# Marcus Villanueva, Head of Product Marketing at Pageloom (47 people, B2B project management SaaS) — read of Weekly AI Benchmark, June 23 2026

> 9 years in SaaS marketing, currently obsessing over why ChatGPT keeps recommending Notion and Asana to our prospects and never us. Commutes 40 minutes each way on the 101, two kids (8 and 5), coaches youth soccer Saturday mornings. Ahrefs, G2, Semrush, HubSpot are my daily stack.

## How I got here

A prospect last week told me he'd typed "what's the best project management tool for a 50-person startup" into ChatGPT before our call. We weren't in the answer. He bought a competitor. That landed differently than usual. So I started Googling "track AI recommendations SaaS" and "GPT brand visibility monitor" and this came up around result 7 or 8. I clicked because the slug matched what I was searching for almost word for word.

## What I clicked first

The hero grabbed me. "Know where your SaaS ranks with AI" is exactly the question I walked in with. And the "Try it Live result" thing piqued me immediately. I clicked it before I read anything else. Whatever that demo shows is doing more work than the entire copy block below it.

"See exactly where AI models rank you first, second, third against every competitor. Updated weekly." Fine, clear, I'm nodding. "Know when rivals move up or down in AI recommendations before they announce anything." That line hit differently. That's the actual pain. Not just "where am I" but "when did they leapfrog me and why."

## Where I paused

The pricing section stopped me cold. "Browse Free. Unlock for $5. Adopt for $99." I read that three times. Adopt? Unlock a dossier? "ICP, MVP scope, first 7 build tasks, 30/60/90 launch plan, GTM, email drip." Wait.

Then I scrolled back up and re-read "Built by Wishdeal Studio" and the sidebar listing "MerchantQA 78" and "RemarkAI 71" with "Yr1 $$-5K (est)" tags.

This is not a subscription tool I can sign up for. This is a business idea for sale. The entire page is pitching me on BUILDING this thing, not using it. I felt genuinely disoriented for a minute. I came in as a buyer for the service and found out I'm being recruited to be the founder.

## What I distrusted

"Estimates only · no live customer revenue claimed." That disclaimer is tucked into the footer in small gray text. The above-the-fold experience reads like a live SaaS product. The "Before / With Weekly AI Benchmark" framing, the feature bullets, "Real-Time Rankings" (but then "Updated weekly" — which is it?) — none of that signals "idea stage." It feels like false-front marketing where you have to read the fine print to understand what's actually for sale.

Also: "Help the right operator find this. We don't get inbound any other way." I actually respect the honesty of that line. But it's strange to include it on a page pretending to be a product page. It's the studio talking, not the product talking. The frame keeps slipping.

The $5 "unlock dossier" tier is doing something weird too. It implies the real content is hidden, which makes me distrust the free content I'm reading as a teaser rather than substance.

## What would convince me

If I'm reading this as a potential builder: show me one real example output. Not a screenshot mockup. An actual prompt run on Claude or ChatGPT, category chosen, real competitors, real rankings, showing what the weekly diff would look like. If the methodology is sound, that one example would sell the idea better than every bullet point on the page.

If I'm reading this as a potential end-user buyer (which I now know I can't be, but I came in that way): show me a live trial where I type in my SaaS name, pick a category, and see my actual ranking in two or three models right now. That would convert me in 90 seconds.

## What I'd ask in an email reply

1. The methodology note says "Updated weekly" but how are you querying the models? Are these fresh prompt runs each week, or are you caching results? Models get updated mid-week and context windows drift. How do you control for that?

2. The idea assumes AI models rank products in a stable, consistent way. In my testing, ChatGPT gives wildly different answers depending on how the question is phrased. How does your scoring account for prompt variance, or is the benchmark based on one canonical prompt per category?

3. If I adopt this for $99, what's the actual first step? "Working code starter" for what stack, hosted where, and how much does the ongoing querying cost to run each week in API fees?

## Verdict: on-the-fence

The core pain is real and I felt it in my chest when I read "Know when rivals move up or down before they announce anything." But I came in looking for a tool to buy and found a business to build. That's a bait-and-switch even if it's technically disclosed. The idea is good enough that I might actually come back. The page needs to pick a lane: product page or idea listing. Right now it's trying to be both and it's fully neither.

---
*Memo by skeptic persona, generated 2026-06-23. Studio breaks own self-grading loop.*