Question about technical audits and execution latency
Hey community,
Quick question for those who regularly test their execution environments. I have been trying to clean up my setup lately and move away from platforms that rely on dealing desks or heavy sales floor pressure.
When you are auditing a new software provider or infrastructure setup, what are the exact technical metrics you look at? I know people talk about stp and dma routing and sub 35 ms speeds, but I am curious how you actually verify things like banking segregation or whether a system is truly automated without human intervention.
I have been looking at how a few independent setups handle this, for example checking technical breakdowns
What kind of testing process do you use for latency and order fills before trusting a new provider with live funds? Would appreciate any practical tips.
I’d separate the evaluation into two areas: execution quality and custody/regulatory risk, because they’re often treated as the same thing when they’re not.
For execution, I’d look beyond claims like “sub-35 ms.” I’d measure actual order-request-to-fill time, spread, slippage, requotes, rejected orders, and execution during high-volatility periods like CPI, NFP, and major market opens. A low ping during quiet market conditions doesn’t necessarily mean good execution when liquidity gets thin.
For STP/DMA, I’d verify the legal entity, execution policy, liquidity arrangements, and regulatory registration directly rather than relying on the provider’s website.
For segregation, I’d confirm which legal entity holds client funds and check the applicable regulator’s rules on client-money protection.
