No matter if you are a trader or a passionate NFT collector, there is always a need to track collections’ health and what’s your tokens’ value on the market. Why’s the floor price soaring? What’s the sales volume? Is there an artificial pump? What do the whales do? Who makes the most profit from the collection?
There are tens of NFT tools out there, free and paid, all providing their own numbers and charts. So it looks easy — just open the first one you can find, take a look at the charts and then sell or buy nice tokens for fun and profit. Wait…
Yes, it can happen sometimes. That does not mean your favorite NFT tool is not precise. There is a number of reasons why the numbers can be inaccurate:
- It can be just expected because different platforms use different data sources. For example, the Total Volume in a collection can differ by a few times just because the tools account for a different set of marketplaces, i.e. OpenSea sales only or LooksRare and other marketplaces too.
- Swap transactions, executed on the swap platforms like GemSwap, can be difficult to account for sometimes, as a platform may act like a buyer who always spends and never earns, which leads to incorrect profit numbers across the whole transfers chain. Such addresses sometimes pop up on different tools’ leaderboards, and it usually does not make much sense.
- The data sources themselves can lie. Unfortunately, some APIs can return not very accurate data sometimes.
- It can be designed to have the data outdated by tens of minutes or even a few hours, up to 1 day, mainly because of the underlying technology and/or high data volumes. Usually, the tools clearly indicate what is their expected data freshness. The difference across tools can be especially visible when there is active trading that dynamically changes the numbers (for example, the number of tokens held by wallets). Some tools catch up quickly, some don’t.
- There can be just bugs. Yes, building correct NFT analytics is not that simple and usually requires thorough quality assurance, even though the analytics tools are not positioned as financial instruments in most cases.
Artacle provides analytics data, all these charts, and numbers that can be used to estimate the collection’s health. So what do we do to ensure the analytics quality? Let’s take as an example the Wallets Leaderboard that has been recently added.
Data Sources
Artacle relies directly on the blockchain data that include:
- Minting: what was the token minting price (that’s accounted to the one’s spending when calculating the profit)
- Transfers: how tokens move across wallets, who is the current owner of each token
- Sales: Artacle reads sale events from the most popular marketplaces (OpenSea, LooksRare, X2Y2, Rarible, and the CryptPunks contract — it has a built-in marketplace) to cover the vast majority of all sales across Ethereum blockchain. There are other marketplaces with lower volumes though, and they can be added if required.
- Swaps: the swap platforms sales usually require special attention to account for the ownership, profits, and losses correctly, as NFTs often traverse these platform’s addresses in the scope of swap transactions. Example: 0x0ff1dc85b391ff8c9efd58348342c3a7c177cf69c305c3ebf5e0982de17b6595
Generally, the blockchain data is quite reliable to build analytics on. Artacle identifies and fixes possible data hiccups with special techniques to make it even better.
Data Freshness
Artacle grabs fresh Transfers and Sales and accounts for them immediately when calculating the Wallets Leaderboard. Although there is internal caching in the application, the data is kept up-to-date with a small delay. There are other tools that update the ownership stats with a delay of a few hours or even a day, as well as tools that keep the data more or less fresh. A small hint: it is easy to validate whether the ownership data is up to date with respect to the number of tokens — most of the ERC-721 Smart Contracts have the balanceOf method that returns the exact value per address. Etherscan allows checking it quickly when in doubt.
Data and Charts Quality
Artacle applies the best practices and advanced quality assurance techniques to deliver good quality given the nature of the data.
- Excessive manual and automated testing
- Continuous data stream monitoring and immediate reporting when there are interruptions (for example, certain API providers have their planned or unplanned downtime, which Artacle detects and then double-checks the data when the API is back in service
- Data consistency analysis (such as possibly lost hops in the transfers chain) — detect and recover possibly corrupted data, basic sanity checks
- User’s issue reports in our Discord channel — we pay special attention to that, it is always a top priority.
Artacle leverages these approaches to deliver nice and clear analytics like this:
Artacle pursuits the best possible analytics quality not only for the PFP collections but for the Generative Art projects, such as ArtBlocks, CryptoCitizens, or BrightMoments. The Generative Art collections are not usually covered by the NFT tools out there.
We are open to any suggestions and feedback. Should you have a feature request, a crazy idea that makes the NFT flow easier for the community, or any kind of feedback — feel free to share with us at feedback@artacle.io or in our Discord.
