Audit Methodology

At launch, healthy projects keep team and advisor tokens locked so insiders cannot sell into the first wave of buyers. This measures how much of that allocation is instead liquid on day one and free to be sold, which lets insiders cash out and walk away early. A lower share is safer, and foundation tokens are left out since some early foundation liquidity is normal for funding the project.

Only a fraction of the supply is actually tradeable when a token launches, and this measures how much of that initial float sits with investors. Investors usually bought in early and cheaply, so when they hold a large share of the thin launch supply they can push the price down fast. A project can still pass with a big investor allocation overall, as long as most of it stays locked at launch.

The initial float is supposed to be freely tradeable at launch. Foundation tokens usually sit with the foundation rather than being liquid, so when a large part of that float is foundation controlled, the genuinely tradeable supply is smaller than the headline float suggests. A high reading means the float has been filled with tokens that are not actually circulating.

This tracks investor control of the circulating supply across the first four years and reads whether it is trending up or down. The more of the supply investors control, the easier it is for them to move the price or swing governance votes, so a climbing trend means power is concentrating in investors instead of spreading out as it should. A low or steady control trend passes, while a rise into elevated territory is flagged, gently for a slow climb and harshly for a fast one.

This scans investor control at each of the 48 monthly checkpoints and flags every month it exceeds 30% of circulating supply. Each of those months is a window where investors hold enough of the supply to move the price or tip a governance vote on their own. The verdict is about duration: a brief breach during peak vesting is far less severe than control that stays over the line for years.

This tracks team, advisor, and foundation control of the circulating supply across the first four years and reads whether it is trending up or down. The more of the supply this group controls, the easier it is for them to move the price or swing governance votes, so a climbing trend means power is concentrating in insiders instead of spreading out as it should. A low or steady share passes, while a rise into elevated territory is flagged, gently for a slow climb and harshly for a fast one.

This scans team, advisor, and foundation control at each of the 48 monthly checkpoints and flags every month it exceeds 45% of circulating supply. The bar is deliberately harsh because many projects sit near it, and control above 45% means a small group holds close to a majority of everything circulating, enough to move the price or carry a governance vote on its own. The verdict is about duration: a brief breach during peak vesting is far less severe than one that drags on for years.

This is the slice of the entire supply set aside for the wider community through airdrops, rewards, ecosystem programs, and staking, with the public sale counted separately. A large community share means ownership is spread across the people who actually use the project, rather than concentrated in insiders and investors. The bigger this slice, the more the token belongs to its users, which is the goal of a fair distribution.

This is the slice of the entire token supply, once everything has vested, that is promised to investors across the seed, private, and strategic rounds. Investors are in it to sell at a profit, so a large investor allocation is mostly supply that exists to be sold onto the market over time. The smaller their share of the whole, the less built-in selling pressure the token carries.

This is the share of the entire supply held by the team and advisors who build the project. A healthy middle keeps the builders motivated and aligned with holders, while too large a share concentrates control and leaves a lot of future selling hanging over the market. Too little can mean the team has little skin in the game, so this rewards a balanced range rather than simply less.

This is the share of the entire supply held by the foundation that stewards the project. There is a healthy middle here, enough to fund development and the ecosystem for years, but not so much that one entity quietly controls the token. Both too little and too much are problems, which is why this scores a range rather than simply rewarding more.

This is the share of supply that ordinary buyers could purchase directly, through a public sale, IDO, or ICO. It is often the only point where regular people get in near the same price as insiders and investors, instead of buying later and higher from those early holders. Having any genuine public sale at all is the big step, so even a small slice beats none.

With a cap, the supply can never grow past a set maximum, so dilution has a clear ceiling. This checks whether the supply is capped or can be minted without limit, where an uncapped supply can grow forever and every new token dilutes everyone already holding. A fixed maximum is the safer structure, since open-ended minting hands the team a permanent lever over the supply.

A readable schedule resolves to the full supply with a defined end for every pool. This measures how much of the total supply the combined schedule actually reaches by its final disclosed month, and when even one pool has no disclosed finish or is left to later governance, the end of the supply is unknown and future dilution cannot be predicted. The closer the schedule lands to 100%, the more honest and predictable the dilution ahead.

The circulating float is the supply genuinely trading at launch, and the unlocked float is all the supply that is free to sell, whether or not it is on the market. Everything circulating is also free to sell, so the unlocked float is always the larger of the two, and the gap between them is phantom float, supply that is unlocked but held off the market. It usually sits in team or foundation wallets, so the unlocked float looks deeper than the supply that is really trading, while that held supply can hit the market at any time. The smaller the gap between the two floats, the more honest the launch.

This measures the share of total supply already circulating at launch. A larger launch float means later unlocks land as a smaller share of what is already trading, so each release dilutes holders less. Higher is better, since a thin launch float makes every early unlock a large shock relative to the supply on the market.

This measures the share of total supply unlocked at launch, free to sell whether or not it is circulating. A larger unlocked float means later unlocks add a smaller share to what can already be sold, so each release weighs less. Higher is better, though read it with phantom float, since an unlocked float that far exceeds the circulating float is controlled supply rather than real liquidity.

This measures how evenly monthly unlocks are spaced, as the spread of monthly unlock sizes where lower is smoother. Smooth release lets the market absorb supply steadily, month after month. Lumpy cliffs create sudden volatility and obvious points for holders to sell into, since everyone can see the big unlock coming. Lower is better.

This measures how evenly monthly unlocking is spaced, as the spread of monthly unlock sizes where lower is smoother. Smooth release lets the market absorb sell-eligible supply steadily, month after month. Lumpy cliffs create sudden volatility and obvious points for holders to sell into, since everyone can see the big unlock coming. Lower is better.

This measures the cliff before team and insider allocations start unlocking. A 12-month cliff holds insider selling back past the fragile launch window, when the market is thin and the price moves on little volume. A cliff under 6 months pushes that selling into the earliest, most fragile days of trading, when the float can least absorb it. A longer cliff is safer.

This measures how long team and insider allocations take to fully vest, start to finish. A 36-month or longer schedule spreads insider selling across years and signals a long-term commitment to the project. Anything under 24 months releases insider supply quickly and falls short of the industry standard, concentrating future selling into a shorter window. A longer schedule is safer.

This measures Year-1 growth of the circulating float, the circulating share at month 12 against the share at launch. New supply hitting a thin early float is what drives the price down hardest, so a smaller first-year expansion is far easier for the market to absorb. Lower is better, and the early-year bar is wide because some launch growth is unavoidable.

This measures Year-2 growth, the circulating share at month 24 against the share at month 12. By the second year the launch float has had time to deepen, so the market expects inflation well below Year 1 and the bar tightens accordingly. Lower is better, since continued heavy expansion keeps diluting holders past the launch window.

This measures Year-3 growth, the circulating share at month 36 against the share at month 24. By the third year the bulk of vesting should be behind the project, so the market expects inflation down in the low tens of percent. Lower is better, and inflation still running hot this late points to a long, heavy unlock tail.

This measures Year-4 growth, the circulating share at month 48 against the share at month 36. This is the strictest year, when a healthy schedule has nearly finished releasing supply and the market expects inflation in the low teens or below. Lower is better, and high fourth-year inflation means dilution is still landing years after launch.

This measures the circulating float at year 4 against the float at launch, so it is the total growth of trading supply across the whole four years, not any single year. A schedule can look mild year by year yet still multiply the launch float several times over by the end, and that full-span growth is the dilution early holders actually carry. Lower is better.

This measures how evenly emission is spread across Years 1 to 4 on a 0 to 1 scale, where 1 is perfectly even and 0 means it is all bunched into one year. Even release lets the market absorb supply steadily, while a single shock year of heavy unlocks creates a predictable price drop. More even is better, since it avoids any single overhang.

This adds up the part of every monthly supply shock that rises above the 15% threshold, the share too large for the market to absorb. A schedule rarely has just one of these spikes, so summing them across the whole schedule makes every spike count, instead of letting a project bury repeated large unlocks behind a single calm month. Even a small total is already a warning, because oversized unlocks do outsized damage. Lower is better.

This adds up the part of every monthly supply shock that stays below the 15% threshold, the share small enough for the market to absorb. No single one of these releases is big enough to notice, so the only way to see the pressure they build is to add them up across the whole schedule. Together they become a slow, grinding source of dilution that wears holders down, the routine drag no single month would flag. Lower is better.

This measures the biggest one-month jump in circulating supply, the worst single dilution event in the schedule. Large one-off shocks create predictable price drops as the market front-runs and absorbs them, and a single month that adds a large share of the float can overwhelm even a deep market. Lower is better.

This measures Year-1 growth of the unlocked supply, the unlocked share at month 12 against the share at launch. New supply becoming sell-eligible against a thin early base is what hangs hardest over the price, so a smaller first-year expansion is far easier for the market to absorb. Lower is better, and the early-year bar is wide because some launch growth is unavoidable.

This measures Year-2 growth, the unlocked share at month 24 against the share at month 12. By the second year the sell-eligible base has had time to widen, so the market expects inflation well below Year 1 and the bar tightens accordingly. Lower is better, since continued heavy expansion keeps adding to the overhang past the launch window.

This measures Year-3 growth, the unlocked share at month 36 against the share at month 24. By the third year the bulk of vesting should be behind the project, so the market expects inflation down in the low tens of percent. Lower is better, and inflation still running hot this late points to a long, heavy unlock tail.

This measures Year-4 growth, the unlocked share at month 48 against the share at month 36. This is the strictest year, when a healthy schedule has nearly finished releasing supply and the market expects inflation in the low teens or below. Lower is better, and high fourth-year inflation means dilution is still landing years after launch.

This measures the unlocked supply at year 4 against the unlocked float at launch, so it is the total growth of sell-eligible supply across the whole four years, not any single year. A schedule can look mild year by year yet still multiply the launch float several times over by the end, and that full-span growth is the dilution hanging over holders. Lower is better.

This measures how evenly unlocking is spread across Years 1 to 4 on a 0 to 1 scale, where 1 is perfectly even and 0 means it is all bunched into one year. Even release lets the market absorb sell-eligible supply steadily, while a single shock year of heavy unlocks creates a predictable wall of selling. More even is better, since it avoids any single overhang.

This adds up the part of every monthly unlocked shock that rises above the 15% threshold, the share of newly sell-eligible supply too large for the market to absorb. A schedule rarely has just one of these spikes, so summing them across the whole schedule makes every spike count, instead of letting a project bury repeated large unlocks behind a single calm month. Even a small total is already a warning, because the unlocked schedule is a leading indicator of selling pressure to come. Lower is better.

This adds up the part of every monthly unlocked shock that stays below the 15% threshold, the share of newly sell-eligible supply small enough for the market to absorb. No single one of these releases is big enough to notice, so the only way to see the pressure they build is to add them up across the whole schedule. Together they are a steady, grinding stream of supply turning sellable that wears holders down, the routine drag no single month would flag. Lower is better.

This measures the biggest one-month jump in unlocked supply, the worst single release in the schedule. Large one-off releases create predictable sell-pressure as the market front-runs and absorbs them, and a single month that frees a large share of the supply can overwhelm even a deep market. Lower is better.

This is the plainest test of whether the token is actually needed. It asks whether the protocol would still work if every use of the token were swapped for a neutral stablecoin like USDC. A token that cannot be swapped out holds important roles such as paying for transactions, securing the network, or serving as the unit everything is priced in. A token a stablecoin could cover instead exists mostly for fundraising or speculation rather than for any real function. The less replaceable the token, the stronger the result.

This measures, across everything the token is used for, the share that is mandatory, meaning you must hold or spend the token, against the share that is optional and merely nice to have. A mandatory use forces real demand, since holders cannot avoid acquiring the token to do the thing. An optional use creates no such floor, because anyone can skip it. When most of the utilities are mandatory, the token sits on a genuine demand floor, while a token whose utilities are mostly optional has a weak one. The larger the mandatory utility share, the stronger the result.

This separates demand that comes from actually using the product, where doing the thing requires the token, from demand that exists only because staking pays a yield. Staking is locking the token up in exchange for rewards, and reward-only demand is fragile, since it falls away the moment the yield drops or the rewards run out, and it is circular when those rewards are paid in the token itself. Usage demand is sturdier, because it lasts for as long as people keep using the product. The more demand rests on usage rather than rewards, the stronger the token.

This tests whether demand for the token is tied to the protocol getting bigger. Demand is structurally coupled when more usage mechanically needs more token, through transaction fees, a required stake, or tokens burned on each use, so demand rises with adoption on its own. It is discretionary when demand grows only if the team decides to route value back to the token, is not automatic, and decoupled when the protocol can grow with no effect on token demand at all. The more demand scales with usage by design, the stronger the result.

A token can create demand in four broad categories, Staking and Lockup, Access and Benefits, Exchange and Spending, and Governance. This counts how many of the four the token actually uses. A category counts only when the token has a real use there, one that is live, not propped up by subsidies, and genuinely creates demand. Covering several categories beats stacking everything into one, because each category is a separate source of demand, so a token leaning on a single category is exposed if that one use case fails or loses interest. The more of the four categories a token uses, the stronger the result.

This compares the utilities the token actually has against how common each utility is across its niche, the group of projects doing the same kind of thing. These standards are drawn from the classifier across our whole database, so the bar is what real peers tend to implement. One gap is fine, since no token matches every peer, but falling short on several utilities its niche have as standard is a real shortfall. Skipping a utility that is rare in the niche is never held against the token. The closer the token sits to its niche norm, the stronger the result.

This measures how much of what the project claims is live today rather than a roadmap promise, so credit goes to working utility instead of vaporware, features announced but never shipped. A utility that is openly held back to a later phase, with a clear reason and a stated date, is noted but not punished like a promise that simply never arrived. A utility that is merely planned with no detail earns nothing. The more of the claimed utility that is actually live, the stronger the result.

This measures where the rewards paid to stakers actually come from. Rewards minted as new supply are the weakest source, since every fresh token dilutes existing holders and adds selling pressure as stakers cash out their yield. A pre-set tokenomics pool is a little better but it is finite and not tied to any real income, and a blend of new minted/released supply and revenue funded buybacks also sits in the middle. Real yield, paid from genuine revenue such as stablecoin fees or revenue funded buybacks, is the strongest because it adds no net selling and grows with actual usage. The more rewards lean on real revenue rather than new supply, the stronger the result.

This scores staking against two marks of genuine commitment, the alternative to costless yield farming. Slashing, misbehavior or an early unstaking can confiscate part of the staked tokens, not just forfeit the rewards. Lockup, the stake is locked for a meaningful period before it can be taken back. Each mark puts the holder's own tokens at risk and creates real skin in the game, so the result counts how many hold. A claim that only pays rewards, with nothing locked and nothing at risk, meets neither. The more marks staking carries, the stronger the result.

This measures whether the token's staking and access gates keep rewarding a holder for accumulating more, or simply switch on once at a single minimum. A binary gate, where you cross one threshold and you are in, gives no reason to hold beyond that minimum, since someone holding far more gets the very same terms, which caps how much demand the gate can create. Tiered or curve-based gates, where larger holdings earn escalating discounts, bigger allocations, or deeper fee cuts, keep pulling holders to acquire more. The more continued accumulation the gates reward, the stronger the result.

A gated benefit is a perk you only get by holding or spending the token, such as a fee discount or access to a feature. This grades how valuable that benefit is against the money a holder has to tie up to earn it, since a benefit only builds real demand when it genuinely changes what a rational user would do. A strong benefit clearly tips that decision, like a large fee cut, a hard grant of capacity, or a feature the user actually needs. A weak benefit is cosmetic or too small to put a number on, so no one would buy the token for it, while a modest one is a real edge that only pays off for heavy users or at high volume. The more the benefit moves a real decision, the stronger the result.

The settlement medium is the asset a transaction is actually paid in. This tests how much an outsider has to touch the token to transact, since a token that every payment must route through earns fresh demand from outside the project each time it is used. A required or exclusive medium forces every transactor to acquire the token, or makes it the base pair every trade passes through, so demand rises with usage. A token that is preferred but bypassable, where users can pay in a stablecoin or have fees sponsored, makes that demand optional, and a freely substitutable token captures almost none of the transaction volume as real demand. The harder the token is to route around, the stronger the result.

This scores governance against three marks of real power. On-chain execution, a passed vote executes on-chain at the smart contract level, not an advisory signal the team can ignore. Key Parameters, votes control the parameters that matter, like the treasury, fees, emissions, or risk, not just cosmetic proposals. No insider override, no admin key, multisig, or council can veto or reverse a vote once it has passed. Each mark is a separate way governance can be real or hollow, so the result counts how many hold. The more marks it meets, the stronger the result.

This checks who is allowed to start a proposal, not just who gets to vote on it. Open access lets any holder above a reasonable stake put one forward, so the agenda is not set by insiders alone. A system that is open on paper but sets the bar at a large slice of supply is closed in practice, since almost no one can clear it. Delegate systems, where holders hand their votes to a meaningful set of active delegates, sit in between. When only the team or foundation can propose, holders are left to approve what insiders choose to put in front of them. The wider and more used the right to propose, the stronger the result.

This counts the protocol's independent revenue sources, where independent means a genuinely distinct user activity or counterparty, not several fee line items of the same underlying action. Each independent source matters to a holder because revenue that rides entirely on one activity, one product, or one customer is a single point of failure no matter how large the headline number. When several unrelated sources each carry weight, the income survives any one of them stalling, while a concentrated stream collapses the moment its single driver slows. The more independent the sources, the stronger the result.

Verifiability grades how much of the captured revenue an outsider can confirm for themselves, judged by where the fees settle. When the revenue-generating activity runs on-chain and every fee passes through the chain, the full figure is independently auditable by anyone at any time. When the revenue mixes on-chain and off-chain sources, part can be checked and part must be taken on trust, so it can be monitored but never fully confirmed. When it is purely off-chain, fiat settlement, off-chain commissions, or undisclosed treasury deals, the reported number cannot be verified at all and the holder is left trusting the team's word. The more of the revenue that settles on-chain, the stronger the result.

This looks at whether the activity that produces revenue repeats across a user's lifetime or fires only once. Recurring usage, a fee on every trade, a per-transaction service, or repeat consumption, means revenue compounds as a returning user base keeps paying. Periodic or seasonal usage still recurs but in bursts with quiet gaps between them. One-off usage, a single onboarding payment, a one-time mint, or a launch-only event, charges each user once and forces the protocol to keep acquiring new users just to hold revenue flat. The more the usage recurs per user, the stronger the result.

This measures the share of net protocol revenue that flows back to token holders against the share the protocol or treasury keeps. Value reaches holders through burns, buybacks, or a revenue share, whether it is paid in the token or in another asset, while whatever stays in the treasury reaches no holder. A holder earns nothing from the revenue a protocol retains, so the captured share is what gives each token a real claim on what the protocol makes. The larger the share of revenue that reaches holders, the stronger the result.

This measures whether the value reaching holders scales as protocol revenue grows, or stays flat while the protocol expands. When accrual is set as a share of revenue, more revenue mechanically sends more to holders, so they ride the protocol's growth. When it is a fixed or capped amount, revenue can climb while the holder share thins out and the upside accrues to the treasury instead. A holder wants their claim to widen as the protocol earns more, not to be left behind as it scales. The more the value reaching holders tracks revenue, the stronger the result.

This measures how widely value reaches holders, whether it lands on every holder just for holding or only on those who actively take part. Passive accrual, through a burn or a broad revenue share, reaches everyone who holds the token with no further action required. Active accrual reaches only holders who stake, validate, or delegate, so a holder who simply holds captures nothing, and value the treasury retains reaches no holder at all. A model that pays every holder spreads the benefit widest and asks the least of them. The more broadly accrual reaches holders, the stronger the result.

This measures where the accrual draws its funding from, across three sources. Revenue funding, paid or bought back from real protocol earnings, adds no new supply and can run indefinitely. Emissions from a pre-allocated pool are finite, they add circulating supply while they last and then run dry. Inflation, freshly minted tokens with no cap, is perpetual dilution dressed up as yield. Revenue funding rewards a holder with value the protocol actually earned, while supply-funded accrual quietly hands them back their own dilution. The more the accrual leans on revenue rather than new supply, the stronger the result.

This tests how certain it is that a declared accrual actually executes, because a burn that might happen is not a burn. The strongest form is hardcoded and automatic, a base-fee burn or an in-contract buyback that cannot be switched off and scales with activity on its own. Governance-controlled execution is real but contingent, it happens only when a vote says so. The weakest form is discretionary, where the team or a market maker decides whether to burn or hold, or where no one is named as the decision maker at all. The harder the accrual is to switch off, the stronger the result.

This catches accrual that is promised but not yet real. A protocol can list a burn or a buyback that is really a roadmap item for later, or one that switches on only if a condition is met first, like hitting a revenue target or reaching a new phase. Each mechanism that already runs today is demand a holder can count on now, while a deferred one is a hope with a date attached. The larger the share that is working now rather than waiting on a future date or a trigger, the stronger the result.

This checks whether the protocol pins down the concrete percentage that each destination receives, not merely that the mechanisms exist. A protocol can name a burn, a buyback, and a revenue share, yet never state what fraction of revenue each one takes or what the treasury keeps for itself. Until those splits are sized, a holder cannot tell how much value actually reaches the token, so the accrual cannot be measured or trusted. The more completely the splits are pinned to concrete percentages, the stronger the result.

This scans the whole documentation for internal contradictions, not just in the revenue splits but across every parameter and mechanism, the rates, the multipliers, the thresholds, and the mechanism descriptions. When the same thing is stated differently in two sections, or related numbers do not reconcile, the design cannot be trusted at face value and a holder cannot tell which version governs. The fewer the contradictions across the documentation, the stronger the result.

This grades how thoroughly the parameters, the splits, the thresholds, and the rates, were tested before launch, not which tool was used. The strongest case validates them across a broad range of revenue, demand, and shock scenarios, so they hold up under conditions the team did not hand-pick. A check against only a few scenarios is partial cover. Parameters set by design intent with no testing are the weakest and cap the score, because untested splits can break under lumpy revenue or a stress the design never modeled. The more broadly the parameters were tested before launch, the stronger the result.

This check applies only when the project's classification flags a buyback that holds the token or pairs it into liquidity, so it stays dormant for any project that runs no buy and hold program. When a protocol buys its token back, what happens to those tokens decides whether the float really shrinks. The strongest outcome durably removes them from circulation, by burning them, locking them in non-withdrawable protocol-owned liquidity, a time-lock, or a sink with no private key. A governance-mandated reserve is weaker, since a vote could release it. Discretionary reserves the team can sell again whenever it chooses are the weakest, because the float reduction is reversible and the held stack hangs over the market as latent sell pressure. The harder the bought-back tokens are to sell back, the stronger the result.

This check applies only when the project's classification flags an existing buyback, so it stays dormant for any project that runs none. It grades how consistently that buyback is actually carried out over time. A gradual program in the TWAP style, or any steady drip on a regular cadence, spreads the market impact thin, removes discretion over timing, and reads as a credible standing commitment rather than a one-off gesture. Lumpy, random, or ad-hoc buybacks with no set schedule hit the market unevenly and are easy to time for optics, firing when the team wants a headline and going quiet otherwise. A predictable drip earns more trust than a discretionary one, because holders can see it will keep running. The steadier and more rules-based the execution, the stronger the result.