Alex Svanevik: Nansen excels in blockchain data attribution, the importance of quality assurance in labeling, and how data harmonization drives insights

Key takeaways

• Transforming raw blockchain transactions into insights involves multiple data processing layers.
• Nansen excels in the attribution layer, crucial for understanding on-chain flows.
• The attribution layer for labeling addresses includes both algorithmic and human efforts.
• Quality assurance is critical in maintaining trust in data labeling processes.
• Every label in the database is backed by compiled evidence to ensure accuracy.
• Public information on blockchains can be used to label addresses, with removal rights for individuals.
• Blockchains are inherently public and transparent, making information immutable.
• Labeling blockchain addresses involves studying behaviors and deterministic smart contract events.
• Data harmonization across different chains is essential for deriving meaningful insights.
• Address labeling has evolved to include agentic and algorithmic methods.
• The integrity of blockchain data labeling relies heavily on quality assurance.
• Blockchain technology’s transparency and immutability have significant privacy implications.
• Nansen’s labeling infrastructure provides real-time insights for investors and traders.
• Understanding blockchain transactions requires a combination of inference and flow analysis.
• The complexity of on-chain analytics highlights the need for advanced data processing techniques.

Guest intro

Alex Svanevik is the CEO and Co-founder of Nansen, a blockchain analytics platform pioneering AI-driven agentic trading. He previously served as Chief Data Scientist at CoinFi, where he built a crypto database and back-tested trading signals. Svanevik holds an MSc in Artificial Intelligence from the University of Edinburgh.

The process of transforming blockchain data

• Raw blockchain data must be extracted into a more efficient storage and compute layer.

• The most basic thing you have to solve is to get the raw on chain data into a more convenient storage and compute layer

  — Alex Svanevik

• Data harmonization across different chains is crucial for effective analytics.

• You need to kind of harmonize all that data

  — Alex Svanevik

• The attribution layer involves labeling addresses to provide context to blockchain transactions.

• The third part… is the attribution layer where you label the addresses

  — Alex Svanevik

• Nansen excels in the attribution layer, critical for understanding on-chain flows.

• We probably are the best in the world at the attribution layer

  — Alex Svanevik

• Transforming raw transactions into insights requires multiple data processing layers.

• You need to get the raw entredata into something that is more convenient for running analytical queries

  — Alex Svanevik

• The process involves both algorithmic and human efforts in address labeling.

• We’ve kind of evolved our approach to labeling addresses over the years

  — Alex Svanevik

The importance of data harmonization and attribution

• Harmonizing data across chains is essential for deriving meaningful insights.

• The second layer is you have to kind of harmonize the data cross chains

  — Alex Svanevik

• Address labeling has evolved to include both agentic and algorithmic methods.

• We also do a lot of algorithmic work that is not agentic but is still super important

  — Alex Svanevik

• The attribution layer is crucial for interpreting blockchain transactions.

• The third part which we i would say probably are the best in the world at is the attribution layer

  — Alex Svanevik

• Nansen’s expertise in attribution helps investors make informed decisions.
• The complexity of on-chain analytics highlights the need for advanced data processing techniques.
• Understanding blockchain transactions requires a combination of inference and flow analysis.

• You can literally send some money to binance and see where the flows go

  — Alex Svanevik

• Labeling wallets involves studying transaction behaviors and deterministic events.

Labeling blockchain addresses

• The process involves studying behaviors and deterministic smart contract events.

• You have to study the behaviors of different entities

  — Alex Svanevik

• Smart contracts provide deterministic events for labeling.

• Some behaviors are deterministic because they’re smart contract driven

  — Alex Svanevik

• Labeling wallets requires inference and transaction flow analysis.

• You can literally send some money to binance and see where the flows go

  — Alex Svanevik

• Quality assurance ensures trust in data labeling processes.

• We do focus a lot on quality assurance

  — Alex Svanevik

• Every label in the database is backed by compiled evidence.

• You cannot add a label to the database unless you’ve compiled the evidence for it

  — Alex Svanevik

• Public information is used for labeling, with removal rights for individuals.

• We rely on public information right information that’s in the public domain

  — Alex Svanevik

Quality assurance in data labeling

• Quality assurance is critical for maintaining trust in data labeling.

• It’s very it’s it takes a long time to gain trust but it’s very easy to lose trust

  — Alex Svanevik

• Every label is backed by evidence to ensure accuracy.

• You cannot add a label to the database unless you’ve compiled the evidence for it

  — Alex Svanevik

• The integrity of blockchain data labeling relies heavily on quality assurance.
• Ensuring high precision in data processing is essential for credibility.
• The rigorous approach to data labeling enhances reliability and reduces errors.
• Nansen’s commitment to quality assurance builds trust with users.
• Accurate labeling is crucial for interpreting blockchain transactions.
• The process of labeling involves both algorithmic and human efforts.
• Quality assurance practices are vital for maintaining data integrity.

Public information and privacy on blockchains

• Public information on blockchains can be used to label addresses.

• We rely on public information right information that’s in the public domain

  — Alex Svanevik

• Individuals have the right to request the removal of labels.

• If you come to us and you say hey I actually want that label removed as an individual you can do that

  — Alex Svanevik

• Blockchains are public and transparent by default.

• Look this is actually on the blockchain like this is not it’s like immutable

  — Alex Svanevik

• Information on blockchains is immutable and cannot be erased.

• Even if we remove it from our database the ens name is always gonna be etched into that address

  — Alex Svanevik

• The transparency of blockchain technology has significant privacy implications.
• Understanding the principles of blockchain data treatment is crucial.
• The distinction between individual and corporate identities is important in labeling practices.
• Ethical considerations are integral to operational policies regarding blockchain data.
• Public and transparent nature of blockchains affects data permanence.