The FIX Trading Community has unveiled new recommended practices for algorithm testing and certification, aimed at enhancing the robustness and consistency of algorithmic trading.
The recommended practices build on regulations already in place to for algorithm testing and certification, and are set to meet the growing demand for standardised, electronic mechanisms to enable these capabilities and communicate certification information to counterparties and regulators.
Specifically, the new practices span workflows and message descriptions for defining system configurations, running tests and communicating the results and approving suitable algorithms for production usage.
Jim Kaye, FIX executive director, said: “Robust regulations exist to mandate the testing and certification of algorithms, but until now there has been a lack of consistency in how this requirement is met and communicated.
“FIX’s algorithmic trading working group has worked through this in detail to define best practices, and guide firms in implementing these.”
Read more – ESMA issues algorithmic trading-focused briefing
By implementing these new recommended practices, FIX aims to ensure that algo certificates include standardised meaningful data, not only to meet legal requirements but also to facilitate stronger information to be stored in in-house algorithm inventory systems.
This data spans: precise definitions of algorithms beyond identifiers, comprehensive documentation of performed test and clear reporting of results.
The recommended practices complement growing demand and interest across the algorithmic trading community for granular venue-level data and sophisticated routing logic analysis to inform smarter execution decisions.
As shown in The TRADE’s recent Algorithmic Trading Survey – Long-only, ‘data on venue and order routing logic or analysis’ saw one of the biggest increases behind reasons for algorithmic usage, up from 3.77% of responses in 2025 to 4.69% in 2026.
These findings indicate that an increasing number of buy-side firms are looking into their algos to ask why they behave in a certain way, and FIX’s new guidelines support this trend, by ensuring standardisation of the documentation and certification for algorithms, their parameters and testing scenarios.
Read more – Beyond the Data: A new era of buy-side expectations in algorithmic trading
The importance of data analysis and performance for algorithmic trading was also echoed by Eric Brown, FX trader at T. Rowe Price at the TradeTechFX US event in Miami in February this year.
Brown asserted: “We may send through a handful of trades a week in the algo space and that just doesn’t get you accurate sample size to be able to draw significant conclusions […] I think data is kind of the holy grail for the future – allowing traders to have consistent data in front of them to be able to know this is the best time to use an algo based on the current conditions.”