I’ve been in a lot of conference rooms over the last two years. Zero Click in San Francisco. SEO Week in New York. BrightonSEO in San Diego. Good rooms, full of people doing real work….
And in every one of them I kept meeting the same person.
A director or a VP at a software company who just got handed “AI visibility” by their CEO. They have a spreadsheet of prompts they run by hand on Sunday nights. They have a theory about why ChatGPT keeps recommending a competitor. They have nobody at their company who does what they do… and nobody at the conference does either, because the person in the next seat is an agency owner, a tool vendor, or a consultant with a deck.
We’re building the room for those people.
Citation Conference happens in Toronto on May 11 to 13, 2027. It’s Foundation’s first marketing event. There are 200 seats, one track, and no expo hall. Tickets go live on October 26, and the waitlist is open now.
Here’s what it is, why it’s shaped the way it is, and the one thing I’d ask you to do if any of this sounds like you.
Search rewarded the click. AI rewards the source.
For twenty years marketing had a scoreboard. Rank, click, convert. You could argue about attribution all day, but everyone agreed what the game was.
Then the answer engines showed up and shortened the runway. If your brand is missing from the answer, there’s no page two. There isn’t a page one either. There’s a buyer who never knew you existed.
We’ve been doing this work at Foundation since before it had a name. We called Reddit a B2B channel in 2018 and got laughed at. We built GEO strategies a year before the industry settled on an acronym. And our research team has spent 2026 publishing the receipts: 380,000 AI answers and three million citation events, with the finding that roughly 90% of what AI cites about a B2B brand lives somewhere the brand doesn’t control.
That’s the context. The conference exists because the people doing this work are figuring it out alone, and alone is slow.
Why 200 seats and not 2,000
Every marketing conference wants to be the biggest one. I understand the math. More seats, more sponsors, more revenue, and a room where the person next to you is a sales rep working the hallway.
I’m doing the opposite on purpose. 200 seats.
Our team reviews every request one at a time, and the target is a room that’s 70% in-house marketing leaders who own the AI visibility budget, with a handful of invited friends. Agencies are mostly outside the audience here, which is a strange thing for an agency founder to write. But I’ve watched what happens to a room when half of it is selling to the other half, and it isn’t learning.
Every table of ten gets filled with people who have skin in the game. That’s the whole design.
The formats were built for candor
I borrowed from the events I’ve loved and tried to fix the thing that made me leave the rest early.
- Off the Record. No cameras, no recordings, no slides posted afterward. Speakers share the tactics they haven’t published and the numbers they can’t put on LinkedIn. These sessions are never sponsored, and that’s what makes them work.
- Table Talks. Ten to a table, facilitated by channel and by problem. YouTube, Reddit, LinkedIn, forecasting, measurement. Peers only, notes traded in the open, nobody pitches. This is the part I expect people to talk about on the flight home.
- The Midday Report. A data session delivered from the main stage. Findings nobody in the room has seen yet. It becomes a published report after the event, so the room gets it first and the industry gets it second.
- Citation Eve. A hosted evening on May 11 for everyone who flew in early. Old friends, new peers, no lanyards yet. Then two full days: Day One on the answer layer, Day Two on what’s working, and a closing keynote worth flying in for.
And yes, there will be a thread of corny Canadian charm running through the whole thing. I’m from Nova Scotia. I can’t help it.
Speakers drop soon
The lineup is coming together now. We’ll announce speakers in the coming weeks, and I can’t wait to share who’s taking the stage.
What you leave with
Four things. A multichannel AI visibility playbook, because the data says you can’t win this one channel at a time. Benchmark numbers from frontier models you can take to a board that’s already asking what the AI plan is. A way to measure your own AI visibility, Share of Model and the frameworks underneath it, so the budget line has something to defend itself with. And a table of people to call when the next model update rearranges your category.
Running a conference is a bad business and a great excuse
I know that going in. The team at Foundation will be stretched in May, first years are messy, and I’ve sat through enough bad conferences to know I’m capable of building one too.
I’m doing it anyway, because the alternative is another year of the best people in this discipline working from separate spreadsheets. The brands showing up in AI answers in 2028 are the ones having the right conversations in 2027. I’d rather host those conversations than read the recap.
Tickets open October 26. Founding Seats go first.
The waitlist is open now, and the first 25 approved requests get Founding Seats at a reduced price. Every request is reviewed personally, so tell us what you own and what you’re trying to figure out.
If you’re the person I described at the top, the one with the Sunday-night spreadsheet, I built this for you. Come find the room.