Every blockchain query you make passes through a chain of hands, each one taking a cut before the data reaches your application. By the time you're paying for access, you're not paying for data. You're paying for everyone who inserted themselves between you and the source.

Here's who's collecting.

The Cloud Tax

Indexing services run massive operations on AWS, GCP, or Azure. These providers aren't offering crypto discounts. Every query you make carries the embedded cost of their margins, and those margins are substantial.

This is pure pass-through cost. Indexers don't run their own data centers. They rent compute from hyperscalers and mark it up. You're not paying for infrastructure. You're paying for infrastructure plus a layer of cloud provider profit baked into every request.

The Rebuild Tax

Indexers re-execute and reprocess blockchain data that validators have already processed. They sync nodes, maintain parallel databases, layer query infrastructure on top. It's an entire stack that exists to reconstruct state that already exists at the protocol layer.

And the output is worse. Reconstructed state is never as reliable as state from the source. Validators have authoritative data because they processed the blocks. Indexers have derived data, watching from the outside and rebuilding what they observe. You're paying more for something less reliable.

The economics don't make sense until you realize this isn't about efficiency. It's about who positioned themselves to collect.

The Overhead Tax

Indexers are structured like enterprise SaaS companies because that's how they raised money and how they scale. Sales teams, marketing departments, customer success managers, legal counsel, HR, office space. All of it gets passed to you.

Validators don't carry this weight. They don't need a sales org to expose data they already have. The infrastructure exists. Adding a query interface doesn't require building a company around it.

The Investor Tax

Indexing companies have raised hundreds of millions. Those investors expect returns—not modest returns, but outcomes that justify the capital. Pricing isn't set by cost-plus-margin. It's set by what the market will bear while generating the returns investors showed up for.

Validators operate on different math. Protocol rewards cover their costs. They don't have shareholder expectations shaping what they charge. Serving data is incremental income on infrastructure they've already paid for, not a vehicle for 10x returns.

How the Taxes Stack

Each collector takes a cut, and each cut stacks on the last. Cloud margin, then rebuild margin on top of that, then overhead, then investor-driven pricing on top of everything. By the time you're paying, you're paying for compounded profits, not compounded value.

It gets worse the more successful you are. Indexers use tiered pricing, and those tiers don't favor growth. The more queries you run, the more you pay per query as you hit enterprise tiers. Your traction subsidizes their margins. Product-market fit becomes a cost center.

Go multi-chain and the problem multiplies. Every ecosystem you support means another indexer relationship, another stack of taxes. Support five chains, pay the tax five times.

Want fresher data, lower latency, better uptime? That's a premium tier. But validators have these properties by default—they're at the source, processing blocks in real-time. The "premium" indexers charge extra for is just the baseline for anyone sitting at the source.

And once you've integrated, you're stuck. You've built around their API patterns, their query structures, their data formats. Migration means rewriting code and risking production stability. That lock-in gives indexers pricing power they wouldn't have otherwise.

The Validator-Native Alternative

What if you could skip the whole chain?

Validators are already running the compute to secure the network. They've already processed the blocks. They're already held to protocol-level reliability just to participate. The work is done—they just need a way to expose what they already have.

That's a different cost structure entirely. No duplicate cloud deployments. No reconstructed state—just authoritative data from the source. No enterprise sales org. No investor return expectations baked into pricing. What's left is data served at marginal cost by the people who already did the work.

The scaling math flips too. Multi-chain stops meaning "multiply the middlemen" and starts meaning "tap into each chain's existing validator set." Freshness and reliability stop being upsells. They're just what you get when you're at the source.

For validators, this is a new revenue stream on infrastructure they've already paid for. For builders, it's data access that doesn't eat your runway. For the network, it's capital staying inside the ecosystem instead of flowing out to third parties.

What Changes for Builders

The current cost of blockchain data is a tax on building. Every dollar going to the collectors is a dollar not going to product development, user acquisition, or protocol innovation.

Say you're building a DeFi aggregator. You need price feeds, liquidity data, transaction histories across multiple chains. Under the current model, your data costs scale with usage, and the more users you get, the more you pay. Success becomes expensive. A lot of that payment goes to margin stacking, not actual infrastructure.

Validator-native indexing breaks that cycle. You're paying validators directly for access to data they already maintain. The savings aren't marginal. They're structural.

Think of it like buying produce from a grocery store that sources from distributors who source from farms, versus buying directly at the farm. Both work. One costs a lot more.

The Path Forward

The indexing industry exists because blockchain architecture created a gap between consensus and data availability. Filling that gap required building infrastructure, and infrastructure requires revenue. The current model isn't irrational. It's just expensive.

But architecture can change. When validators become the indexers, the collectors disappear. The read layer and write layer merge. The supply chain collapses.

This isn't about cutting out middlemen on principle. It's about recognizing that the most efficient data infrastructure is the one that doesn't get built twice. Validators already invested in the hard part. Letting them serve data directly is just good engineering.

The money you're not spending on taxes is money you can spend on your actual product. In an industry where most projects are fighting for runway, that might be the difference between shipping and shutting down.

What if you didn't have to pay the tax? We're building that.
X · Telegram · Discord · GitHub ·  shinzo.network