On-Chain Data Sources: Where the Numbers Actually Come From

Every chart, metric, or alert based on on chain data crypto comes from information recorded on a blockchain. Validators add blocks, nodes store the records, and different software tools process that information before displaying it. But the numbers you see aren't always the same across platforms. Two tools looking at the same blockchain may report different figures depending on how they collect and process the data. That's why it's worth knowing where the data comes from before relying on it.

Key summary

• On chain data crypto begins as raw blocks and transactions recorded on a public ledger, then passes through nodes, indexers and APIs before it reaches a dashboard.

• The raw record is the same for everyone. The numbers differ because of how each provider decodes, labels and aggregates it.

• Common sources include your own node, a node provider, a block explorer, an indexing service and a ready-made analytics platform.

• Most errors in on chain data crypto come from definitions, delays and mislabelled addresses, not from the chain itself.

• SpotX reads public chain data across Ethereum, Solana, Base and Hyperliquid, scores wallet clusters, and attaches the transaction hash to every alert so you can check the source.

What Is On Chain Data Crypto?

On chain data crypto is the information recorded directly on a blockchain. That includes every transaction, every wallet balance, every token transfer, every smart contract call and every block. Because public blockchains are open ledgers, anyone can read it. Nobody grants access, and nobody can quietly edit it afterwards.

That openness is the whole appeal. A bank statement is private and a company's internal sales figure is a claim you have to trust. On chain data crypto is different because you can check it. If a source tells you that a wallet moved a large sum, there is a transaction hash behind the statement, and you can look it up yourself.

What the raw record does not provide is convenience. A chain is stored in the order events happened, and it stores data in compact, machine-friendly forms. Turning that into "this wallet bought this token for this amount" takes work. Our guide to what on-chain data is covers the basics, and this article follows the numbers one step further back, to their origin.

Step One: The Chain Itself

The first source of any on chain data crypto figure is the blockchain's own record. Validators or miners produce blocks. Each block holds a batch of transactions plus a few details such as a timestamp and a reference to the previous block. Once enough blocks sit on top of it, a block is considered final.

Not every blockchain records the same kind of information. Ethereum tracks balances and smart contract activity through its account-based system. Other chains work with unspent transaction outputs instead. There are also networks designed to process more transactions and layers that rely on a larger blockchain for settlement. These differences affect what data you can easily find and what requires more work to retrieve.

This is also where timing enters. A block is recorded before it is final, and on some chains a recent block can be replaced by a competing one. Any provider of on chain data crypto has to decide how long to wait before treating something as settled. Wait longer and the data is safer. Wait less and it is faster. That trade-off sits underneath every real-time number you see.

Step Two: Nodes

A node is a computer that stores a copy of the blockchain and keeps it updated as the network changes. It gives you access to data directly from the blockchain rather than through another service. If you need on chain data crypto without relying on third-party providers, you can run your own node and get the information straight from it.

There are a few kinds. A full node keeps recent state and checks every rule. An archive node keeps the complete history of past states, which is what you need to ask "what was this wallet's balance two years ago?" Light clients keep less and trust others for the rest.

Running your own node gives you control and privacy. It also gives you chores: storage that keeps growing, sync time, upgrades and monitoring. Plenty of teams decide it is not worth the effort and use a provider instead. We compare both routes in blockchain API vs running your own node.

Step Three: RPC Providers and APIs

Most developers and analysts use a provider to access blockchain data instead of running their own nodes. The provider takes care of the infrastructure and gives them an RPC endpoint where they can request information from the blockchain.

This is the point where on chain data crypto first becomes someone else's service. You trust the provider to be synced, to be available and to return honest results. A provider that falls behind the chain by a few blocks will hand you slightly stale data without any visible warning.

APIs layer more on top. Instead of asking for a raw block, you might ask for the token transfers for an address, and the provider returns a ready-made list. That convenience comes from indexing, which is the next step. If you are weighing the options, our crypto data API guide and guide to choosing a crypto data API go through the trade-offs.

Step Four: Indexers

Nodes aren't designed to handle large searches efficiently. Finding every transaction linked to a particular address can mean going through a huge amount of blockchain data. An indexer makes this easier by reading the records, organising them in a database, and keeping them available for quicker searches.

Indexers also decode. A smart contract call arrives as a string of bytes. The indexer uses the contract's published interface to translate that into a named function and readable parameters. That is how a swap shows up as "swapped token A for token B" instead of a block of hex.

Most of the on chain data crypto people actually see has passed through an indexer. Block explorers rely on them. Analytics dashboards rely on them. The quality of the decoding, and how fast the indexer keeps up with new blocks, decides how reliable the final number is. Our blockchain explorer guide shows what the output of one looks like.

Step Five: Labels, Clusters and Analytics

Even after blockchain data has been properly indexed, it only shows wallet addresses, not who owns them. Data providers try to identify the people or organisations behind these addresses. They label wallets they believe are linked to exchanges, funds, protocols, or other known groups. Addresses that seem to be managed by the same person or organisation are also grouped into clusters.

This layer is where on chain data crypto turns into analysis, and it is also where the most judgment enters. Labels come from public disclosures, behaviour patterns and heuristics. Some are strong, and some are educated guesses. Two providers can label the same address differently, or one can label it and the other can leave it blank.

Analytics platforms then compute metrics on top: active addresses, exchange flows, holder concentration, whale activity. Every one of those metrics needs a definition. Is a transfer between two addresses owned by the same exchange a "flow"? Different providers answer differently, and that is why their charts rarely match exactly. For a wider view of the field, see what blockchain analytics is.

Where On Chain Data Crypto Numbers Come From: A Quick Map

Read from top to bottom, each layer is a little further from the raw record and a little more interpretive. That is not a flaw. It is the price of making the data usable. The point is to know which layer you are trusting.

A Worked Example (Illustrative)

Take a made-up situation to see how this plays out. The numbers are invented for illustration only.

Two dashboards both report "exchange inflows" for the same token over the same day. One shows an illustrative 12,000 tokens and the other shows 9,500. Neither is lying. The chain is the same. The difference comes from the layers above it.

One dashboard tracks transfers to exchange addresses using both confirmed labels and inferred ones. The other only counts addresses that have been publicly confirmed. They also count internal transfers differently. The first includes transactions between an exchange's own wallets, but the second leaves them out. So, even though both use the same on chain data crypt, the numbers don't always match.

Before relying on either number, review a few of the transaction hashes used to calculate it. Public blockchain records allow you to check the underlying transactions and see what supports the reported figure.

Costs and Limits of Each Source

Your own node. Maximum control, but real running costs and upkeep. You still need an indexer if you want broad queries.

Node providers. Easy and quick to start. You rely on their uptime and accuracy, and heavy use can hit limits or costs.

Explorers. Free and convenient for lookups. They are manual and read-only, with no alerting built for you.

Indexing services. Fast structured queries. You depend on their decoding and the chains they support.

Analytics platforms. Meaning and metrics ready to use. You inherit their labels and definitions, and you cannot always see how they were produced.

Off-chain data. Some figures sit partly off the chain, such as order books on centralised venues or prices from oracles. Mixing them with on chain data crypto is common, and it should be labelled when it happens.

Five Questions to Ask About Any Data Source

1. Where does this number start? Ask which chain, which node and which indexer produced it.

2. How fresh is it? Find out how many blocks behind the source can run.

3. How were the addresses labelled? If there's no explanation for a label, it's hard to know how reliable it is.

4. What exactly does the metric measure? Figures such as flows and active addresses can differ depending on how they're calculated.

5. Can you check the transaction hash? If you can't trace the number back to the blockchain, be careful about relying on it.

Common Myths About On-Chain Data

"On-chain data is always accurate." The ledger is accurate. The tools that read, decode and label it can be late, wrong or limited. Accuracy at the source does not guarantee accuracy at the screen.

"All providers show the same numbers." They read the same chain but make different choices about labels, definitions and timing.

"On-chain means everything is visible." Only what is recorded on the chain is visible. Off-chain trades, private agreements and activity on centralised exchanges are outside it.

"Anonymous means untraceable." Addresses are pseudonymous. Patterns can still be analysed, and activity is permanently public, which is why on chain data crypto can be so revealing.

"More data means better insight." A flood of raw activity is noise. Filtering and scoring turn it into something you can use.

Why Different Chains Give Different Data

On chain data crypto looks different depending on where it comes from. Ethereum has a long history and a large ecosystem of tools. Solana moves at high speed, which makes freshness harder and volume heavier. Base is a layer that settles onto a larger chain, so its data has a relationship to another ledger. Hyperliquid is built around trading, and its design means position and order activity can be more visible than on many other venues. Our research on how perp venues leak position changes looks at that last point in detail.

For anyone reading across chains, this matters. A metric built for one chain does not carry over to another unchanged, and comparing numbers across them needs care about what each one measures.

When You Can Skip the Deep Dive

You do not always need to trace every number to its source.

• You are checking one transaction or balance. A block explorer is plenty.

• You follow high-level trends for general interest and nothing rides on exact figures.

• You use a provider you already trust and cross-check key claims occasionally.

• You hold long term and act rarely on short-term chain activity.

If you build, trade or report on on chain data crypto and a decision depends on the figure, though, knowing the source is worth the effort.

Where SpotX Fits In

SpotX uses public on chain data crypto from Ethereum, Solana, Base and Hyperliquid to track wallet activity. It groups related addresses into wallet clusters and checks each cluster through a seven-stage scoring process. Clusters scoring below 65 are removed before publication, so you only receive alerts that pass the filter.

Alerts are sent through Telegram, Discord or webhook, with the wallet cluster, score and on-chain transaction hash included. Delivery takes under 30 seconds on Trader, under 25 seconds on Pro and under 20 seconds on Desk, which sends JSON through webhooks. API and webhook access are available from the Pro tier onward.

Because the hash is in every alert, you can follow the same trail described above and verify the source yourself. Our guide to verifying any alert on-chain shows the steps. SpotX is not a wallet, an exchange or a fund, and it does not give trade advice. If you want to see the alerts in practice, there is a 7-day free trial, plus a free weekly email called Signal of the Week.