There's a very interesting phenomenon. You can tell at a glance that many project data are just for fundraising purposes—hardly anyone is actually using them.

The storage project Walrus recently caught my attention for one reason: the usage traces it leaves on-chain are a bit too authentic.

Let's look at the numbers first. During a recent testing cycle, the Walrus network wrote over 3 million data objects. Note, this isn't the number of transactions—it's the actual number of stored data objects. What does this indicate? It's not just a "try it out" phase; there are real projects continuously and repeatedly writing data into it.

Now, look at the second data point. Currently, there are over 1,000 active storage nodes, and their distribution isn't concentrated in a single region. For a storage protocol, this node structure is already resilient to risks. If it were just about incentivizing nodes, such a distribution wouldn't be possible.

The third number involves costs. Public test data show that this storage solution's unit cost has decreased by more than an order of magnitude compared to traditional on-chain solutions. What does that mean? It means that data which previously wasn't worth storing on-chain long-term now makes sense.

Putting these three data points together: node scale can support → data is genuinely being written → cost structure supports long-term use.

This is no longer a story of "what might happen in the future," but something that's already happening. Many are still waiting to hear the story. But the real signal has long been written on-chain.