ERC721A received much attention as part of Azuki’s launch as they dramatically reduced batch mint prices (reduction of ~70-90% for batches between 10-100 NFTs). Don’t get me wrong, it was a pretty nifty trick. It sounded like magic. But is it really all that?

Some people are probably already aware, but the implementation optimizes for cheap txn costs during minting but stores ownership data in an inefficient manner, drastically increasing the costs of transfers (an increase of 182.85%). Any integration with an ERC721A that includes transfers or on-chain ownership checks (eg. staking, buying, selling, sending to a friend) costs nearly 3 times as much as a normal ERC721.

ERC721A is gud for

1. Whale minters
2. Insidooors who wish to mint large batches for cheap
3. Offloading txn costs to NFT buyers in the secondary market
4. Distributing batches of soul-bound NFTs (not meant to be transferred)
5. Burning ETH

ERC721A is bad for

1. NFTs with on-chain integrations or utility
2. Secondary market buyers’ pockets in the long run
3. Reputation as a long-term buildooor

Conclusion

Using ERC721A doesn’t necessarily mean that an NFT collection is bad, or will perform badly in terms of price. But it should prompt you to realize that one of these is probably true:

1. The team does not care about ecosystem’ expenses in the long run
2. The team did not use a technically competent implementer
3. The team values short-term batch minting over long-term user savings
4. The team does not plan on building/supporting on-chain utility in the NFTs

Here’s the sauce, where I compared ERC721A vs solmate’s ERC721: