Digging Into the World of On-Chain Data

How On-Chain Metrics and AI Reveal Hidden Adoption -– Eugene Tsai, Venture Data Analyst at Decasonic 

Introduction

The world of cryptocurrency is radically transparent. Every transaction is recorded on-chain, revealing fund flows, contract interactions, and the fees and revenues collected by protocols. Yet despite this openness, one important element remains obscured: identity. Unless individuals or teams publicly disclose their addresses, the true owners behind these transactions remain unknown.

This tension between complete data transparency and limited attribution is precisely where the opportunity emerges. When we combine on-chain analytics with AI tools, we gain the ability to identify patterns of adoption that are not visible through traditional methods. This allows us to detect early behavioral signals that rarely appear in standard dashboards or widely used metrics.

The purpose of this blog post is to explain how to use the power of on-chain data and AI to find hidden adoption metrics, uncover early signals through the noise, and enhance investment alpha.

First, I will discuss what blockchain and on-chain actually mean, what approaches we can take to derive adoption metrics, and how we can use those metrics to evaluate project valuation and drive investment decisions.

Then, I will look at how to use AI to increase the output of on-chain analysis and move toward adoption metrics that enhance investment alpha.

What is On-Chain Data and How does it relate to Adoption Metrics

In the world of blockchain, everything is publicly traceable, everything is relatable, and everything is disclosed. Because of these three characteristics, we can derive adoption metrics by tracking on-chain data.

The most reliable starting point is a project’s on-chain footprint, which often begins with the wallets and addresses linked to its operations. Once these entry points are identified, we can follow the progression of user behavior as it unfolds on-chain, moving from initial contact to repeated usage to deeper financial engagement. This creates a natural adoption journey: who shows up, what they do, how much value they commit, and how the ecosystem grows around them. We will also discuss scenarios where projects intentionally hide or fragment their on-chain footprint, since this can complicate the interpretation of these signals.

With this journey in mind, we can organize the project’s adoption metrics into five stages that reflect the progression from awareness to usage, capital commitment, retention, and ultimately ecosystem expansion.

Awareness and Entry

1. New wallets interacting with the project for the first time: The number of fresh wallets engaging with a protocol or application. A rising count signals top-of-funnel growth and expanding awareness.

2. Active addresses on the project: The number of unique addresses interacting with a project or protocol over a given time period. Rising active addresses indicate expanding user participation and are a core signal of real on-chain adoption.

Usage and Behavioral Signals

1. Transaction count and frequency: The number and cadence of actions users perform on-chain, for example, swaps, deposits, borrowing events, mints, or redemptions. Higher activity frequency suggests that users are returning and performing meaningful interactions rather than only testing the protocol.

2. Fees paid by users to the project: The total amount of fees users pay to use a project’s products or services on-chain, for example swap fees, borrowing fees, or protocol usage fees. Growing fees suggest real demand and utility, not just speculative activity.

3. Revenues generated by the project: The portion of fees or other income that accrues to the project, its treasury, or its token holders after incentives or rebates. Sustainable and recurring revenues are a strong indicator of product market fit and business viability.

Capital Commitment

1. TVL (Total Value Locked): The total value of assets deposited into a project’s smart contracts, for example liquidity pools, lending markets, staking contracts, and similar structures. It acts like the project’s on-chain treasury, signaling how much capital users trust the protocol with.

2. TVS (Total Value Staked): The total value of tokens locked by users to secure a network or protocol, for example proof of stake validators or staking programs. Higher TVS indicates stronger economic security and greater user commitment to the project’s long term health.

Retention, Value and Stickiness

1. Retention and repeat interactions: The share of users who continue interacting with the protocol after their first usage. High retention signals that users find ongoing value, while low retention may indicate friction, speculative-first behavior, or a lack of compelling use cases.

2. LTV (Lifetime Value): The cumulative value a user contributes over the lifetime of their engagement with a protocol, for example long-term fee generation, staking participation, or consistent capital contributions. LTV is closely tied to retention, since users who remain active for longer periods typically generate substantially more economic value for the project.

Ecosystem Growth and Expansion

1. Number of decentralized applications (dApps): The count of independent applications built on top of a project or its underlying infrastructure, for example rollups, layer one networks, and DeFi primitives. A larger and growing dApp ecosystem shows that developers are actively building, extending, and experimenting on that stack.

These metrics can also interact with one another. Fees, revenues, and the number of dApps are often tightly linked. When more developers build on top of a project, the number of dApps increases, the likelihood of higher fee generation rises, and project revenues tend to grow as more users are attracted. This can ultimately lead to network effects for the project. Examples include ecosystems like Base, Bsc, Ethereum, etc.

Take Base, for example, during the 01/01/2025–06/30/2025 timeframe:

-The growth rate for App Fees on Base was 83.18%.

-The growth rate for Active Users on Base was 85.3%.

Source: Dune

When Projects Intentionally Obscure their On-Chain Footprint

There are cases where projects don’t provide transparency into their on-chain wallets. This can be done by splitting one visible wallet into many smaller wallets or by using additional smart contracts that automatically swap fees collected from users into different assets immediately. When this happens, it becomes nearly impossible to trace the original wallet addresses and directly reconstruct the adoption metrics.

In that case, we can look at other types of adoption metrics for project assessment, along with their definitions and their relevance to adoption.

Community Metrics

These metrics estimate whether a project is gaining early traction. In our previous blog post on SocialFi network effects, we outlined how communities form around culture, collaboration, and co-ownership. Community metrics offer the earliest signals that these pillars are beginning to emerge. They include social media follower growth, post engagement rate, post impressions, and sector mindshare.

1. Social media follower growth: The rate of increase in followers across platforms over a defined period of time. A growing follower base indicates rising interest and awareness. Rapid growth suggests viral appeal and potential user onboarding.

2. Post engagement rate: The ratio of interactions such as likes, shares or reposts, and comments relative to total followers or total posts. High engagement reflects active community involvement and implies real user interest and loyalty rather than inflated follower counts.

3. Post impressions: The number of times a specific piece of content such as a tweet, blog post, or announcement is displayed on users’ screens, regardless of whether they click or engage. Rising impressions signal growing awareness and visibility and can serve as an early adoption proxy that often precedes increases in users, transactions, and on-chain activity.

4. Sector mindshare: The relative share of conversation volume or mentions a project captures within its broader sector or category, measurable via platforms like Kaito, Messari, or LunarCrush. Increasing mindshare indicates the project is dominating sector narratives, attracting attention from influencers and participants, and positioning itself as a leader in the space ahead of on-chain metrics.

Developer and Ecosystem Metrics

These metrics focus on technical activity and ecosystem expansion. They highlight the project’s appeal to builders and integrators. Examples include commit frequency, the number of open issues and pull requests, the number of forks and stars, and the number of contributors.

1. Commit and pull request frequency: The number of code commits or pull requests to the project’s repository over time. Regular commits signal ongoing development and maintenance, suggesting a healthy project that attracts contributors and implies growing adoption.

2. Number of open issues and pull requests: The count of unresolved issues and pending code contributions. Active issues and pull requests reflect community involvement in improving the project and are a sign of collaborative adoption among developers.

3. Number of forks and stars: The total number of repository forks, which are copies created for development, and stars, which are bookmarks that show interest. High counts indicate developer interest and experimentation and act as a proxy for project growth as more builders engage with the code.

4. Number of contributors: The count of unique individuals contributing code or documentation. A diverse contributor base signals broad developer adoption and often leads to more robust tools and integrations for the project.

Market and Financial Metrics

These metrics help infer investor and user interest and serve as proxies for economic adoption. Examples include trading volume trends, liquidity depth, and token holder distribution.

1. Trading volume trends: The average trading volume of the project’s token across exchanges over a specific period. Increasing volume suggests growing market interest and liquidity and indicates more users buying and selling as a proxy for adoption.

2. Liquidity depth: The amount of token liquidity available on exchanges, for example the depth of the order book. Deeper liquidity implies sustained trading activity and user confidence, reduces volatility, and encourages broader adoption.

3. Token holder distribution: The concentration of tokens among top holders, for example the percentage held by the top ten addresses. A more even distribution suggests decentralized ownership and a broader user base, which proxies for organic adoption rather than centralized control.

Partnership and Media Metrics

These metrics help evaluate validation and visibility, both of which drive awareness and credibility. Examples include partnership announcements, media mention frequency, project sentiment in the press, and influencer endorsements.

1. Partnership announcements: The number and quality of strategic alliances with other projects, companies, or institutions. Partnerships expand a project’s reach and utility, signal institutional adoption, and can attract more users.

2. Media mention frequency: The count of mentions in news articles, blogs, or podcasts during a given period. More frequent coverage typically reflects rising relevance and mindshare.

3. Project sentiment in the press: The tone of media coverage and analyst reports, whether positive or negative. Positive press builds trust and momentum, encouraging user and investor adoption.

How to use AI Tools to Dig into On-Chain Data

Next, we can look at how to use AI tools to dig into on-chain data to derive adoption metrics. We can split this into two scenarios. One where on-chain adoption metrics are directly visible and another where they are hidden or only partially visible.

1. When on-chain Adoption Metrics Are Disclosed

If adoption metrics are disclosed on-chain, we can think of AI tools as a data aggregation and insight layer that sits on top of existing analytics platforms.

For example, by connecting an LLM model like Claude to tools such as CoinGecko MCP, an analyst can ask in plain language for comparisons of TVL, fees, and active addresses across several protocols. Through these MCP integrations, the LLM can pull high-level adoption metrics in real time. From there, Claude can use its tool calls to calculate ratios such as fees over TVL, highlight sticky versus mercenary liquidity, detect inflection points around product launches, and automatically generate charts and commentary. This turns raw on-chain transparency into an investor-grade view of real adoption.

2. When Adoption Metrics are Hidden or Obscured

If adoption metrics are not disclosed or are intentionally obscured, we can shift the focus to AI as a way to triangulate reality from softer and indirect signals.

1. One angle is to run sentiment analysis over social channels, governance forums, and community chats to see whether real users and builders sound more positive or negative over time and whether that mood matches the narrative the project team is promoting.

2. A second angle is to feed the project whitepaper, public roadmap, code commits, and release notes into an AI model and ask whether the team is actually shipping what they promised on roughly the timeline they advertised on their social media channels. We can also test whether new features line up with meaningful user value rather than simply token games.

3. A third angle is to run background checks on the project team. This includes scanning founders’ and core contributors’ public histories, such as their past projects, code repositories, profiles on major platforms, old token launches, forum posts, and media coverage. The goal is to see whether they have a track record of actually shipping projects, how previous financing rounds ended, and whether any controversies or rug pulls appear in their history.

From On-Chain Data to Investment Insight

Digging into the world of on-chain data is ultimately about turning radical transparency into practical adoption insight. By combining direct on-chain metrics with off chain signals and layering AI on top of existing tools, investors can move beyond surface level narratives and toward a deeper understanding of who is really using a network, how value flows, and where durable adoption is emerging.

At Decasonic, we are focused on transforming these data driven insights into bold, informed investment decisions. If you are building in this space or want to explore how to apply on-chain and AI driven adoption metrics in your own work, feel free to reach out to us. We are always excited to connect with founders, investors, and analysts who want to push the frontier of Web3 x AI together.

The content of these blog posts is strictly for informational and educational purposes and is not intended as investment advice, or as a recommendation or solicitation to buy or sell any asset. Nothing herein should be considered legal or tax advice. You should consult your own professional advisor before making any financial decision. Decasonic makes no warranties regarding the accuracy, completeness, or reliability of the content in these blog posts. The opinions expressed are those of the authors and do not necessarily reflect the views of Decasonic. Decasonic disclaims liability for any errors or omissions in these blog posts and for any actions taken based on the information provided.