# Ryan Okafor, Independent Builder / Former PM at Lattice — read of ModelRecon, June 23 2026

> "9 years in B2B SaaS product and growth, left my last job 8 months ago, now consulting 3 days a week and building micro-SaaS on the other two."

## How I got here

Saw someone in the Indie Hackers Slack drop a link with the message "AI SEO is the next SEO, and nobody has a tracker yet." That framing hooked me because I've been thinking about the same thing. I was expecting a live product. I Google-searched "ModelRecon" separately to see if it had any other presence before clicking the actual link, found nothing, came back and read the page properly.

## What I clicked first

The hero concept landed. "What Every AI Model Recommends (And Why It Matters)" is actually a clean brief for the idea. The HubSpot stat pulled me in too: "HubSpot was 46% last week, jumped to 54% this week." That's a real number in a real category. That's the kind of thing a VP of Marketing at a CRM competitor would forward to their growth team on a Monday. I wanted to see more of those.

## Where I paused

The "Try it Live result" call-to-action near the top. I clicked expecting a working demo or at least a screenshot of the actual dashboard. What I got was... I honestly don't know what I got. The page doesn't explain what "live result" means. If there's a working prototype behind that button, show me one frame of it. If there isn't, don't call it live.

## What I distrusted

A few things, but the big one: about two-thirds down the page the product flips identity on me. I came here thinking ModelRecon was a thing I could subscribe to and use for my clients. Then I hit "we don't have live customers on this idea yet. We shipped the strategy package; you ship the customer conversations." So I'm not buying a product. I'm buying a blueprint to build the product. That's a completely different offer, and the page doesn't make that distinction early enough. The hero copy reads like a SaaS product. The pricing section reveals it's an idea marketplace. That's a jarring left turn.

Also: "$-17,776 Year-1 take-home (Fermi)" is buried as a parenthetical in what's supposed to be a confidence-building section. Negative year one is fine, lots of real businesses have that. But leading with "64/100 Adoptability" and then quietly showing a negative income number in the same card without explanation feels like the honest disclosure is technically present but not actually legible.

And "financial upside: 1/10" as a named concern, directly on the page, for a product that costs up to $199 to adopt. That's a bold thing to show. I respect that it's there. I'm also not sure it helps the conversion.

## What would convince me

One company, named, that ran ModelRecon data and made a specific decision based on it. Not a testimonial. A mini case study: "Acme used the ranking shift data from week 7 to revise their category positioning page. Their Claude rank went from 4th to 2nd." That's the kind of evidence that makes the $99 feel reasonable. Right now the strongest evidence on the page is the HubSpot percentage stat, which I suspect the team generated themselves to show the format.

I'd also want to see a real screenshot of the Friday email digest. Even one. If the output is genuinely sharp, show it. The copy says "strategic intelligence delivered concisely" but I have no idea what that looks like in practice.

## What I'd ask in an email reply

1. The "20+ B2B categories" - are these all live and tracked today, or is that the roadmap when someone buys the $99 package and builds it? Because that distinction matters a lot to my build-vs-buy decision.

2. The Fermi model shows negative year-one income. What's the assumed pricing and customer count behind that number? I want to see the spreadsheet, not the output cell.

3. "Operate with us, custom" - what does that actually mean? Is this the Wishdeal team running the product as a white-label for a client? Is it a rev-share? What's the typical custom scope?

## Verdict: on-the-fence

The core idea is real and I haven't seen it executed anywhere yet. But the page is selling two different things at the same time and doesn't commit to either one. If I can't tell whether I'm buying a product or buying the right to build a product, I'm not clicking pay.

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