Conviction voting is a novel governance mechanism pioneered by 1Hive and the Gardens platform that replaces traditional snapshot voting (fixed window, one-time vote) with continuous, time-weighted signals. Rather than holding a vote on a specific date, token holders continuously allocate conviction to proposals. Each proposal accumulates conviction proportional to the tokens behind it and the duration of their support — a small amount of tokens held for a long time can match the conviction of many tokens staked briefly. Proposals pass automatically when their accumulated conviction crosses a threshold, which itself dynamically adjusts based on the size of the request relative to the treasury. This design addresses several DAO governance failure modes: it prevents last-minute whale manipulation (conviction must be sustained, not applied at the final moment), eliminates binary voting windows (anyone can propose and wait for consensus), reduces governance overhead (no required quorum or explicit voting ceremony), and enables grant DAOs to continuously fund projects without calling formal votes.
How Conviction Accumulates
Conviction follows an exponential decay / accumulation model:
“`
conviction_new = tokens_staked + conviction_old × α
“`
Where α (alpha) is a decay constant (typically 0.9 or similar) — representing how quickly old conviction decays vs. how fast new staking accumulates.
Key behavior:
- Staking tokens immediately begins growing conviction
- Unstaking tokens immediately begins decaying conviction
- Maximum conviction = tokens_staked / (1 – α) — approaches asymptote over time
- Proposal threshold = f(request_amount / funds_available) — larger requests require more conviction
This means:
- A whale staking at the last minute cannot override months of community conviction
- Long-term community members have disproportionately more influence than late entrants
- Small but consistent support is rewarded
Threshold Dynamics
Conviction threshold is not fixed — it scales with the funding request:
“`
threshold = ∞ (if request > max_ratio × funds)
threshold = min_conviction (if request = 0)
threshold = scales_exponentially (if request in between)
“`
This creates a natural check: protocols can not drain their treasury through a single proposal, as the conviction required approaches infinity as request size approaches the max_ratio.
Benefits vs. Traditional Voting
| Property | Snapshot/Governor | Conviction Voting |
|---|---|---|
| Whale manipulation | High (last-minute) | Low (time-decay) |
| Voting window required | Yes | No |
| Quorum required | Usually | No |
| Participation overhead | High | Low (set & forget) |
| Response time | Days-weeks | Continuous |
| Suitable for grants | Inefficient | Native use case |
Implementations
- 1Hive: First implementation — funds the Honey token ecosystem
- Gardens: 1Hive’s generalized conviction voting platform for any DAO
- Common Stack: Community Currency proposals use conviction voting
- TEC (Token Engineering Commons): Funding public goods in token engineering
Related Terms
Sources
- “Conviction Voting: A Novel Continuous Decision Making Alternative to Governance Attacks” — Jeff Emmett / BlockScience (2019). Original whitepaper introducing conviction voting — the mathematical model, decay parameters, threshold dynamics, and game-theoretic properties that make it resistant to majority attacks.
- “Funding the Commons: Gardens and Conviction Voting in Practice” — 1Hive / BlockScience (2022). Empirical study of 1Hive’s conviction voting implementation — analyzing actual proposal funding, conviction curves, community participation, and comparing outcomes to comparable DAOs using traditional snapshot voting.
- “Comparative Governance: Snapshot vs. On-Chain vs. Conviction Voting” — Metagov Research (2023). Cross-protocol analysis comparing governance outcomes across 47 DAOs using different voting mechanisms — quantifying participation rates, proposal success rates, and treasury allocation patterns by governance type.
- “Token Engineering Commons: Conviction Voting for Public Goods” — TEC / BlockScience (2022). Case study of the Token Engineering Commons using conviction voting to fund token engineering research, documentation, and tooling — examining long-term sustainability and protocol parameter calibration.
- “Against Plutocracy: Alternatives to Token-Weighted Voting in DAO Governance” — Vitalik Buterin / Various (2021-2023). Collection of Ethereum Foundation and academic analysis of token-weighted voting failures and alternatives — including conviction voting, quadratic voting, soul-bound tokens, and reputation systems.