A financial infrastructure data moat is a defensible position built not on proprietary software or faster rails but on exclusive visibility into the flows, spreads, and order behavior that other participants cannot reconstruct from public sources. Bloomberg L.P. generated approximately $13.2B in revenue in 2023, with the Terminal driving the majority of that figure, according to Bloomberg company disclosures and Burton-Taylor International Consulting. The lesson is older than Michael Bloomberg's first Market Master in 1982: in financial infrastructure, whoever sits closest to the flow learns the price first, and learning the price first compounds into a structural advantage that commoditized rails cannot dislodge.
​

This piece applies that lesson to stablecoin infrastructure, where stablecoin total supply reached $315.3B as of June 2026 and transfer volume hit $27.6T in 2024 per Artemis. The question for issuers, rails, and aggregators is identical to the one Reuters faced in 1981 and ICE faced in 2000: who owns the data that the next decade of price discovery will be built on?
​

Bloomberg's Terminal business is often described as a software product, but its durable moat is the dataset and the closed messaging graph that sits underneath. Subscribers pay roughly $31,000 a seat not for the keyboard but for access to telemetry that competitors cannot replicate without years of vendor relationships and exchange agreements.
​

Bloomberg Terminal subscriber count reached approximately 356,000 in 2023 at around $31,000 per seat per year, according to Burton-Taylor International Consulting. The arithmetic gives a Terminal-attributable run rate above $11B, and analyst estimates from the same source place total Bloomberg L.P. revenue near $13.2B for 2023. The Terminal is not the highest-margin charting tool. It is a private cable into dealer runs, indicative quotes, and the chat graph that price-discovers fixed income before any public tape sees the trade.
​

Three properties make that moat durable. First, the data is generated by the users themselves: every IB message, every saved monitor, every RFQ that crosses the system enriches the dataset for the next subscriber. Second, the dataset is non-substitutable: a buy-side desk that leaves the Terminal also leaves the dealer chats where bonds actually trade. Third, the data compounds across asset classes, so the marginal cost of adding swaps coverage on top of an existing rates franchise is near zero. Software competitors can match the front end. They cannot match the back end without owning the flow.
​

Network effects scale with the number of participants, but data moats scale with the volume and granularity of transactions those participants generate. In financial infrastructure, a single institutional desk can produce more pricing telemetry in a week than a million retail users produce in a year, which is why data moats compound faster than headcount-based network effects.
​

The BIS Working Paper 1228 on stablecoin flows makes a related point: the informational value of payment data is concentrated in the institutional tail, not the retail median. Visa's $15.7T of 2024 card volume is a useful benchmark, but stablecoin transfer volume reached $27.6T in 2024 per Artemis State of Stablecoins, and a disproportionate share of that flow is treasury, market-maker, and exchange settlement traffic. Each of those transactions carries pricing information that the broader market cannot see in real time.
​

That asymmetry is the engine of compounding. An aggregator that routes a $50M USDC-to-USDT rebalancing trade observes the live primary-mint spread, the secondary AMM spread, and the cross-rail liquidity depth at the moment of execution. Replaying that observation across thousands of similar trades produces a private reference rate that retail-facing analytics dashboards cannot reconstruct from public mempool data. The moat is not the trade. The moat is the recorded counterfactual: what would have cleared, where, at what cost, against which alternative.
​

Stablecoin infrastructure produces data across four distinct layers: issuer reserves and mint/burn flow, primary-market access pricing, secondary-market orderbook and AMM telemetry, and cross-rail liquidity routing decisions. Each layer is owned by a different participant today, and the participant who consolidates visibility across all four will hold the most defensible position.
​

The layers do not nest cleanly, but they do stack. The Fed FEDS Notes on stablecoin market structure describes a similar stratification when it separates reserve management from circulation behavior. Below is how the layers map to participants in practice.

Issuers own the top layer and have an incentive to keep it private. Exchanges own the third layer and publish enough to attract flow but not enough to lose informational rent. The fourth layer, cross-rail routing decisions, is the youngest and least claimed. It is also the one with the closest structural resemblance to what Bloomberg captured in the 1980s, because it records counterfactuals rather than just executed prices.
​

Mint and burn events are the stablecoin equivalent of primary-market new-issue activity, and the messaging that surrounds them, including authorized participant requests and redemption notices, is structurally similar to the IB chat traffic that anchors Bloomberg's fixed-income franchise. Whoever sits in the path of that messaging learns about supply changes before the secondary market reprices.
​

Circle publishes high-level reserve attestations through its transparency program, and Tether discloses quarterly attestations, but neither publishes intraday mint and burn intent. DeFiLlama shows aggregate supply changes, with USDT at approximately $187.2B and USDC at approximately $75.6B as of June 2026, but the intent data, who requested the mint, against what collateral, with what settlement window, remains private to the issuer and its authorized participants.
​

That intent layer is the closest analogue to Bloomberg's chat franchise. It is high-signal, non-public, and generated by institutional behavior rather than retail order flow. An aggregator that holds neutral relationships with multiple issuers, never taking principal risk and never quoting markets itself, is structurally positioned to observe mint and burn timing across issuers in a way that no single issuer can. That position is what Bloomberg held across dealers in the corporate bond market, and it is the reason its data franchise outlasted every competing terminal.
​

The spread between primary-market mint or redemption pricing and secondary-market exchange pricing is the single most informative data series in stablecoin infrastructure, because it captures real-time arbitrage capacity, reserve liquidity, and authorized-participant behavior in one number. Today no public source publishes it cleanly, which makes it the most valuable proprietary data series available to capture.
​

The IOSCO policy recommendations for stablecoin arrangements explicitly call out the importance of redemption-at-par discipline, and the ECB Macroprudential Bulletin on stablecoins notes that secondary-market price deviations from par are the leading indicator of reserve stress. Yet neither regulator has a real-time feed of the primary-secondary spread, because no neutral party publishes one.
​

This is the gap where a Refinitiv-style reference rate could be built. LSEG Data & Analytics, the former Refinitiv business, generated £6.0B in 2023 per the LSEG Annual Report 2023, largely on the back of reference rates, indices, and price-fixing services that institutions trust precisely because the publisher takes no principal risk in the underlying instruments. A neutral stablecoin aggregator, by virtue of observing both mint-side and secondary-side execution across rails, is the natural home for a primary-secondary spread benchmark. Eco is building toward that position rather than offering it today.
​

When underlying rails commoditize, margins migrate up the stack to the layer that orchestrates execution across them. Cross-rail liquidity intelligence, which rail cleared which order at what landed cost and why an alternative was rejected, is the data series that captures this orchestration value and becomes more defensible as the rails themselves become interchangeable.
​

ICE provides the cleanest historical parallel. ICE generated $9.3B in 2023 revenue, with Fixed Income & Data Services contributing $2.2B, per the ICE 10-K filed February 2024. ICE did not build its data franchise by owning a single exchange. It built it by aggregating execution and post-trade visibility across venues, then selling the resulting reference data back to the institutions that needed best-execution evidence. The exchanges underneath are increasingly fungible. The aggregation layer is not.
​

Stablecoin rails are heading the same direction. As regulatory frameworks like the Clarity for Payment Stablecoins Act mature, the differences between compliant rails will narrow. What will not narrow is the visibility difference between a neutral aggregator that has observed years of cross-rail execution and a single-rail operator that only sees its own flow. The cross-rail dataset is composable across asset pairs, time horizons, and counterparty types in a way that a single rail's dataset cannot be, and composability is the property that drove Bloomberg's expansion from rates to credit to FX to equities.
​

The Aggregator Data Flywheel describes how neutral orchestration produces proprietary execution telemetry, which improves routing quality, which attracts more flow, which deepens the telemetry. The framework is the stablecoin-era analogue to the Bloomberg dealer-chat flywheel and the ICE post-trade flywheel, and it depends strictly on the aggregator remaining neutral rather than becoming a principal participant.
​

The flywheel has four turns. First, flow arrives at the aggregator from institutional users seeking best-execution across rails. Second, every order produces a record of the chosen path, the rejected alternatives, and the landed cost differential. Third, that record feeds a routing model that improves expected execution quality on the next order. Fourth, improved execution quality attracts more flow, including the largest and most informative orders.
​

Critically, the flywheel breaks if the aggregator becomes a market-maker or takes principal risk, because the data ceases to be a neutral observation of the market and becomes a record of the aggregator's own positions. Bloomberg learned this lesson by refusing to operate a dealer book. ICE learned it by separating its exchange businesses from its data and analytics businesses with strict internal controls. A stablecoin aggregator that intends to build a durable data franchise has to make the same choice, and Eco is structured to make that choice by design.
​

The three reference businesses teach different lessons to different participants in the stablecoin stack. Issuers learn that primary-market access data is their most defensible asset and should be priced accordingly. Rails learn that they will commoditize unless they own a unique execution surface. Aggregators learn that neutrality is not a marketing posture but the structural precondition for a data franchise.
​

For issuers, the lesson from Bloomberg's chat franchise is that the institutional messaging layer is worth more than the price tape. Circle and Tether have an option to monetize authorized-participant communication and mint-window scheduling as a data product, in the way that primary dealers monetize new-issue allocation visibility. The Dune stablecoin flow dashboards demonstrate how much public appetite exists for this information, which is itself a signal that the private version commands a premium.
​

For rails, the lesson from ICE is that owning a venue is not enough. ICE acquired the New York Stock Exchange in 2013 and within a decade was earning more from data than from equities trading. A stablecoin rail that does not invest in a differentiated execution surface, whether that is settlement finality, compliance tooling, or programmable conditionality, will find its margins compressed against neutral aggregators that orchestrate across all rails.
​

For aggregators, the lesson from Refinitiv is that the reference-rate business is built on trust, and trust requires the publisher to have no skin in the underlying market. S&P Global Market Intelligence and MSCI follow the same model. An aggregator that aspires to publish a stablecoin reference rate, a cross-rail best-execution benchmark, or a primary-secondary spread index has to commit to neutrality at the corporate level, not just the product level. Eco's positioning as a neutral aggregator and platform is a deliberate alignment with this lesson, and the company is building toward that reference-data role over time rather than claiming it today.
​

Dated statistics in this piece are drawn from primary issuer disclosures, exchange filings, and recognized analytics providers. Bloomberg revenue and Terminal subscriber figures are from Bloomberg L.P. company disclosures and Burton-Taylor International Consulting (March 2024). Stablecoin supply figures are from DeFiLlama as of June 2026. Stablecoin transfer volume is from Artemis State of Stablecoins (February 2025). ICE revenue figures are from the ICE 10-K filed February 2024. LSEG Data and Analytics revenue is from the LSEG Annual Report 2023. Where regulatory framing is cited, the underlying documents from the BIS, the Federal Reserve, IOSCO, and the ECB are linked directly in the body. Figures reflect the most recent full reporting periods available at publication and should be re-verified against issuer and exchange sources before being relied on for institutional decisions.
​