Research

State of POKT Network Q4 2023

Jan 31, 2024 ⋅  14 min read

Key Insights

  • Despite POKT issuance dropping 18% in Q4’23, supply-side revenue (POKT minting) increased 103% QoQ to $1.2 million, due to the POKT-USD price increase throughout Q4.
  • Simultaneously, protocol revenue decreased 23% QoQ to $75,000, in line with activity measured in serviced relays decreasing 45% in Q4’23. As of Q4’23, there are two Gateways servicing relays: Grove (71% of relays) and NodiesDLB (29% of relays).
  • Average POKT staking participation was 53% in Q4’23, down 7% QoQ. Service nodes and users (applications / developers) collectively accounted for ~95% of staked POKT.
  • The Shannon upgrade, scheduled for Q2’24, is set to enable decentralized demand-side access points for Gateways and application developers in 2024.
  • POKT aims to become a broad provider of decentralized data services and support any open-source service or public database. Specifically, its current focus is on supporting open-source LLMs hosted by node runners.

Primer

POKT Network (Pocket Network) is a decentralized node infrastructure protocol that enables users (applications/developers) to access blockchain data. Users submit API requests (relay requests) to service nodes via access points (Gateways) that build on top of the POKT Network. The protocol runs on top of POKT Network’s Proof-of-Stake blockchain, which uses a work token model to capture value. The key participants in the POKT Network are:

  • Service Nodes: Stake POKT to service relays to users (applications/developers); get compensated in POKT per relay serviced; receive 85% of total POKT rewarded for relays serviced.
  • Validator Nodes: Stake POKT to validate the PoS blockchain; confirm blocks that contain proofs of the relays serviced; retain 5% of total POKT rewarded for relays and 1% of transaction fees.
  • POKT DAO: A whitelisted group of POKTDAO tokenholders (on Gnosis Chain) exercising governance on Snapshot. Members undergo a proof-of-participation onboarding process before admission to the POKT DAO. The DAO retains 10% of the total POKT rewarded for relays and 99% of transaction fees.
  • Gateways: Access points for users that optimize for specific use cases and audiences. Gateways pay the POKT DAO a $0.00000085 fee per relay serviced. This fee is then burned by the POKT DAO.
  • Users (Applications and Developers): Submit relay requests through a Gateway.

As of Q4’23, there are two Gateways servicing relays: Grove (71% of relays) and NodiesDLB (29% of relays). Simultaneously, the POKT Network is working towards further decentralization of demand-side access points by onboarding more Gateways. The upcoming Shannon Upgrade will enable permissionless Gateways. POKT aims to become a broad provider of decentralized data services and support any open-source service or public database. Specifically, its current focus is on supporting open-source LLMs hosted by node runners.

Website / X (Twitter) / Discord

Key Metrics

Performance Analysis

The POKT Network is a two-sided marketplace connecting:

  • The supply, i.e., service nodes that enable users (applications/developers) to access blockchain data.
  • The demand, i.e., relay requests from users to service nodes via access points (Gateways).

Network Usage

The interaction between users and service nodes represents the fundamental utility of the POKT Network. Relay requests (calls) to service nodes are usually required by exchanges, wallets, and applications in order to, e.g., retrieve wallet balances, smart contract event logs, or interact with DeFi liquidity pools. Thus, the demand side can be derived by measuring relay requests serviced by the POKT Network.

Relay Requests

Relays serviced by the POKT Network decreased 45% in Q4’23 to approximately 58 billion (i.e., approximately 630 million relays serviced daily). This drop occurred in the context of Grove, POKT Network’s largest gateway, shutting off access to its public endpoints and reducing its free tier. As of Q4’23, there are two Gateways servicing relays: Grove (71% of relays) and NodiesDLB (29% of relays).

Relay Chains

Relay requests are routed to any of the supported blockchains (relay chains). POKT Network can support the routing of relay requests from any relay chain that uses the remote procedure call (RPC) standard.

The distribution of relays by chains shows that Ethereum accounted for 19 billion relays in Q4’23, down 27% QoQ from 26 billion relays in Q3’23. Simultaneously, Polygon accounted for over 13 billion relays in Q4’23, down 52% from 27 billion relays in Q3’23. The remaining relay requests throughout Q4’23 were serviced by Binance Smart Chain, Gnosis, DFK Chain, Avalanche, Fantom, Fuse, Klaytn, and Optimism.

Service Nodes

Staked service nodes on the POKT Network route relay requests from users (applications/developers) to relay chains, then forward the responses back to applications. Nodes that unstake POKT are referred to as unstaking nodes and must wait 21 days before their POKT is unstaked. Staked service nodes that double-sign a block or fail to sign a certain amount of blocks within a set time period are called jailed nodes.

While staked nodes declined over the past four quarters to nearly 17,000 in Q4’23, down 9% QoQ, jailed nodes followed a similar downtrend over the past four quarters. Notably, the percentage of jailed nodes continued to decrease in Q4’23, accounting for 0.15% of all service nodes on the POKT Network.

Validators

Validators are service nodes that also propose and validate blocks on POKT Network’s blockchain. Hence, all validators are service nodes, but not all service nodes are validators. The top 1,000 service nodes in terms of POKT staked are eligible to become validators. Validators earn 1% of transaction fees and 5% of POKT inflation when they are a proposer for a given block.

The number of average daily unique block proposers remained relatively stable at 94 in Q4’23, down from 96 throughout the first three quarters of 2023. This stability shows that POKT is widely distributed among the validator set. Given that block rewards are only sent to the proposers, the mechanism distributing POKT is even further decentralizing the holder base as more unique validators continue to propose new blocks.

Staking Participation

To capture value, the POKT Network uses a work token model, also known as a utility token model. POKT Network has a total supply of 1.6 billion POKT, inflating at <5% currently with further reductions targeted.

Newly issued POKT is distributed to network participants as follows:

  • 85% to service nodes,
  • 5% to block proposers (validators), and
  • 10% to the POKT DAO, instead of paying for each RPC relay.

Historically, the POKT protocol would issue new POKT for each relay, directly paying nodes through minting. With the introduction of the Parameter Update Proposal-32 (PUP-32) in June 2023, the POKT issuance was set to target a maximum of 220,000 POKT per day. Service nodes earn POKT rewards for each relay serviced. The amount of rewards is a factor of how many nodes they have staked in the network and how often gateways select them based on their past performance (based on e.g., latency, correctness, and uptime).

Because relays are paid through the issuance of POKT (i.e., by issuing and distributing rewards to the key ecosystem participants), staking participation reflects the percentage of POKT utilized for services on POKT Network. An average of 53% of the POKT tokens were staked throughout the POKT Network in Q4’23, down from 60% in Q3’23.

Nearly 900 million POKT was staked in Q4’23, down 11% QoQ. In Q4’23, service nodes accounted for 94.1% (~825 million POKT) of the staked POKT, whereas 0.5% (~4 million POKT) was staked by users (applications/developers). The remaining ~48 million POKT is locked by the protocol until the DAO disburses it.

Of note, the Shannon Upgrade (to be deployed to testnet in Q1 2024 and scheduled for late Q2 2024) will enable decentralized demand-side access points for Gateways and applications/developers. Doing so will give users a more direct way to stake on POKT Network and access RPC service by service nodes.

Revenue

The revenue generated by the POKT Network consists of:

  • Supply-side revenue: generated from POKT issuance, where 85% is distributed to service nodes in proportion to the number of relays that they service, and 5% is distributed to block proposers (validators). What’s not recorded as revenue is the remaining 10% that goes to POKT DAO.
  • Protocol revenue: accounts for Gateway fees burned by the POKT DAO alongside accrued fees from transactions. After PIP-29, a Gateway fee of $0.00000085 per relay serviced was introduced where Gateway operators send Gateway fees to POKT DAO to then be burned.
  • Transaction fee revenue: accounts for 1% of the transaction fees rewarded to block proposers, while the remaining 99% goes to POKT DAO.

In Q4’23, approximately 94% of the revenue represented supply-side revenue from POKT issuance, whereas nearly 6% of the revenue accounted for protocol revenue from token burns and transaction fees.

Supply-Side Revenue

Despite POKT issuance dropping 18% QoQ, supply-side revenue from POKT minting increased 103% to $1.2 million in Q4’23 due to the POKT-USD price increase.

The successful introduction of the Parameter Update Proposal-32 (PUP-32) set the minting to roughly 220,000 POKT per day. In Q4’23, this translated to an 18% QoQ decrease in POKT issuance. Despite the lower POKT issuance, the POKT-USD price increase led to the supply-side revenue (from POKT minting) increasing 103% QoQ in USD to $1.2 million. This amount represented over 94% of the revenue by the POKT Network in Q4’23. The rest of the revenue consisted of protocol revenue.

Protocol and Transaction Fee Revenue

Protocol revenue accounts for Gateway fees burned by the POKT DAO alongside accrued fees from transactions. Protocol revenue started to grow significantly as of May 2023, after the successful introduction of POKT burns as part of the PIP-29. Burns consist of the $0.00000085 Gateway fee per relay that is sent to POKT DAO to then be burned weekly. The rationale for including burns in the protocol revenue calculations is that burning POKT accrues value to all POKT tokenholders. In Q4’23, protocol revenue dropped 23% QoQ from $97,000 in Q3’23 to $75,000 in Q4’23. Protocol revenue accounted for nearly 6% of the total revenue of the POKT Network in Q4’23.

Transaction fee revenue accounts for 1% of the transaction fees rewarded to block proposers (validators), while 99% of the transaction fees are sent to POKT DAO. Transaction fees on POKT Network function similarly to gas fees on Ethereum. In Q4’23, total transaction fees amounted to nearly 347,000 POKT, up 20% QoQ. Conversely, transaction fee revenue accounted for nearly 3,470 POKT. Overall, revenue from transaction fees only plays a small part in total revenue.

Qualitative Analysis

Releases

Non-Custodial Node Running (PIP 32)

This successful proposal introduced a reward-sharing feature that enables a staked node to distribute relay rewards across multiple wallets. Node operators can leverage this functionality to offer a reward-sharing pricing model. This can be achieved by introducing a new field "delegators" in a staked node. Details about the working implementation are on GitHub.

Parameter Changes (PIP 33 and PIP 34)

This successful proposal introduced a new parameter that allows the calculation of custom multipliers of relay rewards. The purpose of this proposal is to lay the foundations for onboarding new data types with higher costs per request, such as serving inference requests for open-source AI LLMs like Llama 2 and Stable Diffusion. Details about the working implementation are on GitHub. Simultaneously, this proposal seeks to accurately determine eligibility for relays and rewards.

Transferring a Staked App to a New Account (PIP 35)

This successful proposal facilitates the secure transfer of a staked app to a new account without needing to unstake the app first. The current unstaking process poses a risk that the app slot could be stolen during the transaction process due to the non-guaranteed order of transactions included in a block by validators. To mitigate this risk, the proposal aims to allow the direct transfer of staked apps, thereby preserving the security of the app slot.

Governance

Citizenship Specification (PGOV 4)

This initial proposal defines that Citizenship is attained by completing two quests:

  • The DNA quest - A credential attained by understanding the purpose, vision, and values of the POKT Network.
  • The DAO quest - A credential attained by understanding the governance, grants, impact, and operations of the POKT DAO.

The Citizenship credential will allow a person to become a DAO Voter in the Citizen House (10% of total voting power in the DAO).

Builder Specification (PGOV 5)

This initial proposal defines a set of eight different types of Builder credentials that translate to governance power in the DAO. The eight credentials are as follows: Protocol Builder, Priority Builder, Socket Builder, Proposal Builder, Bounty Hunter, Thought Leader, DAO Scholar, and OG Governor.

Staker Specification (PGOV 6)

This initial proposal aims to introduce three productive uses of POKT with governance power within a single Staker House:

  • Supply-side path (40% of the power in the Staker House): a node runner or a validator staking POKT to serve relays or secure the network.
  • Demand-side path (30% of the power in the Staker House): a Gateway transferring POKT to the DAO - via the Gateway Operator Fee (burn) - to pay for access to the POKT Network protocol on behalf of its end-user developer customer base.
  • Liquidity provider path (30% of the power in the Staker House): a liquidity provider staking wPOKT/ETH (and/or future approved pools) to provide liquidity for the network.

Voting Specification (PGOV 7)

This initial proposal attempts to specify how the voting results of each of the houses (the Citizens - PGOV 4, the Builders - PGOV 5, and the Stakers - PGOV 6) are aggregated. This specification ties these together into a Voting Module, which brings everyone together to vote on proposals and aggregates the results of each house's votes. The proposed voting weights are: Citizens - 10% of total; Builders - 45% of total; Stakers - 45% of total.

Proof of Personhood Specification and Regulating Sockets

This preliminary discussion aims to adopt Gitcoin Passport as POKT DAO's technical solution to Proof of Personhood to ensure that each POKT DAO voter is a real person. Simultaneously, this pre-proposal specifies requirements to be eligible for opening a Socket.

POKT Foundation Transparency Report & Accounts

POKT Foundation published its Quarterly Transparency Report and the Accounts for Q3 2023.

wPOKT V2 Timeline Update

This proposal seeks to extend the current wPOKT LP incentive program from the end date of Jan. 10, 2024, to Apr. 9, 2024, and to launch an incentivized liquidity pool for wPOKT on Arbitrum.

Ecosystem

Recovery of User Funds (PEP 67)

This proposal seeks to elucidate the developments following the closure of staking operations by a major staking provider for the POKT ecosystem – Nodes4All – on October 2, 2023. The primary objectives include designating a representative to engage in negotiations with Nodes4All for the return of investor funds.

Community Support

PEP 65 granted POKTscan $20,000 in POKT every month until the launch of the Shannon upgrade, currently scheduled for late Q2 in 2024. The funds will be used for POKTscan to continue its work supporting the Morse protocol (POKT Network’s first release, formerly known as V0). PEP 66 granted $4,000 monthly in POKT paid at the monthly trailing 30-day average price to fund the cost of having two lead community moderators.

Ecosystem Update and Vision

This forum post featured the key ecosystem developments in 2023 and shared the updated Q2’24 objectives, including community grants and ecosystem rewards, e.g., open RFPs). In a recent post, Grove’s CEO highlighted that POKT aims to become a broad provider of decentralized data services and support any open-source service or public database. Its current focus is on supporting open-source LLMs hosted by node runners. As POKT expands beyond RPC, it aims to become a broad middleware provider.

Closing Summary

Despite POKT issuance dropping 18% in Q4’23, supply-side revenue (POKT minting) increased 103% QoQ to $1.2 million in USD due to POKT-USD price increase throughout Q4. Simultaneously, protocol revenue from token burns and transaction fees decreased 23% QoQ to $75,000 in Q4’23, in line with activity measured in relays serviced decreasing 45% in Q4’23. Average POKT staking participation was 53% in Q4’23, down 7% QoQ.

POKT Network’s Shannon protocol upgrade aims to enable decentralized demand-side access points for gateways and application developers in 2024. POKT seeks to become a broad provider of decentralized data services and support any open-source service or public database.

Let us know what you loved about the report, what may be missing, or share any other feedback by filling out this short form. All responses are subject to our Privacy Policy and Terms of Service.

This report was commissioned by Pocket Network Foundation. All content was produced independently by the author(s) and does not necessarily reflect the opinions of Messari, Inc. or the organization that requested the report. The commissioning organization does not influence editorial decision or content. Author(s) may hold cryptocurrencies named in this report. This report is meant for informational purposes only. It is not meant to serve as investment advice. You should conduct your own research, and consult an independent financial, tax, or legal advisor before making any investment decisions. Past performance of any asset is not indicative of future results. Please see our Terms of Service for more information.

No part of this report may be (a) copied, photocopied, duplicated in any form by any means or (b) redistributed without the prior written consent of Messari®.

Upgrade to Messari Pro

Gain an edge over the market with professional grade tools, data and research.

Already a member? Sign in

Upgrade to Messari Pro

Gain an edge over the market with professional grade tools, data and research.

Already a member? Sign in

Mihai is director of research at Messari. Mihai and his team cover base layers, mid-layer infrastructure, DeFi, and consumer. Prior to joining Messari, Mihai was a tech entrepreneur and worked in AI at UBS and Swiss Re. His background is in computer science and math. Mihai holds a PhD in information systems from ETH Zurich, Switzerland

Onchain Jíbaro. Background: Photography, Quantitative Banking, & Manual Labor.

Mentioned in this report

Read more

Research Reports

Read more

Based on your watchlists

Create a new watchlist
Read more

Research Reports

Read more

Based on your watchlists

Create a new watchlist

About the authors

Mihai is director of research at Messari. Mihai and his team cover base layers, mid-layer infrastructure, DeFi, and consumer. Prior to joining Messari, Mihai was a tech entrepreneur and worked in AI at UBS and Swiss Re. His background is in computer science and math. Mihai holds a PhD in information systems from ETH Zurich, Switzerland

Onchain Jíbaro. Background: Photography, Quantitative Banking, & Manual Labor.

Mentioned in this report