Lots of web3 projects — web3 games especially — have had splashy launches in the last year, selling millions of dollars of NFTs or pre-selling tokens that will be used as a currency within their game or app. They then trumpet that success to investors and use the momentum to catalyze massive growth fundraises that would normally be reserved for much more established companies. Their central claim is that consumer demand for tokens or NFTs is proof that people want what they’re building — that they have strong product-market fit.

This is not always true. In fact, it’s the exception rather than the rule.

Here’s the quick test you can use: does the NFT’s— or the series of NFTs’ — core functionality derive from another product? If no, then you have PMF.

What does this look like in practice? Companies like RTFKT that create NFTs where all the value is in owning the item itself, like a real-world piece of art or a fashion accessory, can count NFT sales as product-market fit. People are buying the NFT for its own value. The NFT’s utility doesn’t rely on its use in a game or an owner’s ability to compose it with other NFTs: it is purely in owning it and showing it off, of being a member to an exclusive club of “people who own this thing and have this taste”. It has value as a pure social signifier, independent of its functionality in a game world.

Items don’t have to be unique or 1-of-1. The same logic can apply to a series of NFTs such as CryptoPunks, Bored Apes, Azukis, and more. The value of a CryptoPunk might be a function of the value of the collection (much as the value of a Van Gogh might be correlated with other Van Goghs), but a CryptoPunk’s value is that it’s a CryptoPunk, not that it has specific functionality within a game world. (I’m going to leave aside the question of “What’s the appropriate valuation for a CryptoPunk or for a company making an NFT series?” That’s outside the scope of this post.)

In contrast, let’s examine spaceships in Star Atlas or land in the Sandbox. The functionality of land in the Sandbox is entirely dependent on the game; without the game and without people playing the game the land is meaningless and has no value. The same is true of spaceships in Star Atlas: with no game the NFTs have no utility and no value (except the probability-weighted estimate of future value). While many web3 advocates maintain that web3 games are better for consumers because because you “own” your assets and can move them between ecosystems (therefore conferring a persistence to their value), there are vanishingly few examples of this in practice. Developers, so far, don’t want to be encumbered with supporting assets designed for different use cases, with different in-game stats and non-compatible models, and from which they derive no revenue.

(As an aside: that might be an interesting user acquisition vector: take the assets from a defunct game, build a new game around them, and notify the asset-holders of the newfound value! That, however, necessitates all the assets be on-chain, and as of yet there are few NFTs that are stored entirely on-chain. A character model from Halo, for example, is far too large and complex to be stored on-chain. You could store the stats — health, damage, whatever — but a tens- or hundreds-of-thousands-of-polygons 3D model, plus all its textures, is far beyond current capabilities to store on-chain.)

One could posit that it’s possible for an asset in a game world to have value outside of that game world — e.g. if you have a dope Star Atlas spaceship, that’s indicative of the fact that you’re really invested in the game — if the game fails, that value would drop precipitously. And when web3 games have shut down in the past, that tends to be what happens. By way of illustration: F1 Delta Time minted rare NFTs representing specific drivers, tracks, and cars, that sold for up to $270K, but most of them have dropped to near-zero value (or have ceased trading, making their value indeterminate) since the game shut down. (While F1 Delta Time went out of business because of an IP dispute, not specifically because of unsatisfying gameplay, it is nonetheless a reminder that NFTs don’t have value without a functioning game. And one imagines that if the game was doing well — lots of users, lots of engagement — that it would’ve been unlikely to shut down in the first place. The people who own the F1 brand aren’t arbitrary and capricious and wouldn’t end a project that was making them good money.)

What are people actually buying when they buy land in the Sandbox or spaceships in Star Atlas? They’re buying a lottery ticket, a speculative bet on the future value of the assets in the game or app. Just because a game sells millions of dollars of NFTs before it launches does not, in any way, guarantee its future success. It might make that future success more likely — NFTs trading for record valuations will get a lot of press and a lot of people talking about a game and that attention may lead to more players — but it is marketing, not product that people are buying into. If the app isn’t launched yet, the only utility of those assets is their speculative value — the ability for the owner to sell it for more later — since by definition you can’t use them in the app itself yet.

Pre-launch NFT sales don’t confer any information about that future product’s growth, retention, engagement, or resulting future value creation. All the company has proven is that it can sell tokens and a vision of a future state — that it’s very good at storytelling, but not that (a) it can ship the product or that (b) when the product ships, it’s something people actually want. In other words: the project has proven it can raise money, but not that what it’s building is actually useful.

Being good at marketing an idea and being good at executing that idea are two very different things. There are many stories of companies that did exceptional marketing but never delivered a useable product, or whose products, when delivered, greatly underperformed expectations: Juicero, Clinkle, Theranos, Coin, Quibi, and many more litter the graveyard of startups with grand ambitions, massive funding, and nothing to show for it. People wanted to believe in them, but they didn’t deliver on their promises. Those who pre-ordered a Coin card saw the product demo video and thought they wanted the product, but when it finally arrived it turned out not to be as useful as they thought. When you were a child, did you ever get really excited by a toy commercial, begged your parents to buy you the toy, and then when you got it, barely played with it? That’s exactly what can happen here. Beware indexing heavily on NFT pre-sales as indicators of consumer demand for an end product. It just might be a mirage.