[Non-Constitutional] Transparency and Standardized Metrics for Orbit Chains on growthepie.xyz Allowing Data-Driven Decision Making

[tobsch]

[Non-Constitutional] Transparency and Standardized Metrics for Orbit Chains on growthepie.xyz Allowing Data-Driven Decision Making

Proposal

Brief background: This proposal has been created initially for the Arbitrum GovHack Brussels and was selected as one of the finalists. This version is an updated version, especially making sure that this proposal is understood as a start to have more usage analytics around Orbit chains, and that the future process of data collection and onboarding of chains is streamlined, and constantly iterated on to make the listings cheaper and more affordable/sustainable for upcoming (and potentially smaller, more niche) Orbit chains.

Abstract

We want to increase the transparency into the Orbit stack by aggregating important metrics on a chain level so that users, builders, and DAO members can make better data-driven decisions, including what revenue an Orbit chain generates. This allows to plan much better, for example around an Orbit liquidity plan or other discussions around Orbit chains and their current/future support, or even retroactive programs (through better historic and comparable data points).

There is high likelyhood that the DAO will want to run an incentive program for Orbit Chains in the future. By having the data ready to go, we can avoid having a short-term program proposal which bootstraps data gathering. A 2% (low estimate) productivity increase on $50 million would have been $1 million. This proposal can likely pay for itself!

It will be in the form of a dedicated Arbitrum Orbit Stack page on growthepie.xyz having 20 chains listed with standardized metrics for each chain as well as an aggregated fundamental metrics view for the whole stack.

Motivation

So far, there is no go-to place that provides an overview of Arbitrum Orbit. Standardized metrics allow a better idea of the performance and outlook of each chain being part of the Arbitrum Orbit. This results in greater transparency for all chains regarding their activity and how the usage of one chain compares to another.

Users can use this information to gather a quick overview of the Orbit ecosystem. Builders can use the different metrics to better track adoption, and DAO members can make more data-driven decisions on future initiatives (e.g., Orbit retrospective funding based on onchain contribution to the DAO).

Also, visibility for the overall Arbitrum Orbit stack is increased, and the whole Ethereum ecosystem can see what is going on in the Arbitrum ecosystem - where to be and interact, build, or allocate funding towards. Additionally, for DAO members and delegates, revenue and cost metrics provide future insight into how the Orbit approach should be developed to make it profitable for the DAO as well (includes more transparency into how much Orbit chains are paying to Arbitrum One).

This proposal is meant as a start for a long-term partnership to provide listing of Orbit chains long-term and on a recurring basis on growthepie. After the listing of the first setup and 20 initial chains, we would like to follow up with another AIP to enable listing of the 21st+ chain.

Rationale

growthepie is an entirely public goods funded analytics platform for Ethereum scaling solutions. The AIP aligns closely with Arbitrum community’s mission and guiding values. At growthepie, Ethereum alignment is at the core of our existence, ensuring that our efforts contribute to the broader Ethereum ecosystem.

We aim to enhance transparency, enabling data-driven decisions that are crucial for tracking outcomes and improving strategic decisions. All our code is open-source, allowing anyone to review and verify our metric logic. This openness fosters trust and ensures the integrity of our platform. We prioritize inclusivity by offering our platform and API completely free of charge, ensuring maximum accessibility for all users.

Our user-focused design philosophy centers on the needs of our diverse audience, making data easy to digest and actionable for all users. We uphold the values of neutrality and openness as we believe in the power of data to increase transparency and objectivity. All information on our platform is freely available, reinforcing our commitment to these principles.

Key Terms

• Revenue: Gas fees paid by chain users
• Onchain contributions to DAO: Orbit chain revenue contribution to the DAO (exact mechanics and details TBD)
• Fundamentals metrics: Transaction Count, Active Addresses, Throughput, Stablecoin Market Cap, TVL (Total Value Locked), Revenue (Fees paid by users), Costs (Settlement/DA), Profit (Revenue – costs), FDV, Market Cap, Transaction Costs

Specifications

growthepie has proven to be a platform for people wanting to understand the overall health, activity, and usage across the Ethereum scaling solutions ecosystem. Arbitrum One was listed as one of the first 5 chains from the beginning. Since then, Arbitrum members have used the platform to compare Arbitrum One with other Layer 2s, and we have provided reliable data and visualizations for Arbitrum’s community. Marketing has also used our charts to provide data-driven insights into the ecosystem and how Arbitrum One is growing, giving it more credibility.

Building on this, the Arbitrum Orbit should be given a dedicated place on growthepie with 20 chains listed in the beginning. This allows us to provide a mostly complete overview of the whole Arbitrum stack. This proposal is aimed at the start of a longer partnership. Hence, as part of this proposal, we will design and set up the initial pages and backend needed to onboard Orbit chains, aggregate fundamental metrics, and especially revenue metrics for each chain.

Which Orbit chains will be considered among the 20 chains as part of this proposal?
Every chain that is above $1 million TVL (Total Value Locked) at time of the listing decision, will be eligible to be listed as part of this proposal. Every chain that is below at the start of the listing process will not be considered. In case a chain falls below $1 million TVL in the future (after the listing decision had been made), the chain will stay on growthepie, however. The chain will not be delisted and available data will be presented as long as no official announcement by the chain’s team has been made that the chain will be shut down.

Each Orbit chain – as Arbitrum One and any other chain listed on growthepie – will have their own set of metrics (in the categories activity, value locked, economics, and convenience) and their own single chain page. Additionally, each chain will appear on each fundamental metrics page for which the specific metric is available.

Stack view for an aggregated Arbitrum Orbit view
On top of that, we will introduce a Stack view (new page) that provides a complete overview of the whole stack aggregated. This will be on a dedicated page which will enable users, builders, and DAO members solely interested in the Orbit to go to this place and learn more. See the prototype which includes:

• Leading chart that aggregates Layer 2s and/or Layer 3s together showing growth over time
• Table that lists all available Orbit chains on growthepie separated into Layer 2 and Layer 3
• Fundamental metrics section with the aggregated view of all chains combined in the Arbitrum Orbit

This set of views will make it possible to easily get an overview and see the growth of the whole Orbit ecosystem but also to compare certain Orbit chains against each other (or against other Layer 2s from other ecosystems).

Steps to Implement

Timeframe starting from the date of DAO proposal approval and weeks/months add up to each other. Total duration ~ 5 months. Chain listing time is also dependent on potential collaboration with an indexing provider and their speed of chain indexing.

Steps

Step	Resources (hrs)	Costs (in ARB)	Timeframes
Design and implement general platform UX adjustments on growthepie.xyz (more chains mean restructuring of UI elements and different filter logic)	Design - 20, Frontend - 40	$10,350.00 USD	3 weeks
Design and implementing frontend views: Orbit specific stack view, Single chain view (for each Orbit chain), Fundamentals view (to compare Orbit chains)	Design – 15, Frontend – 50	$11,200.00 USD	3 weeks
Streamline listing process	Frontend – 10, Backend - 15	$4,350.00 USD	2 weeks
Data infra setup (RPCs, database execution) and Frontend hosting costs	Backend – 30	$5,175.00 USD	2 weeks
Metric aggregation & Logic: Activity (Transaction Count, Active Addresses – daily and monthly), Value locked (Stablecoin Market Cap, Total Value Locked), Economics (Revenue, Costs, Onchain Profit), plus Market Cap and Fully Diluted Valuation where applicable, Convenience metrics: Transaction Costs (median 1-day aggregation), DAO Contributions (how much back to ARB)	Research - 20, Backend - 40	$10,350.00 USD	2 weeks
Listing first 10 Orbit chains (above $1 million TVL) and maintaining frontend + data pipeline for 1 year	Design – 10, Research – 40, Backend – 80, Frontend – 20, Infra - fixed	$84,375.00 USD	2 months
Listing 10 more Orbit chains (above $1 million TVL) and maintaining frontend + data pipeline for 1 year (20 total)	Design – 10, Research – 40, Backend – 80, Frontend – 20, Infra - fixed	$84,375.00 USD	2 months

Overall Cost

The following cost are denominated in USD. The exact ARB amount will be determined closer to the onchain proposal date to reflect the US dollar ask.

Fix setup costs ($41,425.00 USD):

• Design and implement general platform UX adjustments (more chains mean restructuring of UI elements and different filter logic)
• Designing and implementing frontend views: Orbit specific stack view, Single chain view (for each Orbit chain), Fundamentals view
• Data infra setup (RPCs, database execution)
• Metric aggregation & Logic: Activity (Transaction Count, Active Addresses – daily and monthly), Value locked (Stablecoin Market Cap, Total Value Locked), Economics (Revenue, Costs, Onchain Profit), plus Market Cap and Fully Diluted Valuation where applicable, Convenience metrics: Transaction Costs (median 1-day aggregation), DAO Contributions (how much back to ARB)

Initial chain listings (20 Orbit chains in total) and maintenance for 1 year ($168,750.00 USD):

• Integrate brand assets (icon, colors, etc.) for each chain
• Pull raw transaction data and store in our database
• Setup metric logic where applicable (e.g., bridge contracts for value locked, custom gas tokens, DA layer, settlement layer)
• Daily aggregation of all relevant datapoints
• Aggregated data / metrics available via API (for frontend and other data consumers)

Requirement: Arbitrum Foundation provides indexed data. If not, an additional cost for indexing will be added ($3,750.00 USD per chain)

Recurring costs also cover:

• Hosting frontend
• Maintenance backend / data pipelines
• Regular posts on social media about interesting developments (from data perspective)

Total Costs: $210,175.00 USD

Further information

This proposal has been initiated during the Arbitrum GovHack Brussels 2024 for:

• Track Number: 8
• Track Name: Orbit Stylus Infra

Initial Challenge Statement:

How might we help Arbitrum users, builders, and DAO members to make Orbit metrics available in an easy-to-digest manner so that DAO processes can be improved and builders can get an ecosystem overview resulting in more data-driven decisions and transparent processes?

growthepie team members:

Tobias Schreier, Matthias Seidl, plus growthepie team for implementation: Manish Gupta, Lorenz Lehmann, Ahoura Azarbin, Nader Bennour, Michael May, ETH Wave
More on our contributors page: growthepie Contributors

Team Lead:

Tobias Schreier (TG: @tobschcom)

Links:

Pitch Board

Proposal Video

growthepie platform

[cp0x]

I have a few questions:

1. Can you explain the difference between chains.
   How are they fundamentally different?
   Considering that this is a typical chain, all the settings for interaction, indicators, etc. will be identical.
   Therefore, I do not quite understand why $220,000 is required for this part of your work.
2. Why not make a universal module, where to set up a new Orbit chain, you only need to enter the number of this chain, and the rest will work by default. Or explain why this is impossible.

[mseidl]

Thanks for these questions @cp0x!

1. Luckily, through the nitro standard, chain schemas are well defined and static but there are also many variables:

• gas tokens can differ and separate price feeds for those are required
• host chains can differ (additional data like cost data needs to be pulled from these)
• locked Stablecoins depend on bridge contracts and used stablecoins
• we need access to reliable RPCs for each chain
• we need some additional metadata (which we can collect via a form) but would also need to verify it (i.e. infos mentioned above, icons, etc)
• we also have to verify data quality and correctness

Included in these mentioned costs are also infra costs for storage and compute.

2. Valid question! We definitely want to work our way towards an almost fully automated chain onboarding pipeline but we need to better understand the current Orbit chains first, need to see how much the points from above will actually vary across chains, and we don’t want to overpromise on anything (i.e. some could have really high tx volumes and we want to be prepared for that). I can definitely see a future though where chain id + sufficient metadata will allow us to create required database tables, queries, aggregations, pipeline schedules and other needed logic automatically with only a few manual steps involed. Ensuring data quality is a tough one without other similar open datasets but we can create heuristics.

Hope this answers your questions!

[tobsch]

A little update from our side:

We joined the Governance call yesterday and presented the project. We received some feedback from the DAO, updated and incorporated suggestions @DisruptionJoe (Thank you!) and would love to bring it to Snapshot in the following week.

Is there any more feedback or points to clarify?

[Tane]

Hi @tobsch, thanks for putting together the proposal.

While acknowledging the benefits of better data visualization and standardized metric viewing experience, we are struggling to see what the DAO can take advantage of them for by providing a rather large size cost. How do you see the deliverable that you provide will lead to more user and ecosystem growth so that the cost is justified?

Also, as cp0x pointed out, the implementation should cover an easy and fast integration of additional chains rather than asking for 2 months of a full dev team work.

We would suggest this project to go through a grant program with a very limited scope and validate the success of the product for Arbitrum One and a few Orbit chains.

[BlockworksResearch]

Blockworks Research will be ABSTAINING from a possible related vote on this proposal in the future due to a conflict of interest.

As we provide our own data-related services, voting on this proposal could introduce bias or violate our conflict of interest policies. However, we would like to contribute to the discussion by raising questions about the distribution of this product. Specifically, what is the current user base size for growthepie, and what are the expected growth projections for the client base accessing Arbitrum’s data?

Additionally, we would like to know what strategies are in place to drive user engagement and adoption of this platform. How will growthepie ensure that users, builders, and DAO members are aware of and actively using the new metrics provided? Effective user engagement is critical for maximizing the impact and value of the data.

[tobsch]

Hi @Tane and for your questions.

We already see that with other ecosystems, our data and easy-to-understand visualisations are used a lot in marketing (the Arbitrum Foundation itself uses it and confirms their growth and statistics with our charts). Also, we are used a lot by media sites like The Block, Cointelegraph and other blogs to make people aware of what’s happening and where something is happening.

Doing this for a breadth of Orbit chains only gives us more accurate data where users and builders can go. Most of these user groups are not data-experts, hence don’t use “standard data platforms” but need easy to digest and curated view. Here lies our strength that we don’t focus on everything and all chains in crypto, but on the Ethereum scaling ecosystem.

We usually work closely with the teams, coordinate, and also regularly post on our socials about the latest statistics (again, the Arbitrum Foundation can confirm), for which overhead we do not even account for here. We are not asking for a full dev team compensation, but for limited hours over 1 year’s time for infra, maintenance and updates. We curate the platform a lot, so it stays easy to digest, even with a flood of new chains coming up.

We have been in talks a lot with other grants opportunities, and evaluated especially also at the Arbitrum GovHack in Brussels, that the DAO is the best suited for this.

I hope this clarifies our proposal more.

[tobsch]

Hi @BlockworksResearch and thanks for your comments.

First, we actually do not see a conflict of interest with Blockworks, as you – like other sites – could be using our data to cross-reference and also include in reports, news or else. Our data is fully open-source and free to use through an API if preferred, this also applies with the ask for this proposal.

Although we have a small X/Farcaster follower base, other sites and podcasts like The Daily Gwei, Bankless and others reference and use our data to report. We are also organically growing (without purchasing artificially high follower bases) on our social media channels. Also, the Ethereum Foundation, for example, links to our platform as a go-to platform to learn about Layer 2s.

However, here are a couple of things that we have planned:

• we set up a Guild (guild.xyz) where we will be engaging more users to do their own research and learn how to combine insights and post it themselves, starting on our Farcaster channel and then of course on other their own socials
• we are moving to focus on a more “needs based” approach for users, so helping people to understand more where to go and what to use for their specific needs and wishes. This will allow us to also include more new users instead of the current typical crypto users. With this, in the future, more apps and insights about those will be discoverable, always keeping in mind the UX, which we hear from many is superior on our platform, and again, data is easy to digest without reading long reports or having to be a data expert.

I hope this answers the questions! And we would love you to reevaluate your position, as I believe you can greatly benefit from a coherent and third-party data set and standardised metrics around activity, economics, etc. around Orbit chains.

[BlockworksResearch]

As a team that builds dashboards for our Research platform subscriber base, in addition to building dashboards for other protocols, we fully understand the costs associated with this kind of endeavor.

From our perspective, this pricing is very reasonable and is below cost for a firm like Blockworks Research to execute on when you factor in salaries + backend infra + frontend work + misc expenses.

While costs per chain can get amortized at scale, the ~$10k/chain for 20 orbit chains is not an issue from our point of view.

[jameskbh]

I like the idea of preparing ourselves for reporting usage and other relevant metrics on the Orbit chains, specially after we are voting for Change Arbitrum Expansion Program to allow deployments of new Orbit chains on any blockchain.

This is a very detailed proposal and congratulate the proposers for it. However, I believe that this topic is best suited for an RFP process, as the DAO would have other proposals to compare, and a healthy competition between services providers. I would like to ask inputs from ADPC members like @Immutablelawyer, @Bernard or @sid_areta, if possible.

[swmartin]

I generally support this proposal. What are we doing here if there isn’t data transparency for average users? Some may argue redundancy with L2Beat’s incredible dashboards makes this spend unworthy, but having multiple sources to keep information in check is never a bad thing in my opinion. In terms of the price tag - I see it as fairly priced. This is as much a marketing play as it is a data/user transparency play. Some crypto natives go to L2Beat for data, others go to growthepie. Same goes with Coingecko vs Coinmarketcap, why would you not want to be listed on both?

My only qualm with this proposal, which is qouted below, is that it feels like this will be a race for the growthepie team to get up as many Orbit Chains as possible to conclude their contract. This, for me, is enough to turn this from an overwhelming yes to a definitive no.

@tobsch

Timeframe starting from the date of DAO proposal approval and weeks/months add up to each other. Total duration ~ 5 months. Chain listing time is also dependent on potential collaboration with an indexing provider and their speed of chain indexing.

My suggestion is to add KPIs that makes a given Orbit chain eligible to use one of our limited (20) slots on growthepie’s UI. This could prove difficult as tx count and gas fees paid will be easily gamed given the very cheap and high throughput nature of Orbits when deployed as L3s. I think the good old fashioned $1M of TVL mark would be sufficient for now, leaving the DAO the ability to change these parameters at any time. Here is a snapshot from L2Beat to give folks an idea of who would qualify. Do we really want Cheesechain taking up one of our spots (lol)?

[tobsch]

We really appreciate your input and the discussion. Also the general view that the pricing for this is reasonable.

I also want to highlight that we never see ourselves as competition to L2Beat (they do amazing work!), but rather an extension with a different set of metrics and also different user group for the most part.

Of course, it is also always healthy to compare different data platforms and sources to get a complete picture for the users. Also with additional metrics.

In regards to @swmartin’s point in terms of the 20 spots: The initial plan was to have a mostly complete picture of all chains in the long run, and the 20 chains in this proposal will be the start as it looks like there will be many more chains soon. The plan is to learn from launching the first 20 chains, because right now it is also difficult to foresee how much TPS and hence data there will be for the future chains. We also aim to improve the process of launching future chains, which hopefully lets us list chains (21+) more cheaply, this needs to be reevaluated along the way, however.

If the majority thinks it is better to tie this to KPI’s like TVL as mentioned above, we are happy to incorporate this of course.

[cerv1]

I’m a big fan of growthepie’s site and APIs, and believe they bring a lot of value to the ecosystem.

Requirement: Arbitrum Foundation provides indexed data. If not, an additional cost for indexing will be added ($3,750.00 USD per chain)

I too would love to see the Arbitrum Foundation offer certain foundational datasets as a public good for data teams. It doesn’t make sense for every analytics team to build their own indexing infra, especially as more and more chains come online. Best to get out ahead of this problem!

[davidgarcia]

David here from the Arbitrum Foundation. Foundation is finalizing an agreement with an indexing provider. This will allow us to distribute indexed data across the Arbitrum ecosystem and partners.

We expect having indexed data for 10 chains within one month. We will also implement a KPI to determine when Orbit data should be indexed. Based on this, we expect to achieve indexed data for 20 chains within the next 2-4 months, depending on the shown activity levels.

@cerv1

too would love to see the Arbitrum Foundation offer certain foundational datasets as a public good for data teams. It doesn’t make sense for every analytics team to build their own indexing infra, especially as more and more chains come online. Best to get out ahead of this problem!

[AlexLumley]

I really appreciate this proposal and especially how the team went to GovHack and continued progressing.

I have a few concerns.

1. Why is this Orbit chain data so important for the DAO? I don’t understand why it is important for the DAO to have this data? Will this cause more people to switch to the Arbitrum techstack? Will it improve our Orbit chain incentive programs? Simply having “data” is not enough, it needs to go towards our business objectives.

2. Shouldn’t this be an RFP process? There are multiple service providers who can provide this data (L2Beat, Dune Dashboards, etc) shouldn’t this be an RFP process? I would rather we spend $10K to determine the scope and validate the problem and then create an RFP process from there. An additional option is to create this in house.

I’ll note that these questions are more about a general process the DAO is currently using where we have proposals that are “self-serving” as opposed to scoping, defining and validating the requirements of the DAO and then creating an RFP process.

One option is to have someone from the community apply to a questbook for a small grant to lead this RFP process and then have GrowThePie apply to that.

See voting rationale here: [Non-Constitutional] Transparency and Standardized Metrics for Orbit Chains on growthepie.xyz Allowing Data-Driven Decision Making - #17 by AlexLumley

[Larva]

I support the proposal for implementing transparency and standardized metrics for Orbit chains. This initiative would provide clear, data-driven insights into the performance of Orbit chains, improving accountability and decision-making within the Arbitrum ecosystem. By offering standardized metrics on platforms like growthepie.xyz, developers and DAO members can make informed choices, optimizing resource allocation and governance. This transparency will also boost confidence among stakeholders, driving further adoption and growth of Orbit chains, ultimately benefiting the entire Arbitrum network.

[gauntlet]

The Arbitrum Foundation is actively investing in indexing and tooling for Orbit Chains. While this proposal represents an opportunity for the DAO to extend the presentation of that data, it raises the question of whether it is a service the Arbitrum DAO should subsidize for a single third-party provider. Is this a cost that the Arbitrum Foundation/DAO/Offchain Labs/Entropy should hire someone to implement and host on the Arbitrum site?

Orbit chain data is inevitably becoming as fractionalized as its liquidity. Suppose the DAO is to invest in a data platform. In that case, it’s worth considering a self-hosted approach rather than funding relevant onchain data and metrics across many frontends.

For generic chain data, history has shown that data providers are willing to build dashboards independently; whether DeFi Llama, TokenTerminal, L2Beat, or GrowThePie, data providers have historically fronted the development bill for traffic. Further, the proposal’s data deliverables do not align with the granularity of data any growth or incentives program would require (user segmentation, DeFi protocol data, pool-level data, etc.), which is arguably the ecosystem’s greatest current need.

As such, we are unsure of the return on investment here and plan to vote against the proposal in its current form.

[EzR3aL]

I voted NO because of this comment.
I do think having these kind of metrics need to be maintained by the Arbitrum Foundation. Also i do think the cost are heavy.

[maxlomu]

Gm @tobsch thanks for posting the proposal.

Could you clarify if the $10k/year per chain is expected to be a recurring cost or is it a one-off fee?

Also does the new indexing service the foundation is working on help in reducing the cost of adding new chains as data become more standardized?

[tobsch]

GM @maxlomu!

This is $10k/year, hence a recurring cost. The indexing provider definitely helps because we can be sure that the data is also available, however, each data platform also has to do their own aggregations, save data in their database, etc. which means infrastructure and maintenance cost. However, as stated above, we try to make this cheaper once the chains are better known (for example, depending on TPS), so I would assume this will become cheaper in the long run.