← Back to projects
On-chain SQL Python Dune Analytics

RLC token: measuring protocol adoption through on-chain data

Can on-chain transfer data tell us whether a protocol is actually being used ? I explored that question using iExec's RLC token, a project I know from the inside.

Type

Personal project

Tools

Dune, Python, Plotly

Period

March 2025 to March 2026

Data

Ethereum mainnet


Why iExec and why RLC

I worked at iExec as a Product Manager Junior in 2024. During that internship I spent a lot of time on Dune Analytics doing market research on the Web3 ecosystem: DeFi, DePIN, RWA. That is where I first learned to query on-chain data with SQL.

iExec's RLC token has an interesting property: every computation, data access and app execution on the protocol requires RLC. This means transfer volume is not just a price speculation signal, it should in theory reflect real protocol usage. I wanted to test that hypothesis with actual data.


12 months of RLC activity

57K Total transfers over 12 months
356M RLC transferred in total
6.3M USD traded on DEX
Aug 2025 Peak month, 10K transfers
View full dashboard on Dune →

What the data shows

Monthly transfer activity

Two distinct spikes stand out: March and August 2025. Both are driven by different dynamics, which the zoomed views below explain.

DEX trading activity on Uniswap

April 2025 shows a high USD volume with few swaps, suggesting large institutional trades rather than organic retail activity. From August onward swap count increases steadily, consistent with broader market interest.

August 2025: Institutional spike

The peak concentrates on August 27-28. On August 28, RLC was listed on Guotai Junan, a major Hong Kong brokerage. This single event generated 2,932 transfers and 33M RLC in volume in one day, the highest daily activity in the entire period.

March 2025: Community activation

March 2025 shows elevated activity across several days with no single identifiable catalyst. The period overlaps with iExec's 2025 roadmap release and a community AMA on trusted AI. This illustrates a key challenge in on-chain analysis: correlation is easier to establish than causation.

Top 10 RLC receivers

The top wallet received 80M RLC across 4,620 transactions, consistent with an exchange or protocol contract. The wallet with a single 10M RLC transaction suggests an institutional one-off transfer.


How it was built

DuneSQL Python Pandas Plotly Dune API erc20_ethereum dex.trades

SQL queries were written directly on Dune to pull transfer data from erc20_ethereum.evt_Transfer and DEX activity from dex.trades, filtered on the RLC contract address. Data was fetched programmatically via the Dune API into Python, cleaned with Pandas and visualised with Plotly. Charts are exported as standalone HTML and embedded directly in this page.


Key takeaways

On-chain data is honest but noisy. A single institutional transfer can look like a surge in protocol adoption when it is really just a wallet reshuffle. Distinguishing signal from noise requires cross-referencing with external events.

The Guotai Junan listing on August 28 is one of the clearest examples of how an off-chain event produces an immediately measurable on-chain signature. The data does not lie, but it needs context to mean anything.

Having worked at iExec made this analysis richer. Knowing the protocol's architecture, the role of RLC in each computation, and the team's 2025 roadmap helped me interpret what the numbers were actually saying.