Market
Market critique: Caverio first revenue wedge
MARKET dimension only, 2026-09-25. Competitor prices are unverified: web search was quota-blocked this session.
1. The chairman problem kills PLAN-v1 as written
PLAN-v1 §3A rests the wedge on "access = Hayden's own launch + the 139 contacts"; §4 adds "Pilot #1: Hayden's launch. Ask for a paid pilot; fall back to in-kind for a named case study."
Hayden is listed in §1 as "chairman/investor". A cheque from your own investor is not revenue, it is a cap-table transfer. It proves nothing about willingness to pay, cannot be cited without disclosure, and the in-kind fallback means the flagship pilot most likely returns $0 plus a case study about a token that will be worthless by the time you show it. "First paid invoice by 25 Oct" (§6) can be hit with related-party money and teach the business nothing.
2. Launch teams want pumps, and the mitigation does not hold
PLAN-v1 §7 names the risk itself: "Launch teams pay for pumps, not for honesty. Mitigation: sell to the MM and the agency, who must prove performance to the project."
That inverts the incentive. The agency's job is a flattering post-mortem, and you propose selling it an instrument that can prove its KOL spend produced no money: §4 promises exactly that, "what attention produced money". No agency buys its own audit at $15,000. The MM is worse: it already holds fill, inventory and flow data you cannot match, and it buys execution, not dashboards. The real buyer of an honest report is the treasury or its investors, who are the slowest to pay.
3. Price without references
$7,500 pilot / $15,000 standard (§4) is 5-15% of the stated $100-300K budget, asked by a 14-day-old brand with, per §5, no entity, no Stripe, no cleared trademark, no live domain. Nothing explains why a cold name off a 139-contact sheet signs a five-figure SOW with an unincorporated vendor in three weeks ("pilot signed by 3 Oct"). Cold five-figure B2B into crypto agencies runs 45-90 days, so the 25 Oct and 30 Nov ($30-45K) targets are fiction at any normal conversion rate.
4. The $69 tier: the competitor is free, the record is negative
Synthesis §2 justifies the price as "proven by neighbours (Nansen $49/$69, LunarCrush $90/$300/$900, stated-unverified)". Anchoring on a competitor's price is not demand evidence, and it is the wrong comparison set. The memecoin trader's daily tools are GMGN, Photon, Axiom, DexScreener and Cielo's free tier (unverified), all shipping sniper/bundle/first-buyer views at zero cost; Bubblemaps and Arkham do cluster work free. You are not priced against Nansen, you are priced against $0.
Worse, it is bought as a signal service whatever the copy says. Synthesis §6 claims "members stay because the record is what they pay for". No one renews $69/mo to admire a losing ledger. And the ledger is losing: PLAN-v1 §1 says "34 entries, B book gross +$152 / net -$218 today", while synthesis §1 says "v2 at 21 entries, $4,677 net". The two co-founder drafts quote different books. A prospect opens the record first; if the founders disagree on it, no sale survives call two.
The one differentiated asset is PLAN-v1 §2.2, recurring early wallets (0.10 vs 0.27): a single unaudited statistic over 2,670 tokens that a buyer cannot tell apart from the free sniper badge GMGN already ships.
Verdict: ranked
- Launch Desk first, descoped and repriced. A dated event with a budget holder beats retail with no edge and no channel. One $6K launch is ninety founder-months.
- $69 Founding Member: parallel free waitlist only, $39 when it launches. A zero-cost demand test, not the wedge.
- Data/API lane: later, unchanged.
Price: scan $2,500, launch room $6,000, retainer $1,500/mo. Founding tier $39 over a free Now page.
Three changes forced before this goes to the founder
- Define first revenue as the first dollar from a party with no equity in Caverio. Hayden's launch becomes the free reference with a signed right to publish, not the pilot invoice.
- Sell a report before a system. Replace the $15K launch room as the opening offer with a $2,500 72-hour pre-launch scan: zero new engineering, ships in 48 hours. Five scans sold to strangers beats one chairman pilot.
- Freeze one public record, one basis, one start date, before anyone sells. If it is net negative, print that. And reframe the launch pitch from "honest read" to defensive counter-intelligence: who is farming your launch, and did your paid KOLs convert.
Go-to-market
Red team: GO-TO-MARKET
Caverio, 2026-09-25. Channel, motion, founder time, sequencing, conversion math.
1. The 20-DM math does not survive contact
Synthesis line 88 sets "20 DMs, replies logged, 10 paid"; line 36 makes ten paying the day-14 test. That is a 50% close rate on cold outreach from a brand that line 43 concedes has "Paying customers: zero." Honest bands: 25-35% reply if warm, 10-15% if not, 5-10% of repliers pay that week. Twenty DMs yields five replies and one member. Ten paying needs 150-250 conversations: three to four founder-weeks.
Nor is the list warm. Line 108 calls "the 3,488-handle network and 109 tracked traders" a warm list "we already hold with profiles and PnL." We hold their data, not their permission: a cold list with a dossier. And "one X thread a day" pushes content into an account with no audience. Borrow distribution: three group admins who repost.
2. Elie cannot sell $15K without an artifact
PLAN line 33 prices "pilot $7,500; standard $15,000 per launch"; line 37 sends Elie at "the sheet's top-10 first-moves list." That sheet was built for Hayden (line 10), so Elie cold-mails a third party's list with no case study, no reference, no entity, no Stripe (line 51 has those to-do). A $15K agency or MM engagement runs three to eight weeks, and the budget holder is the project. 139 cold contacts at a 5% meeting rate is seven meetings; at 15% close, one discounted deal, late November. Line 35's "first paid invoice by 25 Oct" is fiction unless it is Hayden's.
3. Hayden is a favour counted as a channel
PLAN line 34: "Ask for a paid pilot; fall back to in-kind for a named case study." A related-party purchase validates nothing about price, and every agency discounts it on sight. Charging him buys $7,500 and no pipeline. Run it free, for two written things: permission to publish the numbers, and five named intros inside his launch stack.
4. What a stranger needs before sending USDC
Synthesis line 87 wants "promise, three real cards from the record, price, pay button." Missing: a named human with a face, a live timestamp proving the tape writes, the record above the price, a reversible entry. USDC is itself the conversion tax: no chargeback, anonymous counterparty, $69. Issue a free 7-day passcode, then charge. Drop "50 seats" (line 66): scarcity at zero customers reads as desperation.
5. Running both splits a one-founder team
PLAN line 36 spends "~5 Bolo build days" on the launch room; synthesis line 91 limits Bolo to "the record page and gate bugs." PLAN line 40 starts founding beta 15 Oct; the synthesis wants 10 paid by 9 Oct. Two clocks, one builder, one selling founder: parallel, neither reaches proof.
Verdict: the sequence I would run
W1 (26 Sep-2 Oct). Record page public and reconciling (Bolo). Free-access issuance on the 3047 gate, terms and refund page (Vesper). DNS, entity, Hayden reframed as barter (Thomas). No launch-room build. Target 25 conversations, 10 trials.
W2 (3-9 Oct). Landing page with a face, record above the price, daily DM material (Vesper). 15 DMs a day (Thomas). No outbound from Elie; he cuts the 139 to a top-20 and drafts the one-pager. Target 75 conversations, 30 trials, 3 paying.
W3 (10-16 Oct). Hayden's launch run live and captured as the case artifact (Bolo, Vesper). Trial-to-paid follow-up (Thomas). Elie opens the top-20 on discovery calls, not $15K. Target 8 paying, 5 agency calls.
W4 (17-23 Oct). Publish the Hayden case with real numbers (Vesper). Elie converts one $7,500 pilot on a dated launch. Thomas asks members what they would pay double for. Gate: 12 paying, or price drops to $39. No rebuild.
Three changes I force before this reaches Thomas
- Rewrite the conversion math. The day-14 gate becomes 3 paying plus 25 logged replies out of 75 conversations, and the funnel gains a free 7-day tier and a card rail.
- Hayden's launch is unpaid and contractual: publishable numbers plus five named intros, in writing before the work starts. Do not invoice the chairman for your only case study.
- One clock, one builder. Launch Desk gets no build days and Elie no outbound until the record page is live and one stranger who is not Hayden has paid.
Ops, legal, build
Critic: OPS / LEGAL / BUILD (2026-09-25)
Web search down (quota exhausted, fourth credit exhaustion this month). All rights claims UNVERIFIED.
1. The tape cannot carry a paid SLA
Synthesis line 82 promises "every green window on Now within five minutes of the crossing". DATA-AUDIT: that stream runs p90 30 min, 11% of gaps over 10 min (ticks 17%), 33% exact repeats. You cannot alert on a crossing you never observed. Solana hung 8 days, then again today, and the watchdog was installed today, so seven clean days measures a one-day-old watchdog's luck. Helius: 13,513/20,000 lifetime calls, burn unknown; exhausted, Solana drops to publicnode 4 rps and 429s. X dark since 09-18 already breaks Launch Desk item (ii) "caller mentions".
"7 clean days" must measure: ingest gap never over 15 min; observation-gap p99 and share over 10 min (target 2%, today 11-17%); crossing-to-alert p95 from providerAt, not detect time; missed crossings by replay against an independent refetch; watchdog fires (three restarts is not a clean day); repeats on in-range tokens only; zero print-gate bypasses; days-to-exhaustion on Helius and CoinGecko credits. Restart on incident, and make it 14 days.
2. Redistribution (UNVERIFIED, blocking)
DATA-AUDIT finding 2: "every provenance-bearing price row since 09-23 is dexscreener." Line 124's "a tape they cannot get elsewhere" is false for the displayed prices: rights risk plus misrepresentation risk.
- DexScreener: free API, terms generally bar commercial redistribution and competing products. Most exposed, no permission held.
- CoinGecko Basic: commercial use generally allowed with attribution (absent on Now); bulk resale reserved to higher tiers.
- Fomo Starter: no redistribution grant known; caller/thesis content is proprietary aggregation, not public chain data, so it is the riskiest reuse.
- Helius: derived chain tape defensible (blocks are public); parsed output and reselling the service is not.
- Binance: market-data redistribution normally needs a separate licence.
Sell derived state only. Trap: the raw sidecar (bead 44ht) provenance requires is what must never leave the building.
3. USDC with no entity
Line 116 leaves "which entity, if any, invoices?" open; PLAN-v1 §5 gives it one line.
- Tax: crypto for services is ordinary income at receipt to a person, plus gain on disposal. No entity, no limited liability, no invoice, so PLAN-v1's $15K cannot be billed.
- Refunds: the 7-day refund (line 86) is manual USDC to an unverifiable address, no dispute path, promised by no legal person, beside "price locked for life".
- AML: merchant not transmitter, so MSB registration likely untriggered, but sanctions screening attaches to the individual and USDC is Circle-freezable. Coinbase Commerce and NOWPayments demand KYB, so the day 1-4 payment task fails into a bare wallet with no receipts.
- Custody: unnamed in both drafts. Demand dedicated address, named signer, written recovery, no commingling with the live Robinhood-chain book.
4. Advice and manipulation adjacency
"max safe size" (line 68) is per-token position sizing, a recommendation as to amount; replace with observed depth, never "safe". Green/red plus a push at entry is functionally a buy signal. The strategy vote and private group (line 72) erode the publisher posture.
Launch Desk is worse. PLAN-v1 §3 sells a market maker, for $15K, live identification of who is buying and "when the MM can step back" during a launch it is paid to support: a tool for timing support withdrawal. Add the 20% referral to the MMs and agencies supplying launches and a first customer who is Hayden, chairman/owner, launching his own token (§1). Related-party revenue plus undisclosed kickbacks plus a "public honest record" is self-cancelling.
5. Five build days is fiction
Line 38 records five rebuilds in seventeen days. PLAN-v1 §3 budgets "~5 Bolo build days" for a live 48-hour room fed by chain tape, holdings, social sweep, case builder. DATA-AUDIT: holdings is Robinhood-only with 46% "chain has no tape" stubs, BSC holdings is Bolo item 7 behind items 1-6, social sweep 25% X-dark, Solana on near-empty credits. Three of four inputs do not exist, and shipping collides with PLAN-v1 §1's provenance-first rule. Realistic: 12-20 days. A 48-hour live room is a hard availability promise on the customer's least forgiving day with no SLA, no credit clause, no rota, no named human awake. The synthesis is not clean either: line 91 gives Bolo "only the record page and gate bugs" while its table assigns the payment hook and alerts channel the same week.
Wedge verdict: Founding Member is the right first wedge (lower build, one counterparty class, no availability contract, no related party). Launch Desk must not be first.
6. Blocking before the first dollar
- A legal person: entity or sole trader, tax residence, signer, liability.
- Named wallet custody, no commingling with the trading book.
- Sanctions screening at payment, refusal clause, payer records retained.
- Terms, risk, privacy and refund mechanics live and human-reviewed before the pay link is public.
- Rights file: five sources quoted and dated, field-to-permission table, unpermitted fields cut.
- "max safe size" removed and green/red settled before 50 people pay.
- Conflict disclosure on the record page: live book, Hayden, referral fees.
- Reliability gate passed on the §1 definition, 14 days, with a written service credit.
- Caverion knock-out in writing (PLAN-v1 §5).
7. Non-blocking
caverio.world and the DNS token, landing polish, Stripe, Pro $99, annual pricing, X account, referrals, $69 vs $79, Launch Desk in any form, provenance items 5-9, price-paths compaction, brand assets, the 50-seat cap.
8. Three changes I would force
- Kill "max safe size" and the green/red verdict as sold fields. No disclaimer repairs a green badge wired to a push notification.
- Redefine the gate: 14 days on observation gap and missed-crossing replay, not uptime, with credit runway as a gate item. At ~6,500 Helius calls left the Solana half is unfunded and must not be sold.
- No dollar before a named legal person and a named wallet custodian exist in writing, and no Hayden launch as first revenue without a written related-party disclosure. If the entity takes two weeks, take names, not money.