INTERNAL, DO NOT SHARE PortLev assessment: sources, rationale, STORM
Companion to /assessments/portlev. Built with the /assessment-report skill, July 21, 2026.
Sources and observations
https://portlev.com/ Positioning: "Close Your AI Build Gap. Built. Not Advised." Mid-market CEOs/boards + individual execs (Portfolio Engineering). Palette used for the report: white bg, near-black ink, indigo #575BE8 accent. Rendered footer links exactly ONE social account: linkedin.com/in/yurikruman. JS check: zero analytics, LinkedIn, or Meta pixels.
https://portlev.com/ai-org-audit $997 flat audit, Stripe checkout, refundable if <$10K identified, credit toward build within 90 days. Delivery: intake call day 1, Yuri maps workflows days 2-4, savings model days 5-6, 60-min readout day 7, 8-part Leverage Map. Free 15-min pre-purchase consult via Calendly. Grounds Audit Factory.
https://portlev.com/scorecard FUNNEL WALK 7/21: took the AI Pilot-to-P&L Scorecard (6 failure modes, 0-3 each). Scored 11/18 -> instant "Pilot Purgatory" verdict + priority failure mode BEFORE email ask (good design). Then submitted contentengine@prestonchin.co for the promised "full breakdown + one-pager. No spam."
https://leveragebrief.beehiiv.com/ FUNNEL BREAK, OBSERVED 7/21: the scorecard email arrived as "Confirm your The Leverage Brief subscription" from leveragebrief@mail.beehiiv.com (double opt-in). The confirm link landed on the generic beehiiv subscribe page asking to subscribe again. The promised breakdown and one-pager never arrived. Grounds Scorecard Rescue.
https://portlev.com/labs Platform/labs pages list PortLev-built AI products; grounds the peer-level "you could build all of this; the offer is capacity" line.
https://masterthetalk.com DOMAIN CHECK 7/21: 301-redirects to masterthetalk.squarespace.com which returns 404. This is the domain Yuri replied from (yuri@masterthetalk.com). Stated plainly on the client page.
https://x.com/masterthetalk Old-brand X handle exists (search-verified). Follower count not pulled; listed without a number on the client page.
Why each automation made the cut
01 Audit FactoryTheir audit page publishes the manual production schedule (days 2-6 are Yuri). The audit is the core paid product and the entry to builds/retainers, so capacity here is revenue, not just cost. Confidence: high; the schedule is their own copy.
02 Scorecard Rescue and NurtureObserved firsthand: promised deliverable never arrives; double opt-in breaks the chain; no pixels means the drop is invisible to them. Highest-credibility finding in the report, and the fix directly fills the audit pipeline. Confidence: high on the observation; the "no slower sequence exists" caveat is flagged in assumptions.
03 Content Engine31K LinkedIn followers vs zero bridge to the funnel, 42 podcast appearances used once, weekly newsletter produced by hand. Preston proof links included per skill. Confidence: high.
04 Hot-Lead Speed AgentThe scorecard scores every visitor but treats a 15/18 the same as a 3/18 (observed: same email either way). Four-figure offer -> the speed-to-lead logic from the skill applies. Confidence: medium-high.
05 Cohort Ops EngineThe 12-week, 15-seat Portfolio Executive Cohort exists (site copy). Solo operator + cohort admin is a classic hours sink. Confidence: medium; cohort operations are not publicly visible, flagged in assumptions.
Rejected: book preorder machineThe /preorder page exists but the book funnel is early and less connected to revenue this quarter than the five chosen. Kept for the call.
Rejected: enterprise/PE reporting automationsThe enterprise page is thin on delivery promises compared to the MomentumOS ROI calculator; grounding was too weak to claim manual reporting exists.
STORM pass (condensed, five lenses)
PractitionerA solo expert consultant reads teardown emails skeptically but responds to firsthand walk-throughs. The broken-scorecard finding is the whole door-opener; it must stay first-person and unadorned. Passes.
AnalystStrongest evidence base of the two reports: four firsthand observations (scorecard flow, opt-in break, zero pixels, dead domain) + their own published audit schedule. Weakest claim: "breakdown never arrives" is a single-run observation; the assumptions box carries that caveat. 31,364 is a dated snippet; labeled approximate.
SkepticYuri trains AI models for OpenAI/Meta/Microsoft per his own site; a lesser report would insult him. The capacity-not-ideas line and peer tone defuse it. The double opt-in might be a deliberate compliance choice; even so, the promised deliverable failing to arrive is a defect regardless of intent.
EconomistOrder is right for a solo operator: audit capacity (revenue per week of Yuri) leads, funnel rescue second, content third. All five avoid adding headcount, which matches a solo P&L.
HistorianConsultant-to-consultant free value has the highest bar of any archetype; specifics and humility win. The "no notes" line about his offer structure buys credibility the criticism spends. Keep it.