Friend.tech Trend.check - Part 2 of 2
Table of Contents
Part 1
Part 2
Important Disclaimer
4. To Be Fungible, Or To Be Non-Fungible
Blockchains pave the way for permissionless asset storage services, and there are two widely traded asset types in Web3: Fungibles (FT) and non-fungibles (NFT). In Section 5, we will attempt to provide an answer to the question, ‘between an FT project and an NFT project, which one should rely on Twitter more than the other?’
Analogous to the common emphasis on the community from the NFT side, we assert that NFTs should prioritize Twitter more than FTs. Twitter activities have a direct impact on NFT trading activities, while the correlation between Twitter activities and decentralized exchange (DEX) trading activities is much less pronounced.
Figure 5 – FT/NFT Trading Volume To Twitter Activity
4.1. Fame Does Not Gain
While we anticipate celebrities to earn a substantial amount of money, the correlation between fame and gain is less evident in the Twitter dimension. The two variables are loosely and weakly related as depicted by Figure 6.
Figure 6 – Account Balance per Subgroup
If users with a large follower base had used centralized exchanges (CEX) or router services to fund their wallets, their low balances might make sense.
Figure 7 – Density Distribution Of Untrackable Address
Figure 7 shows that the users with fewer followers are more likely to use CEX or router services as indicated by the discrepancy between ‘Funded From CEX/Router’ distribution and the ‘Original Distribution’ in the right tail.
But it might be too early to conclude a link between the fame and the gains; our criteria for filtering out flash accounts are below 5 transactions, which is easily achievable by a ‘sub-account’.
Thus, the information here could go either way – people with many followers use many wallets, or they are not significantly richer.
4.2. Seeing Is Believing For Tokens
We often come across threads on ‘hidden gems’ on Twitter by influencers who try to shill their bags. One intuitive challenge is to discern whether they are genuinely shilling their bags or are paid marketers.
Table 2 – FT Trading Volume Rank per Subgroup
*Stablecoins, $BTC, $ETH, and their derivatives (e.g. $stETH) are dismissed
Once again our dataset comprises the entire history of DEX trades, thus we observe rather ancient FTs, such as $ERSDL, on the list.
The list also includes various recent memecoins ($TURBO, $PEPE, $BITCOIN, etc.), and tokens that dominated Twitter during the DeFi summer ($OHM) as well. Since these tokens shilled on Twitter are generally ranked higher on the upper end of the Subgroups, we can at least recognize that influencers are at least buying what they are shilling.
Figure 8 – DEX Trading Volume Ranking per Subgroup By Chains
While the 10K+ Subgroup exhibits a radically different chain usage pattern with activities on Arbitrum and Optimism being larger than Ethereum, for other subgroups, the ranking typically follows Ethereum > Arbitrum > Polygon > Base > Optimism.
4.3. Do Not Judge An NFT By Its Description
One strange fact about Twitter accounts and their NFT trading volumes is the non-preference for the ‘Blue Chip PFPs’ by the 10K Subgroup.
A possible explanation could be that 10K PFP NFT communities are generally known to have 1,000 to 3,000 active participants empirically, and they somewhat overlap. These NFT communities following each other on Twitter would result in belonging to the 1K- to 5K+ Subgroups.
Table 3 – NFT Trading Volume Rank per Subgroup (Bluechip = Highlighted)
Figure 9 – NFT Trading Volume Rank per Subgroup
One amusing observation is that those who prefix or suffix their usernames with NFT-related terms* or write in their descriptions as NFT investors tend to trade less NFTs than those who are more reserved.
*NFT-related terms : NFT, or any Blue Chip PFP words (e.g. punk1234)
However, this could simply imply that these members are simply hodling onto their NFTs through market fluctuations, using their NFTs more as memberships rather than trading instruments.
5. Concluding Remarks
The importance of Twitter for Web3 projects as the main networking window is advocated by many, yet its impact was never truly quantified – all numbers before Friend.tech were mere proxies of the off-chains to the on-chains.
With Friend.tech API, we discovered that the activities on the Twitter communities lead to greater participation in on-chain activities, that influencing Twitter results in a larger consumer base, and that farming engagements is worthwhile.
Now that we know our actions on Twitter affect on-chain transactions, we could devise various marketing or business strategies to maximize their effects on the product’s growth.
Although our paper concludes here, numerous new projects are emerging to close the gap between the Web2 and the Web3 audience, opening up opportunities for new sets of data that were previously unavailable.
