Heather O'Gorman's fintech has to be able to say, at the end of every single day, exactly how much client money it holds — with almost no buffer allowed.
Most of the conversation around instant payments treats speed as the solved problem. JPMorgan's own liquidity specialists argue the harder problem is the opposite: money now moves instantly, and Treasury teams still can't always see it, trust it, or act on it fast enough.
"We didn't need a faster car. What we wanted was intelligence about the journey itself."
Amy Eckhoff and Ross Webster jointly run JPMorgan's global liquidity product solutions and commercialization teams; Heather O'Gorman runs treasury, cash management and safeguarding at NEOM, a cross-border e-money firm that has to answer to regulators for every client dollar it holds.
I listened to the full episode so you can skip it. 50 minutes of audio, 16 minutes of reading.
Here are the 8 insights that matter.
👤 Guests: Amy Eckhoff, Global Head of Liquidity Product Solutions Specialists and Global Head of Liquidity and Accounts Commercialization at JPMorgan; Ross Webster, Global Co-Head of Liquidity and Accounts Product Solutions Specialists, Payments, at JPMorgan; and Heather O'Gorman, Director of Treasury, Cash Management and Safeguarding at NEOM
🎙️ Host: Benjamin Ensor, Director of Research and Strategy at 11:FS
📰 Published: 9 September 2026, on YouTube (Fintech Insider Podcast by 11:FS)
🔴 YouTube | ⏱️ 50 min | ✅ Time saved: 39 min
Key Takeaways
Payments now settle instantly, but most Treasury operating models are still built around batch cycles and business hours — so the payment lands before anyone can confirm it did
Eckhoff says the real constraint isn't liquidity itself, it's visibility and trust in the position
Treasury teams deliberately keep cash sitting idle over weekends, "just in case," because funding processes aren't confident enough to fund on demand
This is what the industry calls trapped cash
Local instant payments work well almost everywhere now; cross-border payments still mostly run on SWIFT and can take days
O'Gorman says that gap — instant locally, slow across borders — is the actual crux of the problem
NEOM has to safeguard client money to the penny, with almost no buffer, and report its position at the end of every day
New safeguarding regimes are arriving in Canada, Australia and the UK (CASS 15), tightening the requirement further
Real-time Treasury is an operating model built over time, not a product a bank sells
Webster: simplification, not more technology, is what he calls the real innovation
Banks now offer currency-specific notional pooling so e-money firms can safeguard client funds in the right currency without running an FX conversion in the middle
Eckhoff says this is one of the more sophisticated tools banks use to help non-bank firms meet safeguarding rules in real time
More payment rails — embedded finance, card networks, stablecoins — make cash management harder before automation makes it easier
Webster: the same technology that creates the fragmentation can also consolidate it
A stablecoin is, in O'Gorman's framing, just another 24/7 cross-border settlement rail, not a separate category of problem
New wrinkles: converting between fiat and non-fiat, and new treasury risk tolerances
1. Payments Move, Not Treasury
Host Benjamin Ensor framed the episode's core tension at the outset: businesses now operate around the clock, but the systems meant to manage their cash often don't.
The mismatch is structural, not incidental. Payments increasingly settle in real time across borders, but the liquidity behind them can sit across multiple accounts, entities and currencies, with information arriving on different schedules
Eckhoff said the payment now happens before Treasury can confirm it. Money is moving "very very quickly," but many operating models still run on batch cycles, cut-off times and business hours, so the payment lands first and the understanding of the resulting cash position comes later
The actual bottleneck isn't liquidity — it's confidence in the data. Eckhoff said the shift means Treasury has to answer a harder question than before, and that "the constraint that we see now isn't really about liquidity. It's can we see it? Can we trust the position and can we move the liquidity safely with the right controls to exactly where and when we need it?"
Ensor summarized it as a data and visibility problem dressed up as a liquidity problem, and Eckhoff agreed
2. The Friday Cash Buffer
Eckhoff broke the friction into specific mechanical causes, not just a vague sense that things move faster now.
Fragmentation is the first source of friction. Client cash spread across many accounts and systems makes it genuinely hard to maintain one reliable view of positions and forecast needs
The second is funding mechanics in an always-on settlement world. When the settlement layer runs 24/7 but the rails and internal processes behind it don't, treasurers end up pre-funding accounts and carrying buffers rather than trusting they can fund on demand
Those buffers pile up specifically over weekends and holidays — what the industry calls trapped cash. Teams hold a "just in case" cushion in local accounts from Friday through Monday because they're worried something could hit while systems are less responsive. That cash is safe, Eckhoff said, but idle
Solving it is where automation, particularly threshold-based sweeping, starts to matter — a theme the conversation returned to later
3. Cross-Border's Still Slow
Heather O'Gorman, whose firm NEOM moves money across borders on behalf of other financial institutions, gave the practitioner's version of the same problem.
Paying in and out locally in real time is now genuinely easy in most jurisdictions. There's been a real push toward 24/7 domestic payment rails in recent years, and O'Gorman said that part of the problem is largely solved
Cross-border movement hasn't kept pace. She said plainly that "it's still difficult to move money across borders. It's still difficult to get your money from one country to another. You know, we're still very reliant on a quite dated old process via Swift, etc. That can take days."
The mismatch between local speed and cross-border friction is the actual crux of the problem. Money can be received and paid out instantly in the US, but making sure the same amount is simultaneously available to pay out in, say, Singapore still requires the liquidity-forecasting buffers Eckhoff described
Webster added that large clients increasingly have real mismatches between where cash comes in — by currency, country and time — and where it needs to go out to suppliers and employees
4. From Speed to Control
Webster said the nature of what clients ask for has shifted, and reached for an analogy about how driving directions have changed.
The ask itself has changed. "The conversation has fundamentally shifted. A few years ago, the ask was simple. Move my money faster." Now clients ask him to help them understand where their money sits across every entity, currency and time zone, and to act on that information in real time
His analogy: paper maps versus live GPS navigation. People used to plan a route before leaving the house; now navigation reroutes dynamically around traffic and closures. "We didn't need a faster car. What we wanted was intelligence about the journey itself" — which he said is exactly what's happening with Treasury
For non-bank financial institutions specifically, the ask is increasingly orchestration: a single view of global cash, the ability to sweep or concentrate liquidity dynamically, and the intelligence to anticipate funding needs before they turn urgent — across every rail, currency and operating window, not just domestic payments
5. A Model, Not a Product
Ensor asked what "real-time Treasury" actually means in practice, and Webster drew a line between two eras.
Webster splits Treasury's evolution into two chapters: speed, then control. "I think the first chapter as we mentioned was about speed and now we're moving into the chapter of control." His framing to clients: "simplification is innovation" — the most innovative move for many treasurers is removing complexity, not adding a new rail, API or platform
Faster payments solved one problem and created new ones. If money can leave an account instantly, does Treasury have the visibility to know what that means for its position? If funds arrive 24/7, is the business actually capturing and deploying that liquidity, or is it sitting unseen until the next business day?
Webster's warning: don't put the technology before the problem. "Someone says realtime Treasury and immediately the conversation jumps to API, instant rails, AI dashboard." Most treasurers don't need everything in real time — they need better decisions, faster, with less manual effort, and sometimes that means better forecasting or fewer accounts rather than more speed
Eckhoff agreed you can't bolt a faster rail onto an old operating model and expect a real-time outcome. Real-time Treasury, in her framing, is real-time visibility plus decisioning rules plus automated execution plus controls that hold up around the clock — a staged journey, not a switch flipped overnight
Webster's summary became the episode's clearest line: "real time Treasury isn't a product that you buy. It's an operating model that you build"
6. Safeguarding, to the Penny
The conversation turned to how regulatory safeguarding obligations shape what a non-bank firm like NEOM can actually do with liquidity.
NEOM has to safeguard client funds to the penny, with essentially no buffer. "We are expected to be able to say at the end of every day this is how much we're holding for our clients and this is how much we then have in our safeguarded bank accounts"
That precision requirement forces pre-funding. Making instant cross-border payouts work while also meeting a strict, penny-accurate safeguarding position means carrying operational liquidity specifically to bridge the gap
Safeguarding regimes are tightening globally. O'Gorman named Canada's new payments-firm license, a forthcoming equivalent in Australia, and the UK's CASS 15 rules as recent examples. For NEOM, safeguarding is, in her words, "the first part of my title" — the priority ahead of speed
Diversifying banking partners is a related discipline, not optional. However strong a single bank's coverage might be, O'Gorman said spreading relationships across banks is both a safeguarding necessity and ordinary treasury logic — in short, not keeping every basket of funds with one provider
Eckhoff described the products banks now offer to help. Currency-specific notional pooling lets e-money firms hold client-money accounts and safeguard funds in the right currency without an FX conversion sitting in the middle of the process, helping non-bank movers meet real-time safeguarding obligations more cheaply
7. Data Needs a Driver
Webster pivoted to data with a motorsport analogy, and Eckhoff distilled it into a phrase of her own.
Formula 1 cars generate thousands of data points a race — tire temperature, fuel load, hundreds of sensors. He said the team that wins isn't the one with the most data — it's the one that can translate that data into an actionable outcome, whether that's pitting now or adjusting the wing angle
Raw visibility alone is overwhelming, not useful. Webster said handing a treasurer a list of every real-time transaction across 200 accounts in 30 currencies would be overwhelming rather than helpful; what matters is turning that into intelligent, actionable signals
Eckhoff's caveat: visibility without the right real-time infrastructure to act on it is just a faster way of seeing the same problem. "You can see your problem quickly but you need to be able to act on that quickly as well"
O'Gorman added the practitioner's version of the data problem: getting it from banks reliably enough to use. Choosing banking partners partly comes down to how frequently and accurately they can supply the data NEOM needs to know its safeguarding position and reconcile client funds correctly
8. Just Another Rail
Late in the conversation, Webster raised how the growing number of payment rails — embedded finance, card networks, stablecoins — changes the Treasury workload, and O'Gorman placed stablecoins in context.
More rails make cash management harder before automation makes it easier. Webster said every new payment rail or embedded-finance flow is another source of liquidity movement Treasury has to track, reconcile and manage, and institutions risk ending up with more fragmentation, not less
The same technology causing the complexity can also absorb it. Virtual account structures can consolidate what would otherwise need dozens of physical accounts, and automated liquidity management can respond to inflows and outflows regardless of which rail they arrived on
O'Gorman's framing: a stablecoin is just another 24/7 cross-border settlement rail, not a new category. "The fact is a stable coin is also just a 247 cross-border rail that can be used to drive settlements"
What's genuinely new with stablecoins is the conversion layer. She flagged liquidity fragmentation and the mechanics of converting between fiat and non-fiat forms of money as considerations Treasury teams haven't had to manage with traditional real-time payment rails
Bonus Insights
Virtual accounts are one of the practical building blocks of simplification, in Eckhoff's account. They give granular visibility by entity, business line or currency without requiring a separate physical account for every slice, and a virtual-account hierarchy enables continuous, real-time cash pooling rather than periodic sweeping. Her line: visibility without the ability to mobilize the cash is really just "an expensive dashboard"
Webster's second analogy was a car stereo. Real-time payments, virtual accounts, notional pooling and API connectivity are increasingly table stakes — not individually a reason to choose a bank, but their absence can be a reason to walk away, the way a missing stereo might sink an otherwise fine car
O'Gorman weighs four questions every day: are client funds protected, is money where it needs to be, is the business earning what it should on idle cash, and what is it all costing to hold and move? Answering all four consistently, she said, is genuinely difficult and often involves trade-offs between the priorities
Eckhoff said AI and machine learning have already been used in Treasury forecasting for close to a decade, and that better data tagging is what's making those models understand nuance and spike days more accurately — data quality, not the AI itself, is the limiting factor
Eckhoff's test for genuine Treasury modernization versus "a better dashboard": less trapped liquidity, fewer manual interventions, and humans setting the strategy and risk guardrails while technology executes within them automatically
The panel's shared bottom line is that instant payments solved the speed problem and immediately created a harder one — visibility, trust and control over liquidity — and that no single rail, product or dashboard closes that gap; only a redesigned operating model does.
Products, Companies & Tools Mentioned
JPMorgan (Eckhoff and Webster's employer; its liquidity, payments and notional-pooling products for corporate and non-bank clients ran through the whole conversation)
NEOM (O'Gorman's employer, a cross-border e-money fintech whose safeguarding obligations and liquidity constraints framed much of the discussion)
11:FS (Ensor's firm and the studio behind Fintech Insider)
SWIFT (The cross-border payment messaging network O'Gorman called a "dated old process" that can take days, in contrast to instant domestic rails)
If this was worth your time, send it to someone who has to have a view on this.
Get the latest market chatter as it happens: