Operator economics

The identity every operator is solving

Monthly reward tokens multiplied by the token price, plus any usage fees, against amortised hardware, electricity, connectivity, site cost and the operator's own time. Positive means the node stays on. Negative for long enough means it does not, and no messaging campaign changes that arithmetic.

Multicoin Capital states the mature-state version as a constraint on the network rather than the operator: at maturity the supply side has to cover both capital investment and operating costs in dollar terms.² A network that cannot satisfy that from usage revenue is being held together by issuance.

The cost side, in dollars

Hardware costs vary by category more than founders expect. A vendor cost guide published by Nadcab, a blockchain development firm, puts Helium-class LoRaWAN gateways at roughly $400 to $600 depending on regional frequency band, a mid-range compute node near $5,000, storage nodes at $2,000 to $15,000, and high-end GPU nodes for AI or rendering work at $10,000 to $15,000 per accelerator.³ Those are vendor estimates rather than an audited survey, so price your own reference build before publishing any operator return figure.

Then add the recurring lines, which decide the outcome. Electricity at the operator's local tariff, which varies widely across the countries your operators actually live in. Connectivity. Site rent or rooftop access for anything needing height. Replacement cycles on consumer hardware. None of these are payable in your token.

The revenue side moves and the cost side does not

Helium's HIP-149 is the clearest published account of a reward rate compressing under an operator. HIP-53 had set a $0.50 per GB target earn rate. By the time HIP-149 was written, Nova had cut the payer rate to roughly $0.10 per GB to reflect commercial offload rates, and the proposal set deployer earnings as a band around that price: a floor at half of it, $0.05 per GB, and a cap at three times it, $0.30 per GB.¹

The same proposal shows why the floor was needed. About 13,870 HNT per day of Mobile deployer baseline against about 91,000 GB per day of rewardable volume is 0.152 HNT per GB, which at the roughly $0.27 HNT price it cites is about four cents, below the floor.¹ The proposal states the minimum is already binding, which is another way of saying the schedule alone no longer paid deployers enough.

A worked payback, and why it is uncomfortable

Take the floor rate and the volume as published. About 91,000 GB per day at $0.05 per GB is roughly $4,550 per day across the whole Mobile deployer set, about $1.66 million a year, before hardware, power or a rooftop.¹ Divide that by however many deployers share it. Across 10,000 deployers it is about $166 each per year. Across 100,000 it is about $17.

Now put a device against that. A gateway costing hundreds of dollars, and a purpose-built cellular radio costing materially more, against a per-node annual share in that range, is not a payback period in any ordinary sense.³ It is the calculation a prospective operator runs after their first month, and the one network dashboards rarely present, because an aggregate pool always looks larger than a share of it.

So publish reward pools per active node rather than in aggregate, and model them forward against the node count your growth plan implies. Supply growth is dilution to the operators you already have.

Why operators convert rewards, and the figure we will not give you

Operators convert tokens because their obligations are denominated in a currency the token cannot pay. Power bills, hardware finance and rent settle in fiat, and a device short of payback has a balance still to recover. Volatility adds a second reason: a fiat cost base against token revenue means the number of tokens needed to settle the same invoice rises exactly when the token is weakest.

A specific figure circulates claiming that DePIN operators sell 80 to 90 percent of their rewards. We looked for its source and there is none: no operator survey, no on-chain study, no named dataset. It propagates between glossaries and decks, and one trail leads back to unsourced glossary copy. We do not use it. Size operator conversion from your own reward distribution and your own operators' cost base, then check it against depth on the venues where they sell.