Spencer Lee of MarketAxess on the evolution of automation, liquidity provision and the case for a single trading cockpit.
Creating more effective ways for traders to match orders takes technology, strategic vision and trading experience.
The DESK (TD): How is automation in credit markets moving beyond rules-based RFQ execution?
Spencer Lee (SL): For a long time, automation in credit trading meant one thing: making RFQs faster. You’d send a request, apply a set of rules, and auto-execute based on the responses. That’s no longer where the market is. Traders are now able to execute with tools like Adaptive Auto X (ADX), which is a shift towards intelligent orchestration. Orders don’t just follow predefined logic; they’re dynamically routed across protocols to make or take liquidity simultaneously. And importantly, that’s translating into real beneficial outcomes. Clients are executing tens of millions in a day by breaking trades into smaller components and working them across different liquidity and execution sources. Automation isn’t just about efficiency anymore, it’s about expanding what’s achievable.
TD: There’s been a push from some buy-side traders to engage with the market more passively to optimise bid-ask spreads, does that fit with what you’re seeing?
SL: Absolutely. What used to feel like an outlier is becoming part of everyday execution.
Protocols like Live Markets make it possible to passively interact anonymously with both buy- and sell-side liquidity. Posting interest, accessing axes, and allowing the market to come to you. That wasn’t always widely adopted, but the industry is catching up.
Trader mindset has been a catalyst to this shift. They are no longer thinking in terms of which tool to use but instead they’re thinking in terms of how to combine those tools effectively.
That’s where I think the industry still has work to do. As traders are given more protocol options, this also multiplies the number of decisions needed. The real opportunity is to reduce that complexity and streamline that decision-making process. What I’ve long advocated for is a “smart protocol chooser.” Not just routing orders but suggesting how to approach them in the first place, based on data and probability of success. That’s the ideal state.
TD: Beyond axes and algos, how can clients see matching more directly against liquidity in the platform?
SL: The number of liquidity pools available to traders today has grown significantly—client-to-client matching, dealer axes, all-to-all RFQ, workups, auctions , and more.
Individually, none of these protocols are new. What is new is the ability to access them together, in one execution platform, that is where the real advantage lies.
There’s also been renewed interest in concepts like indications of interest (IOIs) to give clients a way to express intent more discreetly while still finding matches. It’s another example of how liquidity provision and liquidity seeking are starting to converge.
What we’re focused on is surfacing those opportunities at the right moment for each trade. Not burying them in a workflow, but making sure traders see high-quality matches when they matter.
Because in this environment, missing the right opportunity is often the biggest risk a trader may face.
TD: Can the matching process itself be improved?
SL: Yes, definitely, and in two significant ways. First, I believe the surface area for matching has to expand materially. This is really about increasing the number of liquidity pools against which buyside orders or IOIs can be cross referenced. Of course, this has to be done with great care to guard against information leakage and then, additionally, we must also improve the hit rates by qualifying the trading interests.
The next step is about moving beyond the individual bond level, towards matching more at the portfolio level that is based on broader risk attributes Historically, trading has been very CUSIP-specific, traders work on one order at a time, a list or portfolio trade of pre-defined cusips. But that doesn’t always reflect how risk is actually managed. Capturing the broader needs of moving bigger blocks of risk can often mean flexibility to accept a number of different cusips to satisfy a particular need, Trading technology has room to grow in this area and it is something that we work as one of the next areas for innovative growth.
At the same time, when you’re working large blocks of orders, you need better ways to control how those orders interact with the market. That’s where the idea of a trading cockpit comes in.
Instead of navigating multiple tools, traders should be able to: adjust aggressiveness dynamically, monitor progress across multiple orders simultaneously, and shift between providing and taking liquidity as conditions change.
It’s less about executing a single trade and more about managing the posture of an entire book of risk.
TD: How does data fit into making that cockpit concept work in practice?
SL: Data is central to this framework but only if it’s actionable. There’s no shortage of information available: continuous pricing, dealer performance, hit rates, historical outcomes. The challenge isn’t access, it’s cutting through the noise in a timely fashion to ascertain relevance.
If that data sits in a report after the fact, it’s not helping the trader. It needs to be there at the point of decision, guiding how an order is worked.
That’s the real opportunity. MarketAxess has various proprietary datasets that are applied across protocols and delivered to traders when they need it most, at the point of trade. The focus now is making sure that insight is embedded directly into the workflow so traders can make better decisions in real time.
TD: If you could redesign buy-side trading desk, what would you change?
SL: I’d simplify it.
Today, traders are navigating multiple protocols, multiple decisions, multiple workflows. But the direction of travel is clear. The most effective desks will treat liquidity provision as a core capability not an occasional tactic, consolidate access to liquidity through a single interface, and use their own trading data alongside market-wide insights to create a continuous feedback loop. Our goal isn’t more tools, it’s better orchestration of them. Ultimately what traders need isn’t more choice, it’s clarity on what to do, when to do it, and why it works.
©Markets Media Europe 2025
