I’ve been thinking about this for a while. In Web3, there’s this common belief that everything should go on-chain. Every piece of data, every record, everything. It sounds right at first because blockchains are secure, transparent, and can’t be changed. But the more I look into it, the more I realize something simple. Just because we can put everything on-chain doesn’t mean we should.
The main issue is cost. Storing data on-chain is expensive, and it gets worse as things grow. The more data you store, the more gas you pay. If a system has thousands or even millions of users, those costs become huge. At that point, it doesn’t even make sense anymore. The blockchain was never meant to store large amounts of data. It was built to verify things, not to carry all the data itself.
I’ve seen many projects make this mistake. They try to put everything on-chain to look more “decentralized.” But in reality, it just makes the system slow, expensive, and hard to manage. It sounds good in theory, but it doesn’t work well in practice. Eventually, the system becomes inefficient because of its own design.
So I started thinking differently. Instead of asking how to put everything on-chain, the better question is what actually needs to be on-chain. And the answer is simple. You don’t need all the data there. You only need proof that the data is real and hasn’t been changed.
This is where Sign Protocol started to make sense to me. It separates two things clearly: where the data is stored and how the data is verified. The actual data is stored off-chain, like on IPFS, Arweave, or even private storage if needed. On-chain, you only keep a small piece of information, like a hash or a reference, that proves the data is valid.
This approach is simple but powerful. The blockchain stays light and cheap because it’s not carrying heavy data. At the same time, the trust is still there because anyone can verify the data using that on-chain proof. You get the best of both sides without overloading the system.
For example, think about issuing a credential. If you store the whole credential on-chain, it will cost a lot, especially at scale. But with this model, you store the actual credential off-chain and only keep a proof on-chain. Anyone can still verify it, but you avoid unnecessary costs.
Another thing I like is the clarity. You know exactly where your data is and how to access it. There’s no confusion. This is important when you’re dealing with real things like identity, agreements, or official records. You need a system that is clear and reliable, not complicated.
At the same time, not everyone wants to use fully decentralized storage. Some people need control or have rules to follow. Sign Protocol understands that and gives flexibility. You can use decentralized storage or your own system. You’re not forced into one option.
This isn’t just about saving gas fees. It’s about building systems in a smarter way. It’s about understanding what the blockchain is actually good at and using it for that purpose instead of overusing it.
I think the space is slowly moving in this direction. Instead of putting everything on-chain, builders are starting to focus on balance. Use the blockchain for trust and verification, and use other systems for storage. It’s more practical and easier to scale.
For me, the idea is simple now. Not all data belongs on-chain. Putting everything there doesn’t make a project better. It just makes it more expensive and harder to use. I’ve started being more careful about what should go on-chain and what shouldn’t.
Sign Protocol didn’t create this idea, but it applies it in a very clean and useful way. And in a space where things are often overcomplicated, that kind of simple and practical thinking really stands out.