Commitments and consequences
Companion document to the main memo. This document tabulates the consequences of each architecture along three axes:
- UX impact — what each party (streamer, operator, Verifluence, TG3, external deps) has to do differently
- Cost impact — per-campaign dollar costs, who pays, and where they accumulate
- Build cost and cost of ownership — Verifluence's CFO view: one-time CapEx + ongoing TCO over the 15-year SHA term
Operational risk and recovery scenarios close out the document.
The redline assumptions, the high-level option summaries, and the question to TG3 live in the introduction. Engineering details on each architecture live in Options — technical details + diagrams.
1. UX impact
What each party has to do differently under each architecture.
| Today (Base) | Option A — Parallel | Option B — Metadata | Option C — Facade | |
|---|---|---|---|---|
| STREAMER | ||||
| Where USDC arrives | Base wallet | Base or EtherLink (depends on operator's election) | Base wallet | Base wallet |
| Signs transaction? | No | No (gasless on both paths) | No | No |
| Wait time for payout | ~10 s confirmation | ~10 s (Base) / 5–30 min (EtherLink, via LayerZero bridge) | ~10 s, unchanged | 5–30 min (cross-chain msg latency) |
| Manual off-bridge needed? | No | No (Base path) / Yes, every payout (EtherLink path) | No | No |
| Failure visibility | Clean — never fails | Transfer Cap → bridged USDC stuck on EtherLink (EL path) | None new | LZ outage → delayed arrival, can be hours |
| OPERATOR | ||||
| Funding asset | Native Circle USDC or USDT | Native USDC/USDT (Base) / bridged variant required (EtherLink) | Native USDC/USDT, unchanged | Native USDC/USDT, unchanged |
| Chains touched | Base only | Base or Base + EtherLink | Base only | Base only (EtherLink invisible) |
| New wallet/chain setup | None | None (Base) / add chain, acquire XTZ (EtherLink) | None | None |
| Refund flow | One Base tx | One Base tx / one EtherLink tx | One Base tx | One Base tx (EL coordination invisible) |
| VERIFLUENCE | ||||
| Engineering effort | — | Contract port + chain-aware addressing | Small registry contract + backend writer | Two coordinated contracts + cross-chain protocol integration |
| Calendar to ship | — | 4–6 weeks | 2–3 weeks | 8–12 weeks |
| Code added (Solidity) | — | ~50 lines (chain-aware deploy scripts) | ~80 lines (registry only) | ~650 lines + ~1,500 lines of tests |
| Ongoing ops surface | Single chain | Two deployments, two indexers, EL runbooks | Single chain + thin EL writer | Two chains + cross-chain reconciliation + LZ monitoring |
| Recovery runbooks needed | Standard | + Transfer-Cap / stuck-claim | + Backend retry on registry-write failure | + Stuck-message, replay, reorg, LZ degradation |
| Reversibility | N/A | Easy (disable EL path) | Easy (stop writing) | Hard — sunk audit + on-chain state |
| TG3 | ||||
| What they get | — | Real EtherLink HTLC serving real campaigns (volume ∝ EL election share) | On-chain record of every Verifluence deal on EtherLink (audit-grade) | Escrow logic backbone runs on EtherLink (full architectural commitment) |
| EXTERNAL DEPS | ||||
| New protocol dependency | None | LayerZero bridge | None | LayerZero general messaging |
| Liquidity dependency | None | LZ USDC pool on Base (for streamer off-bridging) | None | None at user level |
| Failure surface inherited | None | Bridge incidents (Transfer Cap, LZ degradation) | None | Full cross-chain messaging surface — historically the highest-incident category in DeFi |
2. Cost impact
All numbers per single campaign as defined in Assumptions of the introduction: 20 slots → 19 successful payouts + 1 refund; streamer signs nothing and is paid immediately on approval; operator funds in USDC.
- Verifluence absorbs → platform OpEx, billed against Verifluence's P&L
- Operator absorbs → paid from operator's wallet during funding
- Streamer absorbs → deducted from streamer's payout
| Today (Base) | Option A — Parallel | Option B — Metadata | Option C — Facade | |
|---|---|---|---|---|
| STREAMER | ||||
| Out-of-pocket / payout | $0 | $0 (Base path) / $1.50–2 per off-bridge (EL path) | $0 | $0 |
| Cumulative friction / campaign | $0 | $0–38 (19 off-bridges if EL-elected) | $0 | $0 |
| OPERATOR | ||||
| Bridge cost / campaign | $0 | $0 (Base) / $1–2 one-time bridge-in (EL, per Technical Assumption #2) | $0 | $0 |
| Gas + XTZ acquisition / campaign | no impact* | no impact* (Base) / $3–5 (EL: bridge + XTZ acquisition) | no impact* | no impact* |
| Cost / campaign (worst case) | no impact* | no impact*–$5 | no impact* | no impact* |
| Cumulative cost over 100 campaigns | no impact* | up to ~$500 | no impact* | no impact* |
| * per Technical Assumption #1: per-tx fees ≤ $0.20 USD reported as "no impact." Verifluence relays cover gas, so operators see only the bridge + XTZ acquisition friction. | ||||
| VERIFLUENCE | ||||
| Build cost (one-time) | — | $8–12K audit + ~$20K eng + $5K integration ≈ $35K | $4K audit + ~$10K eng + $3K integration ≈ $17K | $48–100K audit (often two firms) + ~$40–50K eng + $10–15K integration ≈ $130K |
| OpEx / campaign | no impact* | no impact* (Base) / $28.50 (EL path: 19 claim+bridge relays @ $1.50/tx) | $0.27 | $6.60 (22 cross-chain msgs × $0.30, per Tech Assumption #3) |
| Annual OpEx @ 100 campaigns (2K deliveries, year 1) | no impact* | no impact*–$2.85K | $27 | $660 |
| OpEx over 15-yr SHA term @ 100 campaigns/year flat | nil | up to $43K (EL-heavy) | $405 | $10K |
| Hidden ops cost (on-call, incidents) | — | Moderate | Negligible | High |
| TG3 | ||||
| Direct cost to TG3 | — | $0 | $0 | $0 |
| TG3 captures none of the absorbed cost above — it flows to LayerZero / Hyperlane / Verifluence team time | ||||
| EXTERNAL | ||||
| Per-protocol fees absorbed by ecosystem / campaign | $0 | ~$28.50 on EL-elected campaigns (LayerZero) | ~$0 | ~$6.60 (Hyperlane) |
| 15-yr cumulative fee transfer to protocol @ 100 campaigns/year flat | $0 | up to $43K (LayerZero) | ~$0 | $10K (Hyperlane) |
| For perspective | (~20% of TG3's €200K investment) | (~5% on variable fees alone — but engineering + audit-refresh push total much higher; see Build + TCO table below) |
3. Build cost and cost of ownership (Verifluence CFO view)
Table 2 captures per-campaign transaction fees. This table captures everything else — engineering build cost, ongoing engineering overhead, monitoring infrastructure, incident response, and audit refreshes — and rolls it into a single Total Cost of Ownership figure over the SHA term.
Assumptions used in this table (all per the introduction):
- Volume: 2,000 stream deliveries / 100 campaigns per year, held flat for the 15-year SHA term (Assumption #5)
- SC developer: €100/hr ≈ $4,280/week ≈ $171K/year loaded FTE (Market Assumption #1)
- SC audit: $4,000/day, 1-day minimum (Market Assumption #2)
- Cross-chain bridge: LayerZero at $1.50/tx mid (Technical Assumption #2)
- Cross-chain messaging: Hyperlane at $0.30/msg (Technical Assumption #3)
- Audit-refresh cadence: every 2 years over the 15-yr SHA term
- Option A "Annual TCO" assumes a 50/50 split of campaigns between Base and EtherLink; range shown is 0% EL adoption (low) → 100% EL adoption (high)
- Variable OpEx scales linearly with volume. Engineering, monitoring, and audit refreshes are largely fixed.
| Today (Base) | Option A — Parallel | Option B — Metadata | Option C — Facade | |
|---|---|---|---|---|
| BUILD COST (one-time CapEx) | ||||
| Contract development (dev-weeks) | — | 4–6 | 2–3 | 8–12 |
| SC dev time @ €100/hr (~$4,280/wk) | — | $17–26K | $9–13K | $34–51K |
| Audit fees (@ $4K/day) | — | $8–12K (2–3 days) | $4K (1 day) | $48–100K (12–25 days, often two firms) |
| Integration, deployment, docs | — | $3–5K | $2–3K | $10–15K |
| Calendar time to ship | — | 4–6 weeks | 2–3 weeks | 8–12 weeks |
| One-time build total | — | ~$35K | ~$17K | ~$130K |
| COST OF OWNERSHIP (annual, @ 100 campaigns / 2K deliveries / year) | ||||
| Variable OpEx (bridge/messaging fees) | nil | nil–$2.85K (LayerZero @ $1.50/tx) | $27 | $660 (Hyperlane @ $0.30/msg) |
| Contract-engineering capacity required | — | ~10% senior FTE ($17K/yr) | ~3% senior FTE ($5K/yr) | ~20% senior FTE ($34K/yr) |
| Monitoring, RPC, indexer infra | — | $3K/yr (new EL provider) | $1K/yr | $8K/yr (cross-chain monitoring stack) |
| Incident response (on-call, post-mortem) | minimal | ~$5K/yr (Transfer Cap incidents) | minimal | ~$15K/yr (LZ/Hyperlane degradation, stuck msgs) |
| Audit refresh (~2 days every 2 yrs @ $4K/day) | — | $4K/yr amortised | $2K/yr amortised | $15K/yr (cross-chain re-audit longer) |
| Annual TCO (typical case, 50% EL mix) | <$0.1K | ~$29.5K | ~$7.5K | ~$73K |
| Annual TCO (high case, 100% EL) | <$0.1K | $30.85K | $7.5K | $73K |
| CUMULATIVE TCO @ 100 campaigns/year flat | ||||
| 5-year TCO | <$1K | $148K (typical) / $154K (high) | $38K | $363K |
| 10-year TCO | $1K | $295K / $309K | $75K | $727K |
| 15-year TCO (SHA term) | $1K | $443K (typical) / $463K (high) | $113K | $1.09M |
| Build + 15-yr TCO | — | $478K – $498K | ~$130K | ~$1.22M |
| Delta vs Today over 15 yrs | — | +$478K to +$498K | +$129K | +$1.22M |
What this table makes visible that the per-campaign view doesn't
- Engineering carrying cost dominates every line at this volume. Variable bridge/messaging fees are small because deliveries are small. C's $73K/yr Annual TCO is ~99% engineering + monitoring + audit refresh; only ~$0.66K/yr is variable. The commitment is mostly about funding ongoing engineering capacity, not transaction fees.
- B's number is almost all fixed cost. B's $113K 15-yr TCO is essentially engineering overhead — variable fees are negligible at this scale.
- A's range is narrow at this volume ($478K–$498K) — adoption barely matters at 100 campaigns/year because per-campaign variable fees can't dominate fixed engineering. At higher volume the range widens.
- B is the only option where Build + 15-yr TCO stays below $150K. Every other option is in the six-to-seven-figure range, even at conservative volume.
- The numbers above are the floor. They assume volume never grows past 2K deliveries/year. At a more aggressive 50K deliveries/year (= 2,500 campaigns), variable OpEx scales 25× while fixed costs stay roughly the same. Realistic Build + 15-yr TCO at 50K deliveries/yr lands around A: $0.9M–$1.4M / B: $135K / C: $1.4M. The structural ranking (B << A < C) holds at every scale.
Operational risk surface
A quick view of the new failure modes Verifluence inherits under each option, beyond what already exists in today's single-chain Base operation.
| Risk | Today | Option A | Option B | Option C |
|---|---|---|---|---|
| Bridge liquidity exhaustion (LZ "Transfer Cap Reached") | — | Real (EL path) | — | — |
| Cross-chain message stuck / delayed | — | minor (off-bridge UX) | — | Real (every claim) |
| Cross-chain message replay / signature forgery | — | minor | — | Real (catastrophic if exploited) |
| Source-chain reorg vs cross-chain delivery | — | minor | — | Real (small but nonzero) |
| Backend writer key compromise | — | — | minor (registry pollution only) | minor |
| EtherLink sequencer outage | — | EL campaigns halt | — | Platform halts |
| LayerZero validator-set compromise | — | bridge funds at risk | — | escrow logic compromised |
| Operator support tickets ("where's my payout") | low | moderate (EL path) | low | moderate (cross-chain delays) |
Recovery scenarios — what an incident actually looks like
Option A — Transfer Cap incident
- Friday evening: a large campaign settles. 80 streamers off-bridge their payouts.
- LayerZero's USDC pool on Base, which had ~$500K capacity, drains by ~streamer #50.
- Streamers #51–80 see their off-bridge tx revert with "Transfer Cap Reached." Their bridged USDC is still in their EtherLink wallet — not lost, but stuck until pool liquidity refills.
- Support load: ~30 tickets, all variants of "where is my money."
- Recovery: wait for LZ LPs to refill (typically hours, can be overnight); proactively notify affected streamers; consider a "Verifluence-funded buffer" that off-bridges via an alternate route during incidents (adds ongoing cost, not currently in budget).
- Repeat frequency: correlates with large-campaign clustering; not predictable.
Option B — Backend writer outage
- Verifluence backend writer process crashes; nobody notices for 4 hours.
- During the outage, ~50 deals fund + ~80 claims happen on Base. The EtherLink registry is now 4 hours stale.
- Visible impact: the EtherLink registry's
lastSyncedBaseBlockshows a 4-hour gap if anyone queries it. - Recovery: restart the writer; it replays from
lastSyncedBaseBlockand posts catch-up attestations. Costs ~$5 in EtherLink gas for the backlog. - User impact: zero. Operators and streamers were never affected.
Option C — LayerZero degradation
- LayerZero relayers are slow due to a destination-chain gas spike. Messages from EtherLink → Base are delayed from 5 min to 2+ hours.
- During the slowdown: 200 streamer payouts are queued. None complete.
- Support load: 200 tickets, all "where is my payout."
- Recovery: wait for LZ relayers to catch up; emergency rate-limit new claims so the backlog drains; consider manually paying the higher relayer fee on the worst-stuck messages (~$5 each, paid by Verifluence).
- User impact: real and visible. Once it persists past ~1 hour, social-media backlash starts.
The fundamental difference: B's incident scenarios don't reach the user. A's and C's do.
Next
The financial implications of each architecture — how the costs above translate to share-price impact on founders, TG3, and future investors — are in Share performance projection.