The fifteen tasks
Here is what the fleet goes after, one by one. For each: how the reward is paid, and what KIND of figure exists. No expected amount: we have none measured, and an invented average would be worth less than nothing. What we can say is verifiable instead: what sort of figure exists for this task, and why.
Days to weeks
What is earned with no stake, or almost none
Seven tasks out of fifteen require no capital at risk. That is the answer for anyone unwilling to expose funds: the work is done, the reward is announced beforehand, and nothing is locked up.
On these seven tasks no capital is locked up: what you commit is time. Nothing guarantees an award, though: the quantity available is often limited, and work delivered may not be retained.
Paid missions
An issuer publishes a task and its amount before the work: annotation, interface testing, data verification. Payment follows validation.
An amount published by the issuer, known in advance. How many missions will be available is not.
Paid surveys
An answer supplied against an announced amount, paid by the survey issuer after a consistency check.
An amount published before you answer. The number of open surveys changes without notice.
On-chain quests
A sequence of actions on a protocol (a swap, a deposit, a vote) rewarded in tokens or in points.
An amount published when the protocol announces one. Often nothing is announced and the reward stays discretionary.
Learn-and-earn
A module followed, a quiz passed, a distribution announced by the issuer for those who arrive first.
A published amount, but a strictly limited quantity: arriving after it runs out pays nothing.
Bounties
A posted reward for a precise outcome: a flaw reported, a contribution merged, a piece of work delivered.
A published amount. The award rests on human judgement and is never guaranteed.
In-game quests
Progress in an on-chain game, rewarded in the game’s own assets.
Sometimes a published amount, but denominated in the game’s asset: its resale value is unknown and may be nil.
Incentivised testnets
Using a test network, with a possible reward when the real network launches.
No figure published in advance, in the vast majority of cases. It is a bet on a stated intention.
Continuous
What is earned when capital is at work
Six tasks put capital to work. This is where the most readable figures exist, and this is also where the total-loss sentence applies in full.
A stake is committed, so a risk is too: price volatility, illiquidity, lock-up, exit penalty, and partial confiscation on restaking. You can lose everything you commit.
Staking
Locking a token to secure a network, paid by the protocol itself under a published rule.
A RATE per unit deployed exists, published by the protocol. It moves, and the capital stays exposed to the token price.
Restaking
Reusing an already-staked deposit to secure further services, for additional pay.
No stable rate. Pay depends on the services chosen, and each one adds its own confiscation layer.
Lending
Lending an asset on a money market, against interest paid by borrowers.
A RATE exists, published continuously. It moves with every change in utilisation, sometimes hour by hour.
Yield farming
Supplying liquidity to a market, against a share of trading fees and protocol incentives.
No stable rate. And impermanent loss can exceed what the fees pay: ending up worse off than doing nothing is a common outcome.
Liquidity incentives
Temporary programmes that reward the depth of an order book or a reserve.
No stable rate. A programme ends without notice, and yesterday’s posted rate says nothing about tomorrow.
Arbitrage
Exploiting a price gap between two venues, buying at one and selling at the other.
No rate at all. An OBSERVED gap is not a capturable gap: fees, latency and competing bots often close it before execution.
Deferred and uncertain
What might come later, or never
Two tasks do not settle on their own horizon. They build an eligibility, and an eligibility is not a claim.
Published criteria count DISTINCT months of activity and retention. Many distributions never arrive, and those that do reward some and not others. This page says so before you find out.
Distribution eligibility
Sustained activity on a protocol, in the hope of meeting the criteria of a future distribution.
A DISPERSION, not a rate. Observed allocations run from zero to multiples of the median, and a large share of candidates is filtered out before distribution.
Points programmes
A protocol-internal unit, accumulated through use, that might one day confer a distribution.
No figure, by construction. A point is worth nothing until something is converted, and nothing obliges a protocol to convert.
Duration
What time actually builds
What accumulates is not a yield: it is an eligibility. A wallet active three months running is not in the same category as one active three days, because published criteria count distinct months.
- Two tasks out of fifteen carry that effect: distribution eligibility and points programmes. The other thirteen settle on their own horizon and gain nothing from lasting.
- That is why the subscription is prepaid by the quarter or by the year. We promise not one extra cent after three months: we say that continued activity is what the criteria measure.
- An eligibility is not a claim. Nobody owes you anything, and a protocol may never distribute.
What reduces the result
Everything below is subtracted from what you get, on every task. These lines are not stylistic hedges: they are the items that make a gross amount different from a net one.
- Network fees, paid on every operation, including the ones that fail.
- Platform and protocol fees, taken on the payout or on the withdrawal.
- The gap between the quoted price and the price obtained, which widens with order size and thin markets.
- Conversion costs, on every move from one asset to another.
- Illiquidity: an asset received is not an asset sellable, and a locked asset does not sell at all.
- Tax, which remains yours and depends on your country of residence.
- Total loss, which remains possible on any task that commits capital.
What kind of figure exists, task by task
This is the most important table on the page. It states what SORT of figure exists for each task, and it reads in thirty seconds. A rate per unit deployed, the one everybody looks for, exists for two tasks out of fifteen.
A rate per unit deployed
Published by the protocol, variable, and computable in advance on a given amount. This is the only category where "how much on a thousand euros" has an answer, and the answer moves.
Staking (08), lending (10). Two tasks out of fifteen.
An amount published before the work
The issuer states what it pays before you start. What you do not know is how often the opportunity will come round: the quantity available is not published, and it runs out.
Missions (01), surveys (02), quests (03), learn-and-earn (04), bounties (05), games (06). Six tasks.
No stable rate
A figure may show at a given instant; it predicts nothing. The programme ends, the gap closes, the composition of the market changes. A rate read yesterday is stale information.
Restaking (09), yield farming (11), liquidity incentives (12), arbitrage (13), points programmes (15). Five tasks.
A dispersion, not a rate
Observed allocations spread from zero to multiples of the median. An average computed on that describes a participant who does not exist, and quoting it would mislead.
Distribution eligibility (14). One task.
Nothing published in advance
No amount, no rate, no enforceable rule. The issuer states an intention and reserves everything else.
Incentivised testnets (07). One task.
What the fleet does with these fifteen tasks
It sweeps them continuously, investigates what passes its checks, and refuses the rest in writing. You can watch it work before opening an account.