on myself. I call this whole process Case Study Zero. Executing this in full public view is honestly the only fair test of the work. Anyone can sit here and describe a theory, but far fewer will actually run it on themselves, where the results are completely visible for everyone to scrutinise.
Which leads right into the honest baseline. To do this right, I began exactly where I tell all my prospects to start. Because simply put, you cannot improve what you don't accurately measure. So I ran the exact same rigorous audit that I run for clients. I tested buyer-intent queries across all the heavy hitters: Perplexity, Claude, ChatGPT and Gemini. I focused on these because they are the primary AI discovery engines, really shaping how buyers find information right now. Doing this establishes a comprehensive, accurate baseline for my own consulting category.
And the reality? Well, it was completely unflattering. Zero. Out of seven distinct category buyer queries, I was cited zero times. Now, I am highly deliberate about publishing this exact baseline, because I absolutely will not stand behind a method that uses sample data or padded numbers spit out by a chatbot. Zero is a real number, and it's a number that cannot be quietly edited or massaged later.
Moving on to working in the open. Rather than trying to hide that embarrassing baseline, I'm using a completely transparent process to apply my method and fix it in public. I'm running the exact 30-60-90 sequence I hand to my clients. Step one focuses purely on surface presence, which yields fast results and basically just gets you on the board. Step two then moves into citation-shaped depth. That's crucial, because it's what makes your presence truly durable and defensible over the long term.
Now, doing this requires a fundamental shift in approach. I'm completely abandoning those old keyword-stuffed marketing pages. AI engines just don't care about those any more. What they actually cite is long-form, founder-bylined content that contains numbered claims, structured arguments and highly specific reasoning. So that's exactly what I'm building. The visible work taking place right now consists of published methodology essays. I'm publishing these live on my site as I go. They are engineered specifically as citation surfaces, not your typical blog posts chasing cheap clicks. You literally get to read the working here, not just the final conclusion.
And this is the proof that matters. Building this citation profile in real time, starting right from the absolute bottom, matters directly to anyone considering the service. Why? Because it completely shifts the entire burden of proof. It answers the most obvious question on everyone's mind: does this even work? The core risk of adopting any new method gets answered right here, simply by watching it happen to the operator first. The risk is entirely on my shoulders, not yours.
My site is being authored into citation in real time. Remember, this is a method in motion, measured against an embarrassing, publicly recorded baseline. It proves its citation is built one sourced piece at a time. I'm not claiming a finished win here today. I'm not implying a completed result, or trying to tease a specific improved citation number. I'm simply showing you the honest process as it happens. You don't have to take the method on faith. You can watch it run on my own site. You can watch the baseline, the ongoing work, and the real before-and-after as it lands.
If you want this exact same method pointed at your company, it begins with an audit. What will your baseline reveal?
Episode 02
Upcoming
The first citation
Publishes the day an answer engine first cites my own work. Real before-and-after only — so this lands when there's a genuine after to show, not before.
Locked until the first citation lands. No teaser, no placeholder result — it publishes when there's something real to show.
1 of 2 episodes published. New episodes are triggered by real citation events, not a calendar.
The rules of this experiment
Real before-and-after only. The baseline is measured the same way I measure a client's: buyer-intent queries run across four answer engines, multiple times each. The "before" is a real zero.
Surface presence first, then citation-shaped depth. The same 30-60-90 sequence I'd hand a client — fast wins on the surfaces engines pull from, then the deeper, founder-bylined work that earns durable citations. The essays on this site are part of that depth.
In the open, where it can't be edited later. Each episode documents a real movement in the number above. If the method is good, it has to move a figure I published before I knew where it would land.
Want this run on your company?
The Citation Engine starts with the same audit I ran on myself — your citation rate across the four engines, benchmarked against your category, with the gaps and the fix mapped out.