The Operational Architect: Bridging the Execution Gap

A conversation with Olivier Lachaud — Founder of WSquare Advisory and creator of KadoPool — on eliminating “operational debt,” what makes cross-border expansion actually work, and why he built an AI-powered web app with his own hands.

By Louise Servoin · 2026-05-18 · 12 min read

## Olivier Lachaud – Founder of WSquare Advisory and Operating Advisor for Scaling Companies

Embedded operating advisor for founders and CEOs of small and mid-size companies — strategy, operating model, and AI adoption. Alumnus of Centrale Nantes, ESSEC Business School, and Thunderbird, with eighteen years in the COO offices of Scotiabank, BNP Paribas, and Deutsche Bank in New York, adept at running transformation initiatives. Based in West Windsor, New Jersey.

Some executives see the vision. Others see the spreadsheet. Rarely do you find a leader who can act as the “translation layer” between a founder’s ambition and the operational reality of actually running a growing business.

Olivier Lachaud’s trajectory is a masterclass in managing complexity. Across nearly twenty years inside the COO offices of BNP Paribas, Scotiabank, and Deutsche Bank, he built the “operating systems” behind some of the most complex regulated businesses in the world — from coordinating a $30M project to close a Manhattan office and relocate 700 staff to Jersey City, to designing global data integrity frameworks that cut error rates by 99%.

Today, through his firm WSquare Advisory, he works as an embedded operating advisor for founders and CEOs across construction, independent education, early-stage tech, and cross-border expansion — including with French regtech Resilis – on their US market entry.

We sat down with Olivier to discuss why “operational debt” is the silent killer of scale, what makes cross-border expansion actually work, and why he built KadoPool — a working AI-powered web app — entirely on his own, in about 90 days, using the same AI tools he advises clients on.

## The Bridge Builder

Olivier, you often describe yourself as an “integrator” and a “translation layer.” When a European company tries to land in the US — as in your work with Resilis — what is usually lost in translation?

The friction is rarely just about the product - although European founders tend to under-estimate how much they need to adjust their product for market entry - but it’s about the infrastructure of trust. A “Head Office Strategy” from Paris or London often underestimates the “operational debt” required to operate in a US market with different regulatory, commercial, and cultural norms. With Resilis, my role is to bridge that gap — taking the founders’ vision and translating it into a disciplined, scalable US operation if indeed we determine that there is a market for their product here.

In my experience inside the COO offices at BNP Paribas and Scotiabank, I saw firsthand how critical it is to align front-office goals with operational execution. Whether you are a global bank or a scaling startup, if your “operating rhythm” isn’t synchronized with the actual rhythm of the local market, you aren’t just creating divergence — you’re losing trust with the people who decide whether to do business with you.

## Eliminating "Operational Debt"

You’ve spent years "fixing" broken processes in high-stakes environments. How do you define "operational debt," and why is it so dangerous for scaling firms?

Operational debt is the friction that accumulates when you prioritize "shipping fast" over structural integrity. It’s the manual workaround that stays in place for three years, or the disconnected data systems that force teams to spend 80% of their time reconciling spreadsheets instead of making decisions.

At Scotiabank, we tackled this head-on by engineering a client data framework across 15 different systems. By semi-automating the synchronization between client data and transactional systems, we didn’t just speed up the cycle from monthly to daily — we reduced coherence breaks by 99%. Yet the process itself was brought to life as a workaround. This is an expression of how large organizations struggle with remaining nimble.

The principle scales down: I’m using the same lens today with a mid-size consulting firm in the construction industry trying to build out its AI capability and create a real data and information infrastructure. You don’t “fix” these problems with a band-aid; you engineer frameworks that set the principles and guardrails to stay true to the strategic objective but deliver a structured disciplined approach to execution so the business can actually scale. Part of the challenge specific to AI is that this is so novel. The first order of business is typically to bring the understanding of what is possible up so the right decision can be made for the client.

## The Data Pulse

When you stood up the data governance function inside BNP Paribas’ Chief Data Office — in effect operating as a COO of that department without the title, growing the team from 5 to 90 staff in a year — what did you learn that still applies in the age of AI?

The tools have changed, but the fundamental problem hasn’t: data chaos. You can’t deploy effective AI — or any of the agent-style use cases people are now excited about — if your underlying data foundations are weak. At BNP, we were running a $25M budget across the US, Canada, and India, and the entire point was to ensure we would treat data as an asset, not an afterthought. This can be a challenge depending on the culture of the organization and levels of comfort with handling data.

The lesson was clear: data is not a technical byproduct; it is a strategic asset. If you treat data governance as a “check-the-box” compliance exercise, you fail. If you treat it as the bedrock of your “operating system,” you unlock the ability to deploy enterprise-wide metrics that can reduce high-priority issues by 30% in a single quarter. And then this becomes the steppingstone to your AI adoption strategy. That logic isn’t fintech-specific. It applies to any company staring at AI adoption without an obvious place to start. The highest leverage is, in my opinion, for SMEs as they have the flexibility to adopt this faster than large incumbents that more than compensates for their lower investment firepower.

## Applied AI: KadoPool

You built KadoPool yourself, end-to-end, using AI tools — in about 90 days. Why do that as the founder of an advisory firm?

Honestly, I built KadoPool to keep my hands on the tools. If I’m going to sit in a room with a CEO and help them think through what AI can and can’t do for their business, I want to be speaking from real builder experience — not slideware. KadoPool started from a very human friction I had experienced personally: the chaotic message threads and social pressure implied by traditional payment boards when organizing group gifts, on top of the inherent trust issue of sending money to someone one barely knows. So I picked a problem I cared about, set a 90-day box, and shipped it end-to-end on my own using the same AI tools I advise clients on. What started as an experiment might turn into a real business opportunity.

Because of my background in regulated entities, I built KadoPool with financial regulations in mind and a “Privacy by Design” philosophy from the start — individual contribution amounts are masked, so social pressure is removed. The app acts as a digital side-kick for organizers, especially in schools and offices, pairing cash payouts with digital keepsakes. The product matters to me, but the broader point is this: the same operator’s instinct that helps a CEO untangle an operating model is what I applied to the build. KadoPool is, in part, proof that I personally practice the AI-adoption methodology I advise clients on.

## The Golden Thread

We often see a gap between what technology can do and what professionals actually adopt. How do we ensure AI becomes a 'co-pilot' that elevates human judgment rather than a tool that encourages intellectual passivity?

The premise of the question is worth pushing back on. The risk of "intellectual passivity" assumes AI replaces thinking. In my experience it does the opposite — it raises the volume of information and decisions a leader has to engage with, because the tool now does the easy 80% in seconds and you're left to spend your time on the hard 20%. I find myself stretched more intellectually now that I work at AI speed, not less. The calculator didn't stop humans from doing math; it moved the work to the parts of math that actually require judgment. AI is the same shift, one level up.

The real challenge isn't intellectual passivity. It's organizational. A Board or a CEO asking "how do I adopt AI" is asking the wrong question. The right question is "how do I build a transformation capability inside my company that can keep absorbing new tools — AI today, whatever comes next in three years — without breaking the operating model?" That's a multi-year effort, and it goes well beyond installing a few copilots. Done well, it changes how people work and often what the business itself does. But honestly this is nothing new. It’s just that this is putting a lot more pressure on the many companies who underinvested in transformation capabilities or did not do this well.

The same principle that guides how I work with clients applies here: the goal of any good tool — including a good advisor — is to leave the people using it more capable on their own, not more dependent. AI should make your team more autonomous, not less. If a CEO frames AI adoption that way from the start, the passivity problem doesn't really show up.

Tags: Operating Models, AI Adoption, US-EU Expansion, Operational Debt, Scotiabank, BNP Paribas, Resilis, KadoPool, SME