Were NFTs Too Early or Just Wrong?
For a while, NFTs were everywhere. The question now is whether that was the beginning of something durable or the entire story.
The idea was simple: you could own an object online. Not just access it inside a game or platform, but buy it, sell it, and prove it was yours on an open protocol. By 2020, Ethereum’s standards, wallets, marketplaces, and creator tools had matured enough for the entire loop to work. The pitch was no longer theoretical.
COVID created the conditions for that idea to spread quickly. In March 2020, as much of life moved online, the Federal Reserve cut rates to 0-0.25%. A month later, the personal saving rate hit 33%. Across three rounds in 2020 and 2021, the federal government sent more than 476 million stimulus payments totaling $814 billion.
For many people with secure jobs, that meant more cash, fewer places to spend it, and far more time online. The technology was ready just as a lot of people had money and attention to spare.
Top Shot made digital ownership legible to people outside crypto. It opened to the public in October 2020 and caught fire at the start of 2021. By the end of March, it had generated $500 million in sales and more than 800,000 accounts.
It put a new technology inside an old behavior: opening packs, collecting highlights, and trading with other fans. The NFT boom suddenly had a mass-market product.
From there, adoption happened fast. Snoop Dogg announced his first collection in March 2021, and Paris Hilton launched her first drop in April. Tim Berners-Lee sold an NFT of the web’s source code in June. Coca-Cola released its first collection in July.
By the fall, the NFL had announced its own digital collectibles. In November, Harvard Business School professor Scott Kominers wrote that verifiable digital ownership could change how creators, communities, and markets worked. By December, Nike had bought RTFKT and Adidas had issued Into the Metaverse.
Money followed. Chainalysis later estimated that $44.2 billion was sent to Ethereum NFT contracts in 2021, up from $106 million in 2020. In a year, NFTs had gone from a working idea to what looked like a large new consumer market.
The warning signs were visible too. Much of the activity was speculative. The market was full of obvious scams, insider dumps, fake partnerships, and projects selling little more than a roadmap.
But the market was not only scams and speculation. Artists were selling digital work directly to collectors, and the record of ownership was not tied to one marketplace. We thought an open creation and ownership layer was taking shape beneath the speculation. That was the signal that interested us, and we started building Highlight in late 2021.
OpenSea processed $4.86 billion in January 2022. It would be the peak. Then the market turned almost as quickly as it had risen.
As prices fell, some of the fraud came into view. By August, Elliptic had documented more than $100 million in NFTs publicly reported stolen through scams over the previous year.
”Publicly reported” is doing a lot of work there. From what we saw, the fraud was likely much higher. Theft figures also said nothing about insider dumps, fake partnerships, or teams that sold a roadmap and disappeared. The chain could show that a sale happened. It could not show that the buyer had been lied to.
The headlines kept pointing in opposite directions. In September, even after trading had started to fall, Starbucks announced an NFT loyalty program. On December 2, an academic study reported $3.4 billion in artificial wash-trading volume through the prior January. Two weeks later, Donald Trump, then out of office, launched $99 digital trading cards.
The market was shrinking and mainstreaming at the same time.
We saw the warning signs. What we overestimated was how much of the demand would survive them. We thought speculation had introduced people to a real new behavior: buying and collecting digital work. We expected enough of them to stay once prices stopped rising. Some did. Far fewer than the headline numbers suggested.
By June 2026, OpenSea’s monthly volume had fallen to $32.7 million, more than 99% below January 2022. The broad consumer market we imagined had not materialized.
A narrower use did emerge. Collector Crypt, a marketplace for tokenized physical cards, generated $406 million in trading volume that same month. More than 98% of volume on physical-collectible platforms came from randomized “gacha” purchases, not ordinary resale.
That is still a form of gambling. But here the token represents ownership of a physical card rather than being the product itself.
That does not vindicate the story we told in 2021. It shows one narrow use: tokens can work as ownership rails when the underlying object already has demand.
Native digital work faced a harder test: who would still create, collect, or play if resale expectations, token rewards, and a constant stream of new buyers disappeared? If the answer is nobody, the rewards are the product.
A small group passed that test. At Highlight, creators kept releasing work after prices and attention collapsed. We still think digital ownership gives creators another way to sell their work and collectors a way to keep it beyond the platform where they bought it.
That is a much smaller claim than we made in 2021. If NFTs return, it will not be because the label becomes fashionable again. It will be because people want the products even when prices are not going up, and open ownership makes those products better.