Scott O’Brien, head of product, Americas at Liquidnet
Algorithmic trading has long been a cornerstone of equities markets, arguably one of the earliest and most impactful applications of AI in trading. But as benchmark strategies like VWAP and POV have become standard-issue tools, often relegated to “set it and forget it” duties, the challenge now is to rise above the noise. With the commoditisation of these strategies, we’re left with a question: has innovation in this space run its course, or is there room to reimagine what smart execution really looks like?
At Liquidnet, our goal is to move beyond standardisation and deliver differentiated solutions that leverage our unique liquidity access, intelligent models and configurability to better meet client execution objectives.
The commoditisation of benchmark algos and limitations
Originally designed to quietly trade along with the market, schedule-based algorithms helped traders offload the easier orders so they could stay focused on more complex names. In the US and Canada, they’ve picked up an additional job: paying the research bill.
Their commoditisation is no mystery. They’re widely available and largely standardised. They’re used passively. And because they all use similar logic to track a benchmark, it’s difficult to tell one from another in terms of execution quality.
At Liquidnet, we’ve seen this first-hand across the buy-side. Traders want more than benchmark matching, they want differentiated execution that adds value.
That’s why our approach to schedule-based algos is different. By tapping the unique liquidity in our buy-side network, we offer execution that is anything but standard.
Our blotter-syncing capabilities with the largest asset managers allow us to source unique liquidity, both in block size and smaller flow. This liquidity can be woven into the algo’s schedule or used opportunistically to get ahead of schedule when the conditions are right. Members can configure how and when they tap into that liquidity, depending on their goals. The result: natural, clean fills. Less toxicity, lower slippage.
Moving beyond benchmarks – The future of algo trading
What’s driving the next phase of evolution? The short answer: complexity. Markets are more dynamic, fragmented, and data-heavy. Traders want smarter tools that help them make better decisions, not just automated ones.
At Liquidnet, we’ve built an algo suite that integrates real-time market signals and historical trading patterns to inform decisions on the fly, adapting to volatility, liquidity shifts and regime changes in real time.
When volatility picks up, as it has in the past few weeks with tariff announcements, timing risk and adverse selection become more pressing concerns. That’s exactly where our models outperform and shine.
We use a fair value model that dynamically adjusts the price on the order for the block-seeking portion based on where the market is going. If volatility spikes, we can pull back. We’ve also introduced mechanisms we call “cooldown” and “breathing logic” to give the system a beat to reassess before stepping back into the market.
Differentiating modern execution strategies
Liquidity-seeking algos come in many shapes and speeds. Some are cautious, others aggressive. Some lean on a deterministic schedule, some are more opportunistic. Yet the differences between liquidity-seeking algos can be as stark as they are subtle, and they matter.
At Liquidnet, this is our home turf. We’ve built our capabilities to combine wide venue connectivity, unique liquidity and data science that adapts in real time.
Case in point: our next-generation liquidity-seek algo, Barracuda. Relaunched in the US in March and globally in the coming months, it’s delivering results that have outpaced expectations.
Members can use the algos out of the box based on urgency, leveraging the defaults our quant research has determined. Or they can fine-tune their execution strategy through a wide array of customisation options. Some examples are configuring periodic sweeps, inverting price sensitivity for momentum, and choosing which fills count toward your configurable minimum and maximum participation rates.
Barracuda also features a stock-specific minimum quantity model. Not a blunt tool, but one tuned by historical data at the instrument level. Since launch, slippage against arrival has improved while participation has gone up.
We’re bullish on venue connectivity but selective about when and how we tap it. On a normal day, we may have a curated list of venues for each regime. But if the algo is falling behind or needs to cross the spread, we can expand the universe. That includes lit markets, grey markets, ATSs, and even single-dealer platforms. The goal is simple: access to all liquidity, but only when it makes sense.
The impact and benefits of intelligent algorithmic execution
Better execution comes from smarter tools but also expertise. Our approach is not just about automation. It’s about giving traders tools that respond to intent, to market context, and to changing conditions.
Our experienced coverage team works with our Members across strategy design, real-time monitoring and intraday execution decisions, especially during volatile sessions.
We have intraday TCA that allows our traders and execution consultants to monitor performance, talk to our Members, and adjust course as conditions evolve. Our data and analytics platform gives our coverage desk market color and context our Members rely on to stay ahead. And our algo ranking model helps simulate how a stock might behave under different scenarios.
Ultimately, our goal is to make the trading experience as seamless and as smart as possible. In calm markets or volatile ones, it’s about models that do the right thing at the right time, and coverage teams ready to guide and adjust.
In a commoditised algo space, differentiation isn’t just about speed or spread. It’s about liquidity, intelligence, adaptability, and partnership.