BMLL and Exponential launch order-book flow analytics for US equities

BMLL Technologies and Exponential Technology have launched XTech US Equity Flow, a dataset that decodes full-depth US equity order-book data into net buying and selling activity separated by investor type. The product combines BMLL's Level 3 data across all US equity venues with Exponential's proprietary inference methods, built on over 25 years of high-frequency and systematic trading experience, to produce granular, near real-time signals for quant funds, discretionary asset managers, hedge funds, and sell-side desks.

The launch represents a notable escalation in the data arms race that defines modern capital markets. Where most market data products rely on consolidated top-of-book feeds, XTech US Equity Flow reconstructs the full order book, decomposing activity into institutional, market-maker, retail, and high-frequency trader flows on a per-exchange and per-ticker basis. Coverage extends back to January 2020, giving systematic funds six years of history for back-testing, while the product also offers T+1 and 15-minute delayed delivery at one-minute, hourly, daily, and weekly intervals.

Decoding the order book

Morgan Slade, chief executive and head of research at Exponential Technology, said institutional flow decoded from order-book data explains roughly 80% of open-to-close price moves in US equities and reliably anticipates directional performance. That is a strong claim, and one the company says is grounded in published, out-of-sample research validated across multiple market regimes. The product also incorporates an agentic research workflow layer and interactive dashboards, extending access beyond quant teams to discretionary portfolio managers who want to monitor flow discrepancies without writing code.

For discretionary managers in particular, the value proposition centres on distinguishing temporary market impact from genuine informational repricing. When a stock moves sharply, knowing whether that move is a large institution mechanically pushing price as it executes, rather than a reaction to new fundamental information, materially alters the optimal response. That kind of microstructure intelligence has historically been available only to the most well-resourced players; packaging it via a dashboard product lowers the barrier of entry across the buy side.

Capital markets data as infrastructure

The deal is also a read-across for how capital markets data is evolving as an asset class in its own right. BMLL was acquired by Nordic Capital in October 2025 and has raised more than $80m in successive funding rounds since its 2014 founding in Cambridge's machine learning laboratories. Exponential Technology, founded in Chicago in March 2024, is considerably younger but brings a specialist pedigree in market microstructure. The partnership structure, in which Exponential's inference layer sits on top of BMLL's raw data foundation, is increasingly common: data infrastructure firms monetise their pipes by enabling specialist analytics vendors to build higher-margin products on top.

The macro significance extends beyond the two firms. As equity market structure grows more complex and fragmented across venues, the informational advantage of full order-book reconstruction widens. Sovereign wealth vehicles and large multi-asset managers allocating across US equities, volatility, and macro strategies are increasingly demanding microstructure intelligence to understand whether price signals are durable. Regulators, too, have an interest: BMLL already counts exchanges and regulatory bodies among its clients, and the same Level 3 data that powers alpha generation also underpins compliance and market surveillance. The convergence of alpha-seeking and regulatory-use-case applications in a single dataset is commercially powerful and creates a degree of resilience that pure alpha-product vendors lack.

Exponential says it intends to launch additional products beyond XTech US Equity Flow in due course, with the implied roadmap pointing toward non-US equity venues and its existing global macro forecasting capabilities, which it claims call CPI direction correctly 81.9% of the time, ahead of professional consensus.