- Collaboration enables like-for-like benchmarking of market abuse detection rates
- Provides regulator-ready, explainable evidence trails grounded in real-world reconstructed order book behaviour
BMLL, the leading independent provider of harmonised, historical Level 3, 2 and 1 data and analytics across global equity, ETFs, futures and US equity options, today announced a partnership with Features Analytics, an AI-driven trade surveillance and market integrity analytics provider. The partnership will support Features Analytics’ development and commercialisation of new surveillance benchmarking and integrity analytics built on top of BMLL’s high-fidelity historical order book data.
Bringing market integrity analytics to life on high-precision historical order books
Trade surveillance and market integrity analytics require high-quality historical order book data to reconstruct market and trading behaviour, provide context, measure risk and enable explainability.
Combining BMLL’s historical Level 3, 2 and 1 order book data and Features Analytics’ AI-driven detection approach accelerates the launch of new data products. Specifically, the collaboration enables rapid development of proprietary Features Analytics products designed to address model benchmarking and risk measurement challenges in trade surveillance.
As a result, Features Analytics’ solutions help improve surveillance precision and reduce the operational burden driven by legacy, parameter-heavy approaches, where firms can face extremely high false positives and ongoing calibration costs.
Measuring and benchmarking market abuse detection rates
By developing these products, BMLL and Features Analytics open up a new category of surveillance benchmarking data that enables firms to independently measure how their incumbent surveillance stack is performing, both over time and against market conditions. This enables like-for-like benchmarking of market abuse detection rates across existing solutions, while providing regulator-ready, explainable evidence trails grounded in real-world reconstructed order book behaviour.
Paul Humphrey, Chief Executive Officer of BMLL, said: “Market integrity and surveillance are a natural application layer on top of high-quality historical order book data. This partnership reflects our focus on enabling sophisticated workflows on top of BMLL data, now extending beyond market quality into market integrity and surveillance benchmarking use cases.”
Cristina Soviany, PhD, Chief Executive Officer and Co-founder of Features Analytics, added: “Our mission is to help financial institutions stay ahead of regulatory requirements with our unique eyeDES® AI technology and tools that deliver measurable coverage and accurate detection of market abuse. Access to deep, harmonised, high-quality historical order book data through BMLL’s Activate program supports faster development and validation of the new products we are bringing to market.”
Partnership anchored in Features Analytics’ participation in the ‘BMLL Activate: Data Credits Program’
Features Analytics participated in the ‘BMLL Activate: Data Credits Program’, a time-limited initiative designed to enable qualified partners to build and validate new products on top of BMLL historical order book data with no upfront data licence cost during the build period, and a clear conversion path to longer-term commercial deployment. Under the BMLL Activate program, BMLL provides Features Analytics with a BMLL data allowance (credits) redeemable against defined-scope usage of the BMLL Data Lab (Python research sandbox) and access to the BMLL Data Feed.
