Ethereum Wallet Bulk Analysis – Case Study

If you want to know what transactions an Ethereum wallet has made in the past you just need to look at the Ethereum blockchain. The same goes for the balance, NFTs the wallet owns or any other interaction on the Ethereum Blockchain.

But what if you want to check not just one wallet, but more than 52,000 wallets?
And could you do it in under 48h?

That was the situation a blockchain company from London approached us with. In this case study, we show how Wetomate successfully extracted and analyzed blockchain-based data from over 52,000 Ethereum wallets in record time.


For reasons of anonymity, the name of the client may not be mentioned in this case study, but the client agreed that the results can be published and also gave a testimonial.

Financial Industries /

less than 48 hours

Client location
United Kingdom /

The client was jointly responsible for the planned launch of an NFT project, for which the interest of investors was tested in advance by means of the allocation of whitelist spots. Through marketing measures, a large number of wallet addresses had already been collected, but doubts had arisen about the quality of the data and there was a suspicion that entries might have been generated automatically. This can happen when influencers are paid on a performance basis per transmitted wallet addresses from interested fans.

Due to the tight schedule, the client approached Wetomate and asked to analyse the 52,000 Ethereum wallets to find out how many wallets were authentic and how likely the NFT launch was to be a success based on the existing data.


The problem in this case was, on the one hand, the large amount of data records, the short time and the fact that the data important for the decision was not directly available, but first had to be read from the blockchain. Individual wallets can be analyzed very easily with Etherscan by hand, but if you want to check 52,000 wallet addresses, it takes a very long time.

The data sets provided by the customer contained timestamp, email and ETH address of over 52,000 potential investors.

Example of the input data

The goal was to provide the customer with an analysis of the quality of the data and, if possible, recommendations for the next steps.


To do this, the following information was retrieved about each of the 52,000 wallets.

  • ETH Balance: the amount of Ether on the wallet
  • ETH Last transaction: The date of the last Ether transaction
  • N of ERC20 Tokens: The number of different ERC20-based tokens in the wallet
  • ERC20 Last transaction: The date of the last ERC-20 transaction
  • ERC721 NFTs: Number of ERC721-based NFTs (including also ERC1155)
  • N of ERC721 transactions: Number of ERC721-related transactions (including also ERC1155), representing the level of interaction with NFTs

An example of the recovered data can be found in the screenshot below.

A few of the generated data results

Based on this data, every wallet was rated on multiple different criterias, including:

  • Date of the last transaction (ETH or ERC20)
  • Amount hold in the wallet
  • NFT activity

Based on these and other factors, it was now possible to provide detailed information about the quality of the data and to provide answers to the customer’s concerns.


The result was a detailed analysis of all 52,000 wallets, segmentation of the data, and statistical analysis, which yielded the following results, among others:

  • Only a fraction of wallets would currently be able to mint (buy) the NFT due to the low ETH balance
  • A high amount of wallets with exactly the same ETH balance (0. 0024832 ETH) were found. We were able to trace the transactions of these wallets back to a small amount of wallets which have funded thousands of wallets with exactly the same balance.
  • Starting with data entry number 6234, we found emails in a strange format, as the email addresses did not contain numbers, which is statistically very unusual
  • Only two “wrong” Ethereum addresses were found, which is strange considering 52000 people are said to have created entries, this is a very low error rate
  • A lot of inactive wallets were found, indicating a lower possibility that these wallets will actually buy

Based on these findings, it seems likely that the entries were created by automated methods. Furthermore, the likely manually created data shows low quality and indicates with high probability that these wallets will not make the purchase of an NFT.

We advised the project to delay the launch of the project, as a possible flop of the project would be very likely based on the current data situation. Also, the project was advised to levy consequences against the parties that provided the data, as at least some of the data was added automatically.

In return, the client received full access to all analyzed data and the following results:

  • Analysis results of all 52,000 Ethereum wallets
  • Short video summarizing and explaining all the results
  • Concrete recommendations

Through Wetomate’s help, the project was helped in under 48h, all questions were answered and concrete recommendations were made.

I am extremely happy that I stumbled across Wetomate. I had an extremely large data set that needed to be analysed. Stefan dived straight into my data set and provided the most in depth report I have every seen! He really put my mind at ease and went above and beyond! One of the best online services I have had the pleasure to use. Thank you again and I will be a returning customer for life!

Jamie, CMO @ a Stealth Project based in London

This project has shown that Wetomate can also solve enterprise problems in blockchain-based data through automation and data analytics, and demonstrates the ability to deliver incredible results in a very short period of time.

Your personal contact

Do you have a similar problem or questions about this case study? Our experts are happy to consult you on data analysis and automated data retrieval.

Related Post