Protocol Revenue Report - H2-2022

In order to boost transparency into the revenue coming from Pocket’s protocol, we’re sharing our initial Protocol Revenue Report.

These reports will highlight protocol revenue specifically, as opposed to any other PNI sources of revenue. Protocol revenue represents the total RPC fees paid by demand-side users who request network throughput via the protocol. You can see more on the app economics in the docs.

Note: In the future, when the DAO institutes Application Burn Rate (ABR) (which is expected in the maturity phase), these demand-side fees that are used to acquire POKT will be app-staked and burned as requests are consumed. To maintain network throughput, the user will need to top up their stake accordingly. Currently, these customers pay via dilution vs. the future ABR.

We’ll keep it brief and simple with these reports, sharing:

  • Sources and dollar amounts of protocol revenue (dApp RPC sales (B2B), Gateway as a Service (B2B2C), Pocket Portal’s Pay As You Go (B2C), etc.)
  • Amount of POKT that is (1) currently app-staked and (2) set to be acquired & staked
  • Some key highlights and a quick look at the pipeline

This report excludes additional revenue sources of PNI, which include but are not limited to: new network acceleration fees, MEV (maximal extractable value), and node running rewards.

Given Pocket’s focus on protocol monetization not truly beginning until mid-Q3 of 2022, this initial report covers about 4-5 months of time, rather than a true half-year, but we’ve chosen the “H2” phrasing for simplicity’s sake. Going forward, we expect to share these types of reports on a quarterly basis, in the first month following each quarter’s end.

Protocol Revenue and POKT App Stakes:

With the housekeeping out of the way, first up are the different sources and dollar amounts of protocol revenue. With the exception of the Portal revenue, all the other revenue was closed by the Marketplace team via direct selling to companies, projects, and foundations.

Product Offering H2 2022 Protocol Revenue
dApp RPC Sales $427,000
Gateway as a Service (B2B2C) $49,000
Pocket Portal (Pay As You Go) $1,000
TOTAL $478,000

Next, the POKT staking/acquisition numbers as of 1/17/23, defined as:

  • POKT Currently App-Staked: This is POKT that has been acquired in the market by customers or has already been acquired by PNI (via the market maker for POKT) on behalf of customers.
  • POKT To Be Or Actively Being Acquired: This is the remaining POKT to be acquired on behalf of customers.
POKT Currently App-Staked 5,776,785 POKT (88%)
POKT To Be Or Actively Being Acquired 752,355 POKT (12%)
TOTAL APP STAKED DEMAND 6,529,141 POKT

Key Highlights and Notes:

Below are some of the noteworthy points of context to go along with this initial report.

  • The dApp RPC sales revenue included 9 separate deals, with 2 of those being multi-chain and the remaining spread across 6 different networks. These ranged from DeFi and wallets to new L1s, with the Status deal and Klaytn Foundation deal being especially nice wins for Pocket.
    • The average deal size for dApp RPC sales was ~$50,000.
  • Gateways (fka GaaS) revenue included 2 initial deals on the ETH network, an encouraging start for this new go-to-market motion and customer segment that has come into sharper focus recently.
  • Portal revenue from the Pay As You Go (PAYG) tier came from 90 unique paying users since October 2022. PAYG revenue bumped up in December, with that month accounting for about 50% of the Q4 PAYG total.
  • Note: with the exception of Portal PAYG revenue, you can follow the Marketplace team pipeline here.

Overall, these stats represent the first major steps towards monetizing the Pocket protocol, and reflect different adaptations in focus throughout 2022 - with one of those adaptations being an organizational-level shift towards exploring more direct sales. As we expand past the Portal PAYG revenue to more Direct Sales deals based around our Software as a Service pricing, and as we shift towards wholesale relays through our Gateways product (formerly known as “Gateway as a Service”), there is a large pipeline set up for conversion in 2023, with more than 90 open opportunities and a dozen plus partnerships.

You can see the full pipeline dashboard here.

How does this stack up relative to other Web3 infrastructure protocols?

According to web3 index, if we take the prior 90 day fees * 2 (for an H2-22 estimate) that would put the top revenue generator as STORJ. Despite POKT generating 25% more protocol fees over STORJ, our FDV remains 39% lower. It’s a similar, but more pronounced, picture with GRT (which is a more comparable data protocol). There are always many other factors to consider, but this is a high-level proxy for web3 market comparables.

Data reference: STORJ and GRT fees from web3 index, POKT and GRT FDV’s from CoinGecko, and STORJ FDV from CoinMarketCap.

Note: Web3 index doesn’t yet track Pocket’s app stake demand as fees, since these fees are not fully permissionless on-chain. This will be part of v1.

How can you help accelerate demand-side growth?

As with the rest of the Pocket ecosystem, there are a number of ways you can personally get involved in accelerating demand-side growth. One of the key ways to do so is to take advantage of PEP-46: dApp Referral Program, which pays out an ongoing 5% referral award for those that help bring dApps to Pocket’s paid service.

Pocket’s bounty programs are another way to bring additional talent into the ecosystem (developers) and grow awareness of Pocket’s protocol and mission (content writers) - both of which can translate into further demand-side growth.

That does it for this initial report - for the next one, we will cover Q1 2023 in Q2 and get on track with a quarterly cadence.

11 Likes

Hi Kevin,

Thanks a lot for putting this together, much awaited!

Please forgive my ignorance; if I may, I have a few questions:

  1. https://twitter.com/tokenterminal/status/1612138824708280320

Token Terminal differentiates between Fees and Revenue, and that seems to be the industry standard. Does this affect Pocket (the above data) in anyway?

  1. Now that we have started generating the data, would you know how and when we can get the Financial KPIs on this page for Pocket populated?

This will help in transparency and visibility. Token Terminal has become one of the go-to places for protocol data and research.

  1. I think FDV comparisons between POKT, STORJ and GRT are not apples to apples. Ideally FDV = Current Price X Max Supply. POKT and STORJ don’t have Max Supply, and therefore Total Supply is used as a workaround.

Btw, just curious about the relevance of FDV in this report?

Thanks a lot in advance and I hope that I can reach out in case I have more doubts.

Caesar

1 Like

Appreciate the review and comments, always open to collaborating and refining going forward.

Here are some responses:

  1. Based on my understanding of Token Terminal definitions, fees and revenue in this regard are the same for Pocket at the moment. Fees are what the demand-side (i.e. users) of a protocol pay (excluding gas fees) and revenue is fees generated for tokenholders. In this instance, the fees are what has been outlined above and all of those fees have gone towards buy-side POKT demand. Similar to how fees = revenue with ENS - Ethereum Name Service (ENS) - Index | Dashboard | Token Terminal. Definitions: Fees & Revenue - Token Terminal Docs. As mentioned, open to refinement to ensure absolute clarity.

  2. I strongly agree, and we have engaged with the Token Terminal team about interim workarounds until all of these demand-side fees/revenue are on-chain but they won’t make any exceptions. That is why I mentioned that Web3 Index doesn’t track demand side fees yet either, but will post V1. Ultimately, that is why we are publishing these for transparency in the interim. Ideally, this is all covered by these third-party data providers. CC: @jdaugherty

  3. I think it is always important to understand performance relative to market comps. Additionally, one of the key metrics that is outlined in the Tokenterminal link you shared is P/F and P/S which is ultimately what this is showing. Overall, I agree this is not exactly apples to apples and there are many other factors to consider when assessing a protocol but I thought this was a quick high-level comp summary.

1 Like

Hi Kevin,

Thank you for your prompt reply. I don’t have follow up questions at this moment.

Caesar

1 Like

Hello Team PNF: @b3n @Dermot @JackALaing @nelson @Ming

In the future, who would be responsible for tracking/reporting protocol revenue/financials, coordinating with players such as token terminal on protocol data, basically all things protocol data and financials? Will it continue to be PNI or is there any thought behind transitioning such routine protocol level tasks over to PNF, maybe after V1?

I have no opinion either way at this point. I am just curious about how scope and certain tasks will get shifted around given the new structure that is in place already.