tl;dr: This article presents an analysis of secondary sales and trading activity across several notable, multi-edition music NFT drops from late 2021 to mid-2022. As part of our analysis, we also develop a starting metrics framework for evaluating and comparing different music NFT drops over time. In particular, we focus on highlighting music NFT collector activity (both speculative and otherwise) and gauging the overall community health of a music NFT project — which, while ultimately subjective, can benefit from more precise measurements around how collectors behave over time. Our overarching goal is to provide a more in-depth and nuanced understanding of NFT markets beyond reporting purely on overall primary and secondary sale revenue or whether or not a specific drop sold out, which is where most media coverage otherwise stops.

Even taking the bear market into consideration, the last year has been a banner one for NFTs in the music industry. As we reported in our latest music NFT analysis, the token format generated over $86 million worth of primary sales in 2021 across many forms of utility including royalty splits, physical/digital merchandise, and community access.

Beyond introducing new consumption experiences for fans, NFTs also have the ability to kick back a portion of every sale to the original creator in perpetuity, through functionalities programmed directly into the smart contracts that power them. Web3 platforms and proponents commonly cite this secondary sales mechanism as a pro-creator benefit of NFTs that had not existed historically within the world of traditional physical merchandise.

However, there are conflicting reports around the extent to which NFTs drops actually get the opportunity to leverage secondary market dynamics in their respective lifecycles. One Nature article published in October 2021 found that only 20% of tokens in their dataset of 4.7 million NFTs had any secondary sales. In our previous research, we’ve also cited particular examples of disconnects between expectations and reality, such as when buyers of Tory Lanez’s NFT album felt “duped” after discovering that claims around the allegedly valuable secondary market for the collection did not necessarily hold.

Especially under current crypto market conditions, it is critical to understand what the secondary sales market really looks like for music NFTs, and to add accountability measures into public claims that artists and platforms make about the market — especially when fans’ wallets are on the line. Having the right expectations going in is critical for a variety of stakeholders:

• Artists and their teams: Assuming the artist is the primary beneficiary of an NFT drop, having realistic expectations for both primary and secondary sale revenue is important for projecting financial outcomes, especially if NFTs are used as a means of fundraising.
• Collaborators: Similar to artists, it is important for collaborators to be able to know how much their contributions might be worth when they are negotiating fees and splits.
• Buyers: Given the amount of hype around the speculative aspects of NFT trading, it is important that buyers are properly armed with accurate information about what’s happening in the market.

In this article, we analyze secondary sales and trading activity for several notable, multi-edition music NFT drops, and present an initial metrics framework for evaluating and comparing different NFT drops over time. As our NFT database does not yet include data on secondary sales, we decided to structure this research as a deep-dive into a hand-selected list of drops as initial case studies. Timing-wise, this article picks up where our 2021 analysis leaves off, covering a few notable drops that occurred from the end of last year up to the present day in 2022.

We believe that the metrics we’ve uncovered can help provide a more in-depth understanding of NFT markets beyond reporting purely on overall primary and secondary sale revenue or whether or not a specific drop sold out, which is where most media coverage stops. We also believe that these metrics can provide insight into the health of different Web3-native music platforms and communities, by measuring how much interest is sustained over time and what dynamics exist in terms of short-term speculation versus long-term loyalty.

Lastly, this analysis required us to understand the current state of NFT metadata and how transactions are represented on- and off-chain, so we will offer some of our own commentary as we think about the requirements needed to build out a comprehensive secondary sales database on top of our primary sales tracking.

How we selected drops to study

We went through our music NFT database and hand-selected which drops would serve as ideal case studies for a secondary sales analysis, filtering based on a number of criteria:

A. NFTs that were sold as fixed-price, multi-edition drops. In contrast to 1/1s sold at auction, NFT collections that registered a larger volume of units and revenue give us a wider range of data points, and therefore a more appropriate sample size from which to draw conclusions. As a general summary:

• The majority of drops (1,597) in our database are actually multi-edition (i.e. selling at least 2 NFTs as part of a drop), and the majority of artists in our database have dropped a multi-edition NFT (685).
• However, the larger the number of editions per drop, the fewer number of drops and creators represented, with only 68 (less than 10% of artists we have drops tracked for in our database) artists or music organizations releasing multi-edition NFTs of at least 1,000 copies.
• Of these, only 18 artists or music organizations have released multiple drops of 1,000+ editions each.

B. Drops that sold out in primary sales. While there are surely inefficiencies within the secondary market, evaluating secondary sales for an NFT drop would arguably be premature and unreasonable if original editions are still available.

C. Drops that were at least a month old as of May 1, 2022. Since we include metrics centered around time frames like the first month, we only considered drops that met this cutoff for when we performed our analysis.

D. Drops that featured only one main music artist. Name-billing of other major collaborating artists alongside a primary music artist presents a direct confounding variable.

E. Drops whose primary and secondary sales activity could be verified on-chain. With immutable record-keeping as a key feature of the blockchain, we wanted to rely on decentralized transactional data that anyone can access and verify. This disqualified notable platforms like Nifty Gateway, even though we found that that platform represented 64% of tracked primary sales for music NFTs in H1 2021. Nifty Gateway uses a more “hybrid” opt-out custodial model, where users can conduct secondary sales internally (i.e. not on-chain) within the originating platform, but users can also elect to “withdraw” their NFTs to engage in secondary sales outside of the originating platform. Some notable exclusions that we would have liked to analyze from Nifty Gateway include drops from Grimes, Steve Aoki, and early 3LAU and RAC pieces. Other artists with prominent drops in our database that we considered but ultimately left out for similar or other reasons include: deadmau5, Mike Shinoda, Doja Cat, and Tory Lanez.

It’s worth mentioning that continuously consolidating and unifying different representations of metadata and transactions across NFT platforms ended up being an ongoing setback throughout this project, and remains a major obstacle for building out a database of secondary sales (see “Future research” section at the end of this article for more details here).

With this criteria in mind, we tried to select a diversity of artists representing different genres, label affiliations, and audience sizes. We especially prioritized artists who have had multiple NFT drops across time, as this would provide a more controlled comparison of how drops fare for an artist across a temporal lens. Since our research required going through contracts on a drop-by-drop or platform-by-platform basis, we were unable to achieve an ideal level of diversity in platforms, though this does give us another controlling factor in comparing drops occuring on the same platform.

Below is our final drop list that we considered in our case-study analysis. All of these drops occurred on Ethereum, and most of the drops occurred in the later months of 2021 and 2022. For the sake of scope, we analyzed metrics for these drops up until May 2, 2022 as our cutoff.

Secondary sales transaction data and caveats

While we used Etherscan and the platforms hosting the drops to verify transactions, we ultimately relied on the OpenSea API to compile all initial transfers and successful sales for all of the drops in our case study, due to its place in the secondary market and their consistency in metadata across drops. This limited us to the quantity and quality of data available through their events endpoint, which does not always have complete history for older drops. For example, we originally included King of Leon’s YellowHeart drop but had to drop this from our analysis upon finding that the data we could retrieve from related transactions was incomplete.

A few other OpenSea-related disclaimers:

• One drawback of relying on the OpenSea API is that we are not necessarily capturing sales that occur on other secondary marketplace platforms. However, after tracing the full sales histories of a number of our drops, we believe these sales are negligible in relation to the overall volume of secondary sales.
• The platform allows the ability to bundle assets in a transfer (example), but because any assortment of assets can be bundled together, we removed bundle sales involving our drops of interest because we cannot isolate the price of a single NFT out of a bundle. Snoop Dogg’s Sound drop was the only drop affected by this.
• OpenSea also provides full events for bidding and offer-related activity, which could be used to enrich an analysis of the secondary market, but we chose not to factor that into our analysis if it did not lead to a material sale.

Regarding financial metrics, since we’re making comparisons between initial sales and secondary sales mostly using the same cryptocurrencies, we are not factoring in conversions to USD or other fiat equivalents, and will only be making ETH-to-ETH comparisons, acknowledging that the price of ETH has dropped over time since the initial mints of most of these drops.

ANALYSIS: A framework for measuring the lifetime value of music NFTs

The metrics we used to analyze the above music NFT drops can be broken down roughly into three categories: Summary revenue stats, collector-level activity, and gauging speculation.

All screenshots below are taken from the accompanying interactive dashboard for this article, which we invite you to play around with yourself.

Relevant metrics:

• Total revenue brought in by secondary sales — including the OpenSea split (multiplying the total secondary sales revenue by OpenSea’s 2.5% take rate on secondary transactions) and the creator split (multiplying total secondary sales revenue by any contract-level parameters around what % of future sales are kicked back to the original creator[s].)
• Average secondary sale price — Calculated by dividing the total secondary sale revenue by the number of secondary sales.
• Ratio of secondary sales revenue to primary sale revenue — A measure of the extent to which secondary sales bring in more revenue relative to primary sales.
• Secondary ceiling/floor prices — Maximum and minimum prices for any secondary sale of any edition at any point in time.

In the drops that we analyzed, we found that secondary sales figures frequently eclipsed the initial revenue brought in by primary sales. Part of this may be due to the use of fixed-price offerings for multi-edition drops, allowing initial prices to be deflated relative to market demand (as opposed to auctioning mints to the highest bidder from the jump).

For example, Sound has thus far kept a consistent primary sale price of 0.1 ETH per NFT across all drops. However, across all six Sound drops included in this analysis, the average secondary sale price per drop was at least triple that initial offering — ranging from oshi’s CHILDHOOD drop going for an average secondary sale price of 2.06 ETH to Snoop Dogg’s Death Row Mix: Volume 1 fetching an average secondary sale price of 0.34 ETH. With the exception of Snoop Dogg’s drop, even the floor price across all Sound drops starts at 0.22 ETH (RAC’s Sinners), securing profit for any secondary sale relative to the initial minting price.

By these numbers, Snoop Dogg’s drop may look like it underperformed the rest, but it’s important to factor in that Snoop Doog’s edition size of 1000 represents 10x less relative scarcity than the next Sound drop we considered (from RAC), which may partially explain some of the differences in these metrics. Among the Sounds drops, it appears that larger editions generally had a lower average secondary sale price.

Monte Booker’s Kolors drop via Soulection was the only drop we analyzed where secondary sale revenue has not surpassed primary sale revenue to date (26 ETH versus 40 ETH, respectively). We can observe from our data that unlike in other cases, the average secondary sale price was hardly greater than the initial offering price of 0.08 ETH, and in fact actually represents a loss for the reseller after taking platform and creator fees into account. In one case, a seller even appeared to list an NFT for free, leading us to record a floor price of 0 for the drop.

“Per-token sold” normalization

While the above metrics help summarize the financial scale of a particular drop, different drops have different edition quantities and different proportions of editions that get resold. Thus, to evaluate and compare performance, it may be more helpful to normalize some of these metrics on a “by-token” basis, i.e. dividing secondary sales by the number of editions resold.

Relevant metrics:

• Median/average number of secondary sales per token sold — A measure of how many successful sales we can expect a given token to generate. While we account for the ratio of secondary sales to primary sales later on, we consider this on a “per token sold” basis here to quantify the practical outcomes on the secondary market
• Median/average lifetime revenue per token sold — With lifetime revenue defined as the sum of all revenue over all sales for a token, we can summarize the post-primary tail of financial value created for tokens that resell. (If a primary buyer does not resell their token, the lifetime revenue would only be the primary sale price).

By looking at the number of sales per token, we can break down how secondary sales generate revenue — are sales driven by a long string of transactions and participants, high resale value, or both? One interesting similarity we found was that the median number of secondary sales per token sold was just 1 across all the drops we analyzed. This means that the same NFT does not tend to get sold over and over again, suggesting that while secondary sale revenue can outpace primary sale revenue, this is not necessarily occuring due to a high volume of secondary sales activity. Moreover, given that we are starting with a biased sample of relatively notable, multi-edition drops — which inherently drive more hype and participation on social media — we suspect that this median number represents the higher range of resale activity among these kinds of drops.

This metric naturally drives toward quantifying a lifetime value of music NFTs on the secondary market. Since we saw that per-token resale volume was relatively consistent across the drops we looked at, we can observe a loose correlation between average sale price and average lifetime revenue per token sold. If we look at the median lifetime revenue per token across drops, we can see some slight deviation, chiefly between Matthew Chaim’s first and second Sound drops, Nice Guy and Climb: 0.7 vs. 0.82 ETH, indicating that the distribution of lifetime revenues is skewed much more positively, suggesting a significant concentration of high resale prices and/or volume in the Climb’s secondary sales. Snoop’s drop also outperformed RAC’s drop in this metric (0.49 vs. 0.38 ETH) despite having a lower average secondary sale price overall(0.34 vs. 0.55 ETH), perhaps buoyed by having such a higher average number of secondary sales per token sold (1.78).

Relevant metrics:

• Unique wallets involved in primary sales (mints) — How many people are minting NFTs?
• Number of editions — How many unique tokens are included in a drop?
• Ratio of unique minting addresses to editions (minter:edition) — How many people are minting NFTs relative to how many are available?
• Ratio of transfers to sales (transfer:sale) — How many transfers (whether part of a sale or not) are occurring per secondary sale?
• Ratio of unique sellers to buyers (seller:buyer) — How large is the pool of sellers relative to the pool of buyers?

Our previous work on fan sentiment indicated that financial accessibility and inclusion were among the top perceived areas for improvement in the Web3 ecosystem. While multi-edition drops at relatively lower price points seemed to be one way in which artists have been trying to address this issue, they still remain inaccessible to the majority of music fans (the median USD equivalent primary sale price for a music NFT was $825 in 2021).

We wanted to make sure we accounted for this accessibility issue in our analysis — by looking not only at the number of unique wallets that end up being able to mint, but also at those who end up participating in the secondary market. In considering both primary and secondary sales, we can build a full picture of parties involved by counting all unique buying and selling wallets, which we can normalize against the total number of editions.

One important caveat to translating this as an estimate of how many people are actually behind the wallets: If a transfer occurs without a linkable sale, our analysis assumes that this was not part of any monetary exchange. When inspecting some of these occurrences, we noticed that many of these transfers appeared to involve buyers transferring NFTs to separate vault wallets or potentially gifting NFTs to others. However, without additional context, we cannot verify the nature of individual transfers.

As such, we have also included the ratio of transfers to sales. One might expect that a relatively consistent proportion of buyers are making non-monetary transfers with their purchases, so deviations in this metric could highlight irregularities in our data. For reference, each sale should lead to a transfer, so any transfer:sale ratio above 1:1 would indicate some non-sale-related transfer activity.

While these ratios didn’t vary widely among the drops we looked at, oshi’s first Sound drop notably had the highest ratio at just over 2:1, meaning for every sale in the data, there were 2 transfers. Since this drop also happened to be the first drop that the platform ever did, one plausible explanation is that buyers were more likely to move their NFTs to alternate wallets or vaults as a secure keepsake, or to display the NFTs as an extra special collector’s item.

Another ratio that helps shed some light on the secondary sale ecosystem is the ratio of unique seller addresses to buyer addresses. In a sense, this measure captures the extent to which the secondary market is expanding or concentrating participation. For example, in Monte Booker’s case, many people minted and then resold multiple editions of the NFT to a broader pool of buyers, leading to a seller:buyer ratio of less than 1:1 (1:0.96). A ratio of 1:1 would suggest that each seller ends up selling to a unique buyer, and the collector ecosystem of editions does not become any more concentrated or diluted.

Matthew Chaim’s second Sound drop exemplifies the other side of this spectrum. With a seller:buyer ratio of over 2, editions of this drop are becoming incredibly concentrated relative to the starting state of 30 unique addresses for 30 unique editions during primary sales. Upon closer inspection, we can see that one wallet (degendavinci.eth) bought 11 editions of this drop, largely driving this consolidation. While these NFTs do not represent any form of ownership akin to how stocks in a public may work, this kind of behavior is definitely something to monitor where NFT ownership confers some sort of influence, such as providing additional votes for DAO governance.

To combat problems such as botting and other user abuse, platforms such as Sound and Royal have implemented queueing systems that try to limit mints to one sale per wallet, or at least give everyone who’s there at drop time a fair drop. However, a look at the actual transfer data reveals that this doesn’t always work.

With one exception, all of oshi’s, Matthew Chaim’s, and RAC’s drops have an equal number of unique wallets minting to editions. (With Matthew Chaim’s second drops, there were 7 failed mints that had to be reverted, but we could not derive the reason for this.) In 3LAU’s Worst Case drop, these NFTs were part of a free giveaway celebrating the initial launch of Royal. A glance at this ratio here shows that they were fairly successful at limiting users gaming the system for multiple NFTs, with 306 wallets minting 333 editions.

However, with Snoop Dogg’s drop, there are 668 wallets minting 1,000 editions. And contrasting from the more-established platforms of Sound and Royal, it appears that Soulection had a harder time preventing repeat buyers: 286 unique addresses minted 500 editions — almost two on average per wallet.

It’s unclear whether the sole presence of wallets minting multiple editions always correlates to a high concentration of speculative primary sales. Even for drops with homogenous editions, one could always argue that a buyer might be buying “extra” editions for gifting or collecting purposes, and there are some utilities, such as royalty splits, where it might make sense for a buyer to amass many of the same NFTs. That said, it is certainly suspicious to see this phenomenon occur as prevalently as we see in some of our drops.

In the next section, we go over how looking at other factors around a drop, such as changes in price and velocity over time, can help us better contextualize how the secondary sales market is trending relative to the age of an NFT, and assess whether the primary buyers of a given NFT collection end up holding onto their purchases or immediately selling them.

Relevant metrics

• Number of unique tokens sold on the secondary market (in the first week/month) — We cannot assume that all tokens are available for resale. Looking at the number of unique tokens that resold gives us a gauge of how many have been available on the secondary market.
• Percentage of unique tokens that get resold (in the first week/month) — Normalizing the number of tokens sold to the edition size.
• Overall ratio of secondary sales to primary sales — Considering the entire drop, how many secondary sales occur per token?

While music NFTs ostensibly offers utility to fans of artists and their music, we cannot discount the speculative aspect of NFTs as a commodity, as it can be easy to flip an NFT for profit if you are able to get access to a primary sale. At the same time, it’s difficult to determine without asking buyers directly whether they are purchasing an NFT with the purpose of collecting or the purpose of speculating/reselling. A buyer who may have originally intended to collect an NFT could change their attitude over time, especially if the resale price goes up dramatically. This section addresses a number of data points available to help us assess what might be going on.

It’s worth reiterating here that selling out was a prerequisite for a music NFT drop being included in this analysis, which would logically imply that the demand is greater than the supply for these drops at their initial fixed price. This can lead to activity similar to that in some other resale markets, such as with concert tickets or sneakers, where there are a fraction of buyers who are purchasing with the intent of immediate resale. In general, with or without anti-abuse restrictions, sellouts can also happen as quickly as a site’s infrastructure allows, so we do not consider time to the secondary market after an initial drop as an important factor.

In order to make comparisons, we have timeboxed certain measures such that we’ll only be comparing equivalent time periods across drops; in particular, we took a look at the number of unique tokens sold in the first week and first month after our sale, and interpreted these numbers in relation to the proportion of editions they represented.

Snoop Doog’s Sound drop immediately stands out here as a drop that quickly saw a lot of secondary market activity: A whopping 74% of minted tokens (739/1000) in the collection were resold within the first week alone, a number that goes up to 80% when we look at the first month. The ratio of secondary sales to total primary sales was 1.47:1, meaningthe pure total volume of secondary sales has already surpassed the primary sale volume (the only one of our drops for which this happened): a noteworthy feat considering roughly 20% of tokens are still being held by the original buyers.

It’s important to acknowledge that looking at sales is a results-driven perspective, perhaps a skeuomorphic inherent of the music industry’s emphasis on first week sales and streams. Looking at early resale percentages does not necessarily indicate healthy demand. Monte Booker’s drop saw 44% of minted tokens (216/500) resold within the first week — but when we synthesize this with our earlier callout that the average secondary price hardly breaks even, we might hypothesize that initial demand was not very high, and speculators may have had to quickly cut their losses. Again, these are only potential inferences we can try to draw from the data, but comparing different drops helps to provide multiple data points to analyze differences we observe.

Relevant metrics: All of the above, compared across multiple collections for the same artist

Because the music NFT market is still in its relatively early stages, there are only a handful of artists who have successfully sold out multiple collections, let alone experimented with drops across multiple different marketplaces. That said, for an extra controlled comparison, since we have multiple data points for oshi and Matthew Chaim, we can get a hint of how the secondary markets of these Sound drops and artists may have evolved due to time. In both cases, we find a consistent set of patterns:

Higher total secondary sale revenue

• (oshi: 22.64 → 24.08; Matthew Chaim: 13.17 → 17.47)

Lower average secondary sale price

• (oshi: 2.05 → 1.25; Matthew Chaim: 0.82 → 0.73)

Larger percentage of unique tokens resold within the first week / month

• oshi: 12% → 37%; Matthew Chaim: 32% → 50% (week 1)
• oshi: 28% → 60%; Matthew Chaim: 36% → 70% (month 1)

It’s impressive to consider that the higher total secondary sale revenue pattern we observe here occurs despite the fact that the drops for these artists were both released roughly two months after the first, meaning there has been a lot less time for secondary sale activity to occur. In Matthew Chaim’s case, the pattern persists even when controlling for the number of editions. We can see from the average secondary sale price and percentage of unique tokens resold metrics that this revenue was driven by the greater propensity of second drop collectors to resell — enough so to compensate for a lower resale price.

oshi and Matthew Chaim’s first drops were some of the earliest drops Sound did (oshi’s CHILDHOOD was actually Sound’s first ever overall), which may help explain both why the average sale price was higher for the first drops. This could also help explain why more initial minters held onto their drops as collector’s items — these NFTs offer collectors the prestige of owning an “original” edition.

There’s an interesting dynamic at play here: If an artist or platform is able to create this kind of prestige (or help generate any kind of value for that matter), then we can say that they have successfully created a sense of value. This value, expressed financially, can be measured through primary sale sellout prices followed by sale prices on an open secondary market. Through our case study, we have presented a framework for evaluating the financial outcomes of music NFTs on the secondary market. But is that generated value necessarily the healthiest measure of success and longevity for a given artist?

There are many ways to tie secondary sales to the “health” (or lack thereof) of Web3-native platforms, artists, and their communities. For off-chain activities, we might consider continued engagement and participation to be signals of good health for a community. Translating this to on-chain financial activity, sustained interest in the secondary market effectively reinforces the value of NFTs over a longer term. However, as people continue to buy, the composition of collectors also changes. In a time when community-building seems to be increasingly important for artists, what is the role of continuity needed in maintaining a healthy community? For example, if an NFT provides access to an exclusive Discord server, but members frequently resell their NFTs, this could have negative ramifications for building a strong, lasting community.

While the kickback mechanism of NFTs ensures that artists are continuously rewarded for secondary sale activity, it also means that artists have a financial incentive to encourage reselling over collecting and holding NFTs. Under this paradigm, one could even argue that artists economically benefit more from attracting speculative parties over loyal fans when it comes to NFT participation.

For example, let’s say an artist gets 10% of every secondary sale. If a fan initially mints an NFT at price of 1 ETH, with no intention to sell until they are eventually tempted by a price of 10 ETH on the secondary market, then that NFT would only generate 2 ETH in lifetime value for the artist up to that point. Now let’s say that the same NFT falls into more speculative hands instead, and it changes hands every time the secondary sale value goes up by 1 ETH, selling at 2 ETH, 3 ETH, and so on all the way until the market reaches 10 ETH. In this case, this NFT would have generated 4.4 ETH more than the previous case. This is a significant divergence from a company’s stake in how securities are traded. A company might want to have an ever-increasing stock price, but it does not have the same incentive to want its stocks to be frequently exchanged along the way.

If NFTs continue to provide a new valuation model for fandom or the artist-fan (or even platform-fan) relationship, then these trade-offs are important to wrestle with when thinking about building communities. How do we balance providing economic sustainability for artists with stability for their communities?

In the future, we hope to enrich our database with enough data to do a more comprehensive analysis along these kinds of lenses. In taking a case study approach like this, we cannot definitively generalize trends across the whole market, and there are many interesting and useful questions we are not able to answer, including:

Secondary sales activity for 1/1s and other NFTs sold at auction

As we accumulate more on-chain drop history in the music/Web3 ecosystem, we may be able to reproduce the above analyses at scale for NFTs that are not fixed-price — e.g. potentially pooling together multiple 1/1s over time to perform similar large-sample analyses, or doing a more general analysis of the relationship between initial drop price and secondary sale activity.

Secondary sales trends by utility

There are a myriad of ways to categorize artists and their NFT drops, including but not limited to utility, genre, label affiliation, and tenure in the crypto community. Even our handful of case studies feature add-ons such as mystery direct “special utility” (Monte Booker), Sound’s Golden Eggs, and royalty splits (Royal) through NFTs, which may theoretically affect the worth of an NFT over time.

In our previous research, we explained how we could not comfortably analyze correlations at scale between different kinds of utility and the ultimate success of a drop, given the lack of standardization in how this utility is registered on- or off-chain across NFT marketplaces. For our analysis, we did not separate out different tiers or traits within drops, though this would be a natural next step for any continuation. A similar case-study approach analyzing the long-term health and evolution of several drops grouped by similar utility might be fruitful for uncovering subtle economic differences among various subsets of the music/Web3 ecosystem.

Network analysis of music NFT collectors

Another area that was too ambitious in scope to tackle in this analysis was mapping out the overall network of activity among wallets across drops to not only understand if participants in one drop were more likely to participate in others one, but also to understand what other NFTs and tokens music NFT market participants have activity in. What does the community of holders for a particular drop look like? While the data required for such an analysis (especially one that can incorporate historical trading activity) would be rather unwieldy, this is arguably one of the best examples of how the transparency of blockchain could be used to unlock a wealth of information about the participants involved in NFTs transactions.

Network analysis could also help us assess the prevalence of wash trading — the practice of sellers being involved in the buying of their own assets in order to artificially inflate their values (e.g. buying via an alternative wallet or paying a friend to purchase the asset). The same Nature article we cited earlier also found that the top 10% of NFT traders perform 85% of all transactions. While this statistic alone does not necessarily indicate that wash trading is rampant, it does reflect a certain power law distribution that provides context for how striated activity in marketplaces currently are. Previous analyses focused on specific marketplaces or collections have also attempted to identify the prevalence of wash trading. A recent academic preprint looked at the top 52 collections by volume and estimated the lower bound of wash trading on Ethereum to be 2.04% of all sale transactions. Last April, it was reported that 95% of trading volume on LooksRare, another NFT marketplace, can be attributed to wash sales. While wash trading is illegal in the case of securities, it is currently unregulated in crypto markets.

Market conditions

As noted, we only considered NFT activity prior to May 1st, 2022, as the original analysis for this article took place in the months of May and June 2022. Therefore, results and trends may not be indicative of the current state of these drops at the time of publishing, although we do observe that the majority of activity occurs within the first month of drops (which we do fully cover here for each drop).

In addition, the crypto market has also entered a severe bear market alongside increased turbulence in the financial market, which will likely affect NFT trading activity for the foreseeable future. However, since we are primarily using the drops in our case study to exemplify how to assess and interpret our proposed metrics, we believe that the metrics presented here can be applied regardless of broader market conditions.

Wallet holdings

At the time of publication, co-author Michael Zhang holds one music NFT (3LAU) that was gifted to him on Nifty Gateway in 2020, and currently have no plans of purchasing or selling music NFTs. His wallet address is 0x9eA786aB6Cbe2e76501BAaC11B814C86210a1059.

Co-author Alexander Flores currently holds 4 music related NFTs: “Ace of You” by Autograf, a Chaos Festival Pack, “It’s Your Time” by Jagwar Twin, and a COLORSxCOMMUNITY Founding Pass NFT. His wallet address is 0xe85ca182dfc6e183492623d96975ea67eb1fb03f.

Editor Cherie Hu owns a COLORSxCOMMUNITY Founding Pass, a Chaos Festival Pack, a Rattle Society Genesis NFT, and several other NFTs for projects and brands outside of music. Her wallet address is 0x03e9206f2A1234fEFaD2C0B07A86B1E0f5cB8d0D.