Executive Summary
Yesterday, Aave Labs released the initial Aave v4 hub and spoke configurations. You can review them here.
At genesis, v4 launches with 3 hubs, Core, Prime, and Plus, supporting 10 spokes. Core is the general purpose liquidity hub. It launches with 7 spokes and serves as the main source of inter hub credit lines. Prime targets suppliers who want ETH and BTC collateral exposure without being pulled into borrower driven rate dynamics. Plus is designed to let Ethena adjacent strategy flow scale behind dedicated caps, pause conditions, and risk controls.
With the base configuration set, the real question is how the DAO should grow v4. How should expansion be paced, how should credit lines be managed, and how should adoption quality be measured across hubs and spokes without outgrowing risk operations?
Aave v3 has already proven durable supply and borrow demand across cycles. What limits v4 is not demand, but the DAO’s ability to provision capacity safely and quickly under a new architecture, fast enough to capture existing demand before competitors with stronger execution do. That calls for a risk adjusted growth approach for v4. Capacity should expand only as monitoring, controls, and operational maturity keep pace.
This framework formalizes that approach. It defines pacing logic for expansion, sets criteria for allocating and adjusting credit lines, and lays out three adoption paths for 2026 to 2028 evaluated through a risk first lens.
Starting Point: What the Initial Configuration Establishes
For v4 to win, adoption should be paced by risk, not TVL. The gating question for any expansion is simple. Has v4 earned the right to grow safely?
Separately, early adoption metrics will be noisy. v4 will run in parallel with v3 for 24 to 36 months, so initial TVL will be dominated by migration flows. Incentives can shift deposits without real preference, and rate differentials can inflate borrows through looping and arbitrage without creating net new credit demand. That means the DAO needs adoption measures that distinguish real usage from mechanical volume.
With that context, the next section presents 3 viable adoption paths, each with its own definition of winning.
Three Adoption Paths
2026 Conservative Path
Core Narrative:
The Conservative path treats v4 as a protocol that must earn the right to grow by demonstrating safety first. This path is designed to prove that v4 can sustain deep liquidity across a 3-hub and 10 spokes with a deliberately bounded asset set. In early v4, concentration is a feature: simpler surfaces and tighter risk scope make it faster to learn whether the system remains stable at meaningful scale.
The real proving ground is a major volatility event when liquidations are triggered under v4’s variable bonus mechanism and the protocol performs as designed end-to-end. Until that has happened and been rigorously postmortemed, spoke launches remain intentionally slow and cap increases follow a conservative cadence.
In 2026, the work is operational proof: demonstrating that the architecture already in place can be run safely under real conditions including through stress at scale.
KPIs that matter in 2026
KPIs should be reported by hub rather than aggregated, because Core, Prime, and Plus have different mandates and different failure modes. A single blended number will routinely mask the signals that matter most for pacing growth and managing risk.
For the Core Hub, the primary KPI in 2026 is cap utilization by spoke. Under a conservative operating posture, most spokes should remain below roughly 60% utilization, with only a small number intermittently touching ~80%. The 4 e-Mode spokes are the most likely to approach ceilings first given loop-driven demand, and that can be acceptable so long as liquidation infrastructure and parameterization have been validated under stress. In parallel, Core’s supply composition should be legibly “healthy”: USDC and USDT should lead stable supply in the Main Spoke, with wETH serving as the default collateral baseline. The Gold Spoke and Forex Spoke should not be judged on volume in 2026; they exist primarily as behavioral observatories, where the goal is to study liquidity quality, usage patterns, and operational load rather than to optimize growth.
For the Core-to-Plus credit line, the key KPI is draw rate relative to cap, because this is the primary systemic coupling at genesis and it extends Core exposure beyond Core boundaries. If cUSDT, cUSDC, cfrxUSD, and cRLUSD inventory is running consistently above 70% of the cap, that should be treated as an explicit governance trigger requiring a formal decision: either raise the cap with evidence that the incremental exposure is justified and safely monitorable, or require Plus to build its own organic stablecoin supply depth rather than leaning on Core capacity. A 2026 dashboard that does not surface this draw-versus-cap number prominently is incomplete.
For the Prime Hub, the central KPI is whether the non-borrowable collateral posture actually delivers the withdrawal availability and predictability it promises. If a volatility event produces meaningful queue dynamics or unexpected rate behavior, that should trigger a postmortem regardless of whether losses occur, because it indicates the core Prime value proposition is not holding under stress.
On the borrow side, borrow-to-TVL per spoke is a weak signal in 2026 because governance caps, not demand, are the binding constraint. GHO should start to appear as a small share of stable borrows, but there should be no target to hit; the question is whether it emerges organically without incentive distortion.
Service provider seat:
In 2026 the work is research-heavy, governance-facing, and operationally strict because the system is still validating an ambitious launch configuration.
-
Primary SP lead: a risk provider like @LlamaRisk that publishes asset-by-asset collateral assessments, maintains a versioned cap methodology governance can reuse, and delivers fast postmortems for every liquidation and any incident-level anomaly across all three hubs.
-
Support: @ChaosLabs stress-test initial spoke and e-Mode setups, propose conservative curves for newer spokes (Gold/Forex), and provide simulations for credit line cap sizing.
-
Observability: @TokenLogic v4 dashboards, standardized KPI analytics, alerting, and incident rubrics designed for the three-hub model.
-
Tooling: @AaveLabs v4 operational tooling, including instrumentation that surfaces credit line draw rates and cross-hub exposure to governance in real time.
Success Measurement:
A Risk Profile:
The main risk in 2026 is that 10 spokes create 10 monitoring surfaces simultaneously and the operational infrastructure to cover all of them at the required cadence does not yet have a track record in v4.
2027 Base Path
Core Narrative
By 2027, v4 has enough live history to be judged on evidence. Liquidations have cleared through stress, the e-Mode spoke mechanics have behaved as designed, the Prime Hub isolation promise has been tested, and the Core-to-Plus credit line has a real utilization history for governance to read. The question shifts from whether the initial configuration can be operated safely to whether the architecture can absorb new spokes, new credit line configurations, and potentially new hubs without turning governance into continuous firefighting.
The expansion in 2027 is about introducing more spoke categories if the incremental headroom justifies the operational load. Credit line governance becomes a routine DAO function in 2027 rather than an architectural novelty. By mid-2027 the Core-to-Plus credit line has a utilization history, and governance has to decide explicitly whether the right answer is to raise the cap, hold it steady, or require Plus to develop independent stablecoin supply depth. That decision sets the precedent for how credit lines are used across the system as new hubs are introduced.
KPIs that matter in 2027
A clean success signal for 2027 is that v4 Ethereum TVL overtakes v3 Ethereum TVL on-chain and stays there, because that shows the market is choosing v4 as the main venue, not just experimenting with it. Active hubs should expand beyond the single hub model, with a reasonable target range of 6 to 8 hubs depending on how strongly the DAO prioritizes controlled diversification versus operational simplicity. Live spokes should expand into 12 to 20 range, but the number alone should not be treated as success because the point is that each new spoke is added with a clear reason and a clear risk boundary.
Integration share should climb toward 20% to 40% as the developer surface area of v4 broadens. Borrow to TVL per spoke should move higher into the mid 50s to low 60s as confidence improves and caps expand, while still staying disciplined rather than maximizing leverage. Here, Aave v4 borrows outstanding can plausibly grow from roughly $2.1B toward $8B through the end of 2027 if capacity is expanded on a steady cadence.
GHO should begin to show up as a meaningful part of stable borrows, with a target range around 3%-8% inside Hubs/Spokes borrow mix if peg stability and integration depth are strong enough.
Credit Line in 2027:
By 2027, three conditions should be evaluated before any credit line cap raise: whether Plus Hub’s organic direct supply has been growing alongside credit line draws, or whether the credit line is substituting entirely for organic supply depth; whether the correlation between Core Hub stress and Plus Hub stress has been characterized empirically; and whether credit line utilization has shown stable or improving patterns under normal conditions rather than spiking during stress.
If Plus is consistently drawing against the full cap without building organic supply, the right answer is to require Plus to bootstrap its own stablecoin liquidity before the cap is raised further. If organic supply and credit line draw are both growing proportionally, a cap raise is well supported. This logic should be the explicit governance framework, written into a credit line policy rather than decided ad hoc.
Service provider seat
In 2027, risk coverage should be co-led (not a single SP) because the system has more surfaces and the questions evolve.
-
Co-leads: @LlamaRisk and @ChaosLabs owns credit line policy and new collateral assessments and quantitative cap sizing/stress testing for new spoke types.
-
Growth analytics: @AaveLabs and @TokenLogic focus on cohort retention by collateral type, incentive payback periods, and integration decomposition to separate sticky organic usage from incentive-driven TVL. If the first external spoke deployers go live, their spoke-level metrics need to be tracked against the same quality standards as governance-initiated spokes.
Success Measurement:
Risk Profile
The most difficult risk shifts to hub-level concentration. Even if each spoke is isolated in its own parameters, the hub remains shared settlement liquidity, so correlated exposures can accumulate across spokes without appearing dangerous in any single spoke view. The e-Mode spokes in Core Hub are the most likely place for this to manifest, because wstETH, weETH, rsETH, and LBTC are all ETH-correlated and their liquidation dynamics under a severe ETH drawdown interact at the hub level even when per-spoke caps look fine.
2028 Aggressive Path
Core Narrative
By 2028, Aave v4 has earned the capacity to carry more complex collateral types and more externally-developed spokes without losing systemic risk visibility. It also means governance has built the automation infrastructure that keeps pace with a protocol operating across dozens or hubs/spokes, because manual cap review cycles simply cannot scale to that surface area.
An important decision crystallizes in 2028 that the initial configuration left open: which architectural family governs new hub expansions. The three models from the initial configuration framework, Model A monolithic with v3-style shared solvency, Model B risk-profiled with separate hub boundaries, and Model C asset-centric with venue boundary aligned to a specific domain, are no longer theoretical. They are live options with real trade-offs that have been observed in production.
RWA hubs are the most likely first case where the DAO is forced to make this choice explicitly, because RWA collateral does not fit naturally into Core’s general-purpose solvency boundary and its NAV-based oracle design requires domain-specific liquidation assumptions. A Model C asset-centric hub is the most likely candidate, and the governance framework for authorizing Model C hubs needs to exist before the first RWA hub proposal lands, not after.
In 2028, the top assets list should start to show items that do not feel native to v3, and they usually appear on the supply side first. RWA spokes tend to be supply heavy and utilization light, which can distort blended borrow to TVL and make the protocol look less efficient than it really is unless reporting is done per spoke and per hub.
KPIs that matter in 2028
2028 success is when Aave v4 total TVL across all chains overtakes v3 total TVL. Protocol-wide v4 over v3 dominance often lands in mid-2028, but the aggressive phase can start in late 2027 if dominance is already visible and the trajectory looks durable.
Aave v4 should be capable of running 12 to 18 active hubs and roughly 40 to 80 live spokes, but the important qualifier is that a meaningful share of spokes are external deployers, with a target around 30% or more. Borrow to TVL per spoke should sit around 60% in the mature spokes, and v4 borrows outstanding can plausibly sit in the $9B to $16B range if the system is scaling safely.
GHO borrow share should become meaningfully visible, with a target around 10% to 18% of stable borrows inside Aave if peg stability and deep spoke integration are both real.
The operational expectation is that hubs and spokes can be launched quickly, including multi chain or multi mandate hubs, while the risk system keeps up through frequent automated updates that react to volatility, liquidity depth, and oracle quality, because this tier assumes growth speed that human review alone cannot safely match.
The hub/spokes authorization framework, meaning the documented standard for which architectural model a new hub uses and what evidence a proposer must present, becomes a KPI in its own right. If the DAO is approving Model C hubs without a consistent authorization standard by mid-2028, operational complexity is already running ahead of governance capacity.
Success Measurement:
Service provider seat
- In an aggressive 2028, prioritize per-spoke incremental revenue vs. risk, routing forecasts, and cap sizing backed by a governance-approved dynamic cap methodology that’s automated via guardrails. Expand stress testing to the Reinvestment module and vault positions, which need new models and liquidation assumptions.
- Keep a qualitative risk lead, but shift the job to standards and enforcement: Hubs/Spokes deployer authorization, hub-level concentration reporting, incident accountability, and a consistent decision framework for Model A/B/C hubs.
- Add an adoption-quality SPs to distinguish real usage from incentive TVL, and stand up a structured external contributor program for parameter research and monitoring under standardized templates.
Risk Profile:
The failure mode of the Aggressive path is not a single bad asset or a single liquidation incident. It is operational complexity overwhelming the system’s monitoring capacity. Governance focus shifts toward hub level exposure caps, loss stack design, spoke authorization standards, and credit line management, with standardized reporting and automated guardrails becoming the default rather than optional.
Learning from Uniswap
Uniswap is a useful analogy, but only if we treat it as directional. Liquidity migration between major protocol versions is not automatic; it often requires both a compelling technical upgrade and targeted incentives. Historically, Uniswap’s v4 launch was paired with an explicit governance-driven incentive plan designed to kickstart migration via short tranches and ongoing recalibration.
The takeaway for Aave v4 is that a successful protocol migration is about earning trust, aligning incentives with real usage, and allowing liquidity to follow demonstrated safety and yield. Aave’s initial configuration is already more complex than Uniswap’s early v3 migration surface, which means the patience required for genuine adoption to build is higher, not lower. Mercenary liquidity attracted to a 10-spoke protocol with active credit lines is also harder to distinguish from sticky adoption than in a simpler pool structure, which is exactly why adoption quality tracking is not optional across any of the three paths above.
The Elephant in the Room
We can’t close an Aave v4 adoption framework without addressing operational continuity in the post-BGD Labs and ACI environment. Much of what made v3 dependable wasn’t just the codebase, it was the day-to-day execution across security reviews, risk operations, governance throughput, upgrades and maintenance, and incident response. With those functions now shifting to Aave Labs and other service providers, the protocol needs clear, comparable or better coverage to avoid creating operational gaps at exactly the moment the protocol is becoming more complex.
Trust will have to be re-earned through visible outputs. That means constant delivery, measurable results, a track record of following through on what was promised, and fast transparent communication when things slip or break. Low communication, ambitious product line narratives, or pushing past community feedback won’t substitute for the operational work that underwrites safety.
In the end, Aave v4 adoption will be constrained by whether the DAO and the broader community can continue to trust Aave operations enough to stay motivated and aligned through the transition. The three paths above are only realistic if that foundation holds..
Conclusion
The DAO does not need complicated Aave v4 KPIs. It needs a clear adoption path that prioritizes safety, supports sustainable growth, and improves revenue density. The initial configuration has been defined. What remains is the harder work of proving that architecture under real conditions, expanding it with discipline, and building the governance infrastructure that keeps pace with its complexity.
In that world, a smaller Aave v4 TVL with proven liquidation behavior across all 3 hubs, a well-managed credit line between Core and Plus, and clear spoke authorization standards is a better signal than a larger v4 TVL pushed onto untested surfaces.
The same logic applies to the rest of Aave’s product line. Before the DAO expands further, whether that is the Aave App, the Aave Card, or other distribution products, it should be confident that growth and safety are moving in the right direction in the core protocol first. These products should be judged the same way v4 should be judged: through measurable outcomes, clear ownership, and real operational readiness, not branding narratives.
At the end of the day, Aave wins by keeping liquidity deep, managing risk better than everyone else, and shipping reliably. Everything else is just leverage on top of that.
Disclaimer
This analysis presents conceptual adoption paths for governance discussion. All Aave v4 KPIs and Metrics remain subject to Aave governance, including AAVE token holder votes and established DAO Service Providers.
This analysis was produced independently by me. I did not receive payment, grants, or compensation from any party in connection with this work, and I am not affiliated with any contributor, service provider, or organization referenced herein.
Copyright
Copyright and related rights waived via CC0.