Proof cohort — 10 clients per round
Become the proof that GEO works.
Six weeks for $500. Normally $2,000.
We run a controlled trial to measure how fast AI visibility moves when MentionLayer is run end-to-end on a real business — and we measure it against a matched control group, so the lift isn't our opinion, it's the data. Each round we take 10 businesses. You pay $500 — a 75% discount on the standard six-week engagement — in exchange for becoming a data point in the published results.
That's the deal. No subscription. No upsell. We do the work for six weeks, measure the lift against your control, and your business goes into the study.
Not ready to commit? Start with a free AI visibility audit — see exactly where you stand across ChatGPT, Perplexity, and Google AI Overviews before you decide.
Why we're doing this
Two studies measured. This one intervenes.
We've done this twice already.
Study 1 (Q1 2026) — 1,004 businesses, 10 industries, 5 AI models, 95,392 mention checks. Headline finding: 66% of businesses are completely invisible to AI. Findings, methodology, and data are public.
Study 2 (Q2 2026) — 2,729 businesses, 14 industries, 4 markets, 278,000+ mention checks. Tighter controls. We learned what actually predicts AI visibility: directory presence beats Domain Authority, and the "you need to be on Reddit" advice collapses to r = 0.000 once you control for everything else.
Both of those were observational. We measured. We didn't touch anything.
The controlled trial is different — and it's running right now. For the first time, we intervene. We take a cohort of businesses, run the full MentionLayer program for six weeks, and measure the before-and-after on every major AI surface against a matched control group that gets no intervention. That control is the whole point: it's how you separate "we did work and the number moved" from "we did work and proved the work caused the move."
The causal proof isn't finished — it's being built, one cohort at a time, and the businesses in these cohorts are how it gets produced. That's the jump from "what correlates" to "what causes," and you'd be inside the dataset that makes it.
We discount six weeks of work from $2,000 to $500 so the businesses in the trial are real — recruited at a price that filters in serious operators and filters out the curious. That's the only way the data we publish holds up.
What $500 buys you
Six weeks of real GEO work. Done for you.
This isn't a tool you have to figure out. We do the work. Below is the menu we work from — not a checklist we mindlessly run.
Week 1 — Baseline audit + control match
We measure where your brand is showing up across ChatGPT, Perplexity, and Google AI Overviews on every buying-intent query that matters, and we lock in a matched control group to measure you against. You see the starting line — and so does the data.
Week 1–2 — Entity foundation
Entity sync across Wikipedia, Wikidata, Google Knowledge Panel, and the AI-citation sources we've found move the needle. If you don't have a canonical entity, we build one.
Week 2–4 — Citation engine
Off-page work on the exact sources that LLMs are pulling from in your category. Real placements, not scraped junk.
Week 3–5 — Review & social proof signals
Review velocity campaign tuned to the platforms LLMs weight. Most businesses skip this. Most businesses don't get cited.
Ongoing — Weekly AI Monitor
Every week of the engagement, we scan your target queries across the major AI surfaces and track share-of-model movement. You get visibility into what's working in near real time.
Week 6 — Delta report vs. control
Side-by-side: where you ranked the day you started, where you rank now, and how that compares to the matched control group that got no intervention. Tracked queries, citation count, share-of-model delta. The data we publish in the study is the data we hand you.
Same work. 75% off.
The only difference between paying $500 here and $2,000 outside the trial is that you let us match you to a control and include the results.
Standard engagement
$2,000
Six weeks. Standard rate.
- Full six-week GEO program
- Baseline + week 6 delta report
- Weekly AI Monitor
- Private results
Cohort seat
$500
Six weeks. Same work. 10 seats per round.
- Full six-week GEO program
- Baseline + week 6 delta report
- Matched control group comparison
- Weekly AI Monitor
- Results published in the study (anonymized if you prefer)
After the six weeks
Want to keep going? Study alumni keep our best rate. For good.
By week six you've got real data — share-of-model lift, citation counts, the delta on every tracked query. Most businesses see momentum they don't want to give up.
Study participants get the option to continue at $1,000 per month — roughly a third off our standard done-for-you rate. No contract. Cancel any month. The rate is yours for as long as you stay.
We're telling you this up front so week six isn't a surprise pitch. It's a pre-framed option, not a hidden upsell. If continuing isn't useful, you leave with your delta report and we don't chase you.
This is for you if…
- You run a real business that wants to show up when buyers ask ChatGPT or Perplexity a buying-intent question.
- You'd rather we do the work than figure out GEO yourself.
- You're okay with the results being included in the published study (anonymized if you'd rather not be named), and with being matched to a control group.
- You can start now. Each cohort is capped, and the current one closes as soon as the seats are claimed.
This isn't for you if…
- You want a tool license. This is done-for-you work, not software access.
- You can't share even anonymized before-and-after data. The whole point is the controlled trial.
- Your business is brand new with no website. We need a baseline to measure against.
Claim your seat
$500. Six weeks. We do the work.
Fill this out. You'll go straight to a secure checkout. Once payment clears, we email you within 24 hours to schedule the kickoff call and lock in your control match.
Want to see your starting point first? Run a free AI visibility audit — it's the no-risk first step, and there's no obligation to claim a seat afterward.
The guarantee
If your AI visibility doesn't move on a tracked query in six weeks, you pay nothing. And "it moved" isn't our opinion — it's the data.
Here's the part no other GEO provider can credibly offer: because we measure your results against a matched control group — comparable businesses we don't touch — the lift on your delta report is the difference your intervention made, not noise that would have happened anyway. The control is the referee.
So the guarantee has teeth: six weeks in, we send the delta report. If your tracked queries haven't moved relative to your control, you get your $500 back. No paperwork. No "let me talk to my manager." We don't think it'll come to that — but you shouldn't have to take that on faith, and with a control group, you don't have to take our word for it either.
Honest answers to the obvious questions
Why so cheap?
Because the trial only works if the businesses in it weren't hand-picked by us. The $500 price is what makes each cohort self-selecting and what makes the published data credible. It's the standard rate ($2,000) minus the value of what you give us: permission to match you to a control group and publish the result.
What's the control group, and why does it matter?
When we start your six weeks, we also identify a set of comparable businesses we leave completely alone — same kind of starting position, no intervention. At week six we compare your movement to theirs. That's what lets us say the lift came from the work and not from something that would have happened anyway. It's also why our guarantee is stronger than anyone else's: 'it moved' is measured against a control, not asserted by us.
Is this a one-time thing, or ongoing?
It's ongoing. We run the controlled trial in cohorts, and we take a fixed number of proof clients each round so every business can be matched to a control and run properly. When the current cohort fills, that round closes and the next one opens. The offer itself stands — the scarcity is real seats per round, not an artificial countdown.
What happens to my data?
Your before-and-after numbers — and how they compare to your control group — go into the published study. You choose whether your business is named or anonymized. If you pick anonymized, even we won't put your name in print.
Is this a subscription? What happens after week six?
The $500 is one payment for six weeks of work. No auto-renewal. After the delta report lands in week six, cohort participants get the option to continue at $1,000/month — roughly a third off our standard $1,500/month done-for-you rate. The alumni rate is yours for as long as you stay, with no contract and no minimum term. If you'd rather walk away with your data and stop there, that's fine too. We're telling you up front so it's not a surprise pitch.
Who's actually doing the work?
Joel House and the MentionLayer team. Joel founded Joel House Search Media and has been doing this kind of work for over a decade. The MentionLayer platform runs the daily ops; humans do the strategy and the placements.
What if my business is in a tiny niche?
We'll still take the seat if we think we can measurably move you in six weeks and find you a fair control match. Email [email protected] before paying if you're unsure — we'll tell you straight.
What if I want in but $500 is too much right now?
We're not going lower than $500 — at that price the program is already running at a loss. If it's a budget issue, the right answer is probably to wait for a later cohort and pay $2,000 when it makes sense. In the meantime, run the free AI visibility audit so you at least know where you stand.
One more time
$500. Six weeks of GEO. 10 seats this cohort.
When this cohort fills, it closes and the next one opens. If you want to be in the round that's running now — and in the dataset it produces — now is the moment.
Claim a cohort seat — $500Or run a free AI visibility audit first — see where you stand, then decide.
P.S. — Study 1 (1,004 businesses) hit Forbes, became a working paper, and is the reason MentionLayer exists. Study 2 (2,729 businesses, 4 markets) proved that some of the most-cited GEO advice doesn't survive proper controls. The controlled trial you'd be joining is the first time we intervene rather than observe — the first to measure GEO against a matched control group. It's being built one cohort at a time, and the businesses in it are how the causal proof gets made. Being inside the dataset for $500 is the cheapest credible AI visibility move you'll see all year.