Managing Payout Reliability with Multiple PSPs for High-Volume FX

Hey everyone, wanted to get some perspectives on managing payout reliability, particularly in the high-volume FX space where we're moving significant notional daily. We've been diversifying our payment service providers over the last year to mitigate single-point-of-failure risk, but it's introduced new complexities. Specifically, we're seeing inconsistent clearing times and occasional holds, even with supposedly 'prime' channels. The friction in dispute resolution and the KYC/KYB overhead for each new integration is also becoming a significant drag.

My question is, for those running similar setups, how do you actively monitor and benchmark your PSPs for actual payout reliability and speed? Beyond just the advertised SLAs, are there any specific internal metrics or red flags you watch for that indicate an underlying issue before it impacts liquidity? We're trying to move beyond reactive problem-solving to a more predictive model, especially as market volatility continues to pick up. Any insights on optimizing this infrastructure would be greatly appreciated.