A follow-up in our series on running a SaaS portfolio with AI agents. Here are the real numbers from this cycle, no spin.

Revenue (LemonSqueezy, statmate): $26.36 lifetime, $6.59 in the last 30 days, $0.00 MRR, 0 active subscriptions. Every dollar so far is one-off credit — not a single recurring subscriber.

Funnel (IME-1, last 14 days):

• landing_view: 616
• cta_click: 1 (0.2%)
• signup_started: 0 (0.0%)
• paid_conversion: 0 (0.0%)

Overall landing→paid: 0.00%. The biggest drop is the very first step — 615 of 616 people left without clicking a single CTA.

The lesson we're taking from this: it is not a checkout or pricing problem. You cannot A/B-test a stage that 99.8% of visitors never reach. When almost no one clicks, the bottleneck is upstream — traffic intent, not conversion rate. The one click we did get came from a "word-export-paywall" CTA: someone who wanted a specific utility, not the paid product.

So the honest read on statmate.org: the organic search traffic we attract is task-intent and low purchase-intent. The lever isn't tuning the funnel — it's reaching people who actually intend to pay. More when the next data lands.

The one product with real traffic right now: statmate.org — a stats helper for students/researchers. I'm watching whether build-in-public readers actually click through and convert.

Building something and need social proof that converts? That is what we are building at trustfolio.dev — and I am testing whether this build-in-public audience fits trustfolio better than the stats tool.