GnosisDAO · EVRO · March 2026

EVRO capital deployment

RaidGuild built EVRO · Elementary Complexity helped · For GnosisDAO and NOCTUA

March 2026

Executive orientation

GnosisDAO inputs liquidity; Noctua runs execution. RaidGuild holds continuity; EVRO DAO steers the interestRouter stream, dynamically.

This plan is a draft "working map" for the deployment — clear enough to argue from, light enough to read in one pass.

What's in here (in order):

1) How much — The Stability Pool / Dex / Branche splits and why those numbers where chosen.

2) WhenFoundation first, Deepen after an explicit checkpoint, then post-launch rhythm for incentives and continuation. (and a one shot version).

3) Where — Stability Pools across six branches; EVRO/Euro-e depth on Balancer v2; and the governance moves to redirect fees when needed.

4) Why this shape — Stabilty Pool depth first testing the protocol mechanics; Balancer for EVRO-EURe liquidity pair to open incentive routing options.

Let's treat this as a step for coordination: I've suggested precise numbers so nothing is fuzzy, but all can be discussed.

Decision logic

tl:dr: A structured capital allocation plan taking into account all branches, EVRO-EURe liquidity provision, Dex Choice, Governance and incentivization consequences in 2 versions: phased and one-shot mode.

If this document works as intended, we can move fast without ambiguity. (Sorry that it took a minute to make)

A note from the human

Hello Dave and Julian! Elco here: When tasked with the creation of a plan on a topic that I knew very little about, I did what any sane human would do: I built an agentic consulting company mirroring the way I would go about it and used it to learn as much as I could about the topic. The following plan is the best of my efforts. It is the result of human-machine collaboration, but every word has been authored.

Methodology: Gathered documentation and contracts of all the protocols and actors involved and market facts we could verify directly (how the system works, fixed constraints, and market/execution reality). Those facts where retrieved separately from judgment calls (staging, reserve size, and rollout timing), compared options against the same risk and execution criteria, and selected the strongest overall policy. AI support helped speed up synthesis and consistency checks, but final assumptions were reviewed by humans and written in plain language so stakeholders can see what is certain, what is a decision, and what is still open.

Core assumptions and allocation modes

Core assumptions

ItemValueMeaning
Total capital envelopeEUR 5,000,000Full treasury program size for this deployment scope
Branch setWXDAI, GNO, sDAI, wWBTC, osGNO, wstETHAll branches are, of course, taken into account in the deployment architecture
Interest routing75% SP / 25% interestRouterFixed in code; governance allocates only the 25% stream - But can be used for incentives via the Evro DAO

3.1 Options tested

OptionModeSPDEXReserveWhy this option exists
AStaged72%26%2%More DEX depth with checkpoint control
BOne-shot balanced74%23%3%Simpler execution with larger buffer
COne-shot conservative76%21%3%Higher SP and reserve if uncertainty is very high

3.2 Active policy

ModeSPDEXReserveAmounts (EUR)
Default (active): staged72%26%2%3,600,000 / 1,300,000 / 100,000
Fallback: one-shot74%23%3%3,700,000 / 1,150,000 / 150,000

3.3 Why split changes by mode

Why reserve differs?

Staged

A checkpoint between phases means the reserve only needs to cover the Foundation window blast radius. Overcommitment risk is bounded — you validate before you deepen.

One-shot

Without a checkpoint, the full deployment is live from day one. The reserve must absorb a full-deployment shock on its own, which is why the buffer is sized larger.

Why DEX differs?

Staged

Once Foundation validates pool behavior, more capital can be committed to DEX depth in Deepen. The checkpoint unlocks deeper liquidity without blind risk.

One-shot

DEX is sized more conservatively because there's no checkpoint to catch a routing failure mid-flight. The trim preserves margin against early pool underperformance.

Operational burden?

Staged

More checkpoints, more coordination calls, more recommit actions. The ceremony is intentional — it's what limits the blast radius. But it costs attention and time.

One-shot

Fewer meetings, fewer decisions mid-flight. Execute once, monitor, and act only if triggers fire. Lower overhead, higher trust in the preparation.

Main tradeoff?

Staged

Higher process overhead — but if assumptions are wrong, damage is contained at each gate. You catch mistakes before they compound into a full-capital event.

One-shot

Lower process overhead — but if assumptions are wrong, the full deployment is already live. Speed and simplicity at the cost of mid-flight intervention room.

Position map

4.1 Full deployment by position (EUR 5,000,000)

The table below is the full capital envelope by line item. Staged default (72% / 26% / 2%) and one-shot fallback (74% / 23% / 3%) match canonical policy in §3; SP branch one-shot targets scale the same within-SP weights from EUR 3.6M to EUR 3.7M (× 3.7 / 3.6).

PositionLaneStaged targetOne-shot targetSP share% of 5M (staged)% of 5M (one-shot)Notes
SP-WXDAISPEUR 900,000EUR 925,00025%18.0%18.5%Native stable anchor and lowest-volatility reliability lane
SP-sDAISPEUR 720,000EUR 740,00020%14.4%14.8%Stable/yield lane for steady utilization
SP-GNOSPEUR 630,000EUR 647,50017.5%12.6%12.95%Ecosystem-relevant collateral with moderated volatility exposure
SP-wstETHSPEUR 540,000EUR 555,00015%10.8%11.1%High-liquidity diversification beyond Gnosis-native assets
SP-wWBTCSPEUR 450,000EUR 462,50012.5%9.0%9.25%Non-ETH beta diversification at controlled initial weight
SP-osGNOSPEUR 360,000EUR 370,00010%7.2%7.4%Ecosystem-aligned staked exposure with correlation control
DEX-EVRO-EURe-CSPDEXEUR 1,300,000EUR 1,150,00026.0%23.0%Primary EVRO <-> Euro-e liquidity venue — in staged mode: EUR 0.8M in Foundation, then EUR 0.5M in Deepen after checkpoint
OPS-RESERVEReserveEUR 100,000EUR 150,0002.0%3.0%Controlled execution buffer — not a yield lane; release only with rationale, expected effect, replenishment plan
TotalEUR 5,000,000EUR 5,000,000100%100%

4.2 Branch guardrails

GuardrailRule
Single-branch concentrationNo branch above 30% of SP total in first 30 days
Correlated GNO exposureGNO + osGNO not above 35% without explicit sign-off
Rebalancing disciplineRebalance at scheduled checkpoints or incident triggers

4.3 Reserve usage policy

Allowed usesNot allowed uses
Emergency rebalance, slippage defense, bridge/timing mismatch, incident stabilizationDiscretionary side bets, permanent yield allocation, undocumented ad hoc spending

DEX venue

Balancer v2 vs Curve

Balancer v2 is the chosen DEX venue because it combines stable-pair fit with better composability and a clearer incentive/governance path for EVRO on Gnosis. This is not a "Curve is bad" claim. It is a "Balancer is better for this mission setup" decision.

5.1 What matters for this mission

Decision factorWhy it matters for EVRO deployment
Pool math fitEVRO and EURe are same-currency euro stable assets
ComposabilityNeed routing depth beyond one isolated pair
Incentive pathNeed realistic gauge and vote-routing strategy
Gnosis deployment qualityMust run on confirmed infra, not assumptions
Governance/onboarding pathNeed a practical path to gauge listing

5.2 Side-by-side comparison

FactorBalancer v2CurvePractical conclusion
Stable pair fitComposable Stable Pool (stable-swap style math)StableSwap poolBoth fit stable pairs
ComposabilityNestable BPT enables broader routing designLess composable structure for this patternBalancer advantage
Gnosis deployment/infrastructureConfirmed Vault/factory/gauge stackAvailable but less integrated in this mission pathBalancer advantage
Incentive routingveBAL + gauges + Aura/bribe markets, documented pathveCRV/Convex ecosystem pathBoth possible, Balancer path clearer for this mission
Partner onboardingDocumented onboarding pathLess structured onboarding path in this workflowBalancer advantage
BIP19-style upsidePotential fee-to-bribe reinforcement path if qualifiedNo direct equivalent in this framingBalancer advantage

Deployment plans by mode

6.1 Staged mode plan (default)

PhaseWindowCapital actionsGate condition
FoundationDays 1-10SP: EUR 2.8M across 6 branches; DEX: EUR 0.8M; reserve untouchedBranch coverage >=4, acceptable pool behavior, no critical incidents
DeepenDays 11-21Add SP: EUR 0.8M (to 3.6M total); add DEX: EUR 0.5M (to 1.3M total)Protocol is stable, safe to launch.
Post-launchCyclic (TBD)Re-evaluate incentive weighting / consider governance incentivesGrowth must outweigh the incentive cost.

6.2 One-shot mode plan ("I'm feeling lucky")

One-shot mode is in this plan because staged deployment, while safer, can be a real pain in the ass operationally: more checkpoints, more coordination calls, and more chances to lose momentum in process overhead. One-shot is the lower-ceremony option when we trust the preparation and need to execute in one clean window. We choose it when team bandwidth is constrained, governance timing windows are tight, or the coordination burden is more costly than the value of mid-flight checkpoints. It is not the default, because if assumptions are wrong the blast radius is larger, but it is a valid mode when speed, simplicity, and execution certainty are the priority. Up to Dave.

ItemPolicy
Execution windowTarget 48-72h band
Capital actionDeploy full SP and DEX targets for selected one-shot split
Reserve behaviorKeep untouched unless trigger conditions occur
Post-window (days 3-10)Monitoring and incident handling only; no additional deployment actions
Use one-shot only whenTeam bandwidth is constrained, governance windows are tight, or lower process overhead is preferred over checkpoint flexibility

Validation layer

Soft launch cohort policy (suggestion)

Without invited users, you can validate mechanics but not real user behavior. The soft launch cohort closes this gap before broad awareness.

Aim for a group that covers the main behavioral lanes — operators who can stress-test execution readiness, LP-oriented participants who'll probe depth and liquidity behavior, and borrower or swap-route testers who exercise the user-facing mechanics.

Somewhere between ten and twenty participants gives enough signal diversity without turning coordination into a project in itself; a leaner group still covers the essentials if participation is constrained.

Composition matters more than headcount.

The suggested distribution roughly mirrors the validation priorities: a few operators, a larger LP cluster, and a handful of borrower and swap testers.

8.1 Invite profile and validation target

Stakeholder typeWhy inviteWhat they validate
Treasury-aligned LPsPractical liquidity feedbackDepth quality under live conditions
DeFi-native borrowersCollateral/debt behavior experienceTrove usability and branch behavior
Market makers/active swappersFast execution quality feedbackSlippage, routing, market function
Ecosystem operatorsKnow Gnosis operations contextMonitoring and escalation readiness

8.2 Expected contribution

Expected actionPurpose
Execute real transactions in defined windowsValidate actual system behavior
Follow simple scripts (borrow/swap/deposit/unwind)Standardize comparable results
Report friction quicklySurface operational bottlenecks early
Respect risk caps and test-size limitsKeep launch validation controlled

Timing logic and hypotheses (what each phase tests)

9.1 Foundation (days 1-10)

ItemSummary
HypothesisInitial SP + DEX depth should produce stable first behavior without full commitment
TestsExecution quality across branches, early routing quality, concentration/peg stress signals
Why this windowLong enough for signal, short enough to avoid overcommitment
Success modePositions live, no critical incidents, acceptable market quality, no persistent peg stress

9.2 Deepen (days 11-21)

ItemSummary
HypothesisDeepening should improve quality and resilience measurably
TestsDepth delta effect, branch weighting quality, readiness for full target exposure
Why this windowEnables pre-deepen vs post-deepen comparison while preserving intervention room
Success modeQuality improves, guardrails hold, no critical incidents post-deepen

Ten days is a deliberate choice: long enough to observe real market behavior across a complete cycle — including a weekend, when liquidity conditions can differ materially from weekdays. Short enough that if something is wrong, you haven't overcommitted to the observation window before you can act. Foundation is the stretch that either justifies Deepen or stops it. The window can be adjusted; what matters is that the gate passes explicitly, not automatically.

Everything behaving according to plan? Early users satisfied, positions live, peg stable? → Launch.

Governance

Early liquidity incentives

Launch is also when the 25% interestRouter stream starts to matter for how depth gets funded over time. On Balancer, sustained depth runs through gauge listing and the vote economy—where EVRO DAO can eventually point that quarter of borrowing interest. A few governance and coordination moves should line up before and right after go-live: light scaffolding so incentive routing stays legible, owned, and adjustable as the pool and branches prove themselves out.

Know who decides before Foundation starts.

We (Noctua, RaidGuild, GnosisDAO, and EVRO DAO) do not need a thick playbook — But we do need a shared map of who can commit what, and where execution sits. At Launch, RaidGuild will control 30% and Gnosis, 15% of Governance Power. The remaining 55% are for the DAO to manage - the token, named "RETVRN" hints at the mandate of the DAO: To manage the interestRouter

Post Launch reality for a competitive protocol:

The numbers in this memo attempt to define the deployment we actually execute with and their rationale. But that's just the beginning. This section is a draft of the governance consequences and how we might help the protocol deepen accrued TVL and volume over time. Balancer was chosen in part because it would couple our pool design with a mature, well-documented emission governance layer (veBAL → gauges), which is a credible long-term liquidity story and here we leverage personal relations already built with Balancer operators. (Our team does not currently have the same access on Curve)

How the Balancer incentive lane would plug in:

  • 75% of borrowing interest → Stability Pool (immutable in code).
  • 25%**interestRouter**EVRO DAO (Moloch v3 / Baal on DAOhaus). Governance proposes and passes initiatives that spend this stream—including LP incentive direction.
  • Stewardship of the split and the 25% is ongoing DAO work. Early Foundation phases might emphasize direct LP-facing uses of available incentive budget (e.g. how the 25% is deployed, day one). Longer term implies locks, vote-directed liquidity (veBAL / vlAURA), RETVRN utility design, ecosystem partnerships, and treasury participation on Balancer and Aura with bribes (e.g. Hidden Hand / Redacted) where it improves emissions efficiency.

Our Shared Stewardship

If the euro stable works and attracts TVL, the protocol earns in the obvious ways. Interest, though, is not the whole opportunity (it's the floor). The larger upside for everyone who carries this past launch is RETVRN: how it captures growth, how value accrues in a way holders and contributors can reason about (treasury, locks, fee surfaces, emissions disciplines), and how it stays aligned with usage on Gnosis. The name is a hint at the mandate: stewardship of the interestRouter stream and the decisions that turn a live pool into sustained depth and durable protocol equity.

But the people who lead partnerships, risk, integration, and the unglamorous work of governance, pitching, tokenomics design need to be considered on the next step of our plan for a post launch reality. The 25% is one lever; but RETVRN design and incentives are how we reward the cohort that builds for the long game, not only the quarter where the "peg held".

That is a battle for another day... what y'all think of this deployment plan?

EVRO was built by RaidGuild. Copy on this page is just orientation for final decision making. Elco has written the whole thing, understood the calls and trade offs - it's way less machine made than you think... so I'll put a joke on the footer:

Knock knock. Who's there? Not a model.