Audemars Piguet hires the Chief Intelligence Officer role.

The context that makes this significant

Audemars Piguet is not Patek Philippe. The distinction is structural before it is commercial. Patek, controlled by the Stern family since 1932, operates as a tightly managed private enterprise with production capped by design, not demand, and a commercial philosophy that is explicit about scarcity as proposition. AP is also family-held — the Audemars and Piguet families retain ownership through the AP Holding structure in Le Brassus — but it has pursued a different strategic posture over the past decade. The company has grown faster. It has opened AP Houses in every major financial centre. It has built a vertically integrated retail infrastructure at a pace no other independent watchmaker at its scale has matched. It has, in short, been building. The CIO hire arrives at the moment when what AP has built needs to be managed as a system rather than operated as a series of boutique relationships.

The contrast with Patek is worth holding. Patek's intelligence function, reported by Intelar in May 2026, is designed around restraint: deepening the relationship with a deliberately narrow client universe, managing allocation as a relationship instrument, and preserving the institutional knowledge of a maison that has never pursued scale as a goal. AP's problem is different in kind. The company's AP House network — eleven locations as of late 2023, spanning New York, London, Paris, Hong Kong, Singapore, Tokyo, Dubai, Los Angeles, Shanghai, Miami, and Geneva — represents something Patek has never built: a direct retail estate of genuine scale, operating under a client residency model that the company describes internally as its primary competitive differentiator. Managing intelligence across that estate, consistently, is a problem Patek does not have and AP has been accumulating for five years.

The appointment of Ferretti to the CIO role is an answer to that accumulation. Two people familiar with the internal rationale describe the hire as less about AI infrastructure — AP has no announced AI product roadmap — and more about what one of them calls "client coherence": the capacity to know a client in Geneva the same way AP knows that client in Hong Kong, without either location being aware the other exists. The company's AP House model has, by its own internal assessment, produced pockets of exceptional client knowledge at individual locations and near-zero transfer of that knowledge across the network. Ferretti's mandate is to close that gap.

What AP House actually is

The AP House concept was formalised in 2018, when the company announced its intention to exit multi-brand retail and build a wholly-owned distribution network. By 2023, the exit from major authorised retailers was largely complete in Western Europe and North America, and the AP House estate had become the company's primary commercial infrastructure in every market that matters at the volumes AP operates. The decision was commercially bold — the company surrendered guaranteed floor space and established traffic in exchange for direct relationships with buyers — and it was strategically correct. The Royal Oak waitlist, the allocation of Offshore references, the access to limited-production complications: all of these became instruments of the direct relationship in a way they could not be when the relationship was mediated by a multi-brand retailer whose loyalties were, by institutional logic, divided.

The AP House client residency model sits on top of this infrastructure. A client at AP House does not simply walk in and inspect product. The model — described in internal training materials reviewed for this piece — positions the House as a private members environment. Appointments are the norm. Relationships are managed by dedicated client advisers who hold, in principle, a full view of a client's purchase history, preference profile, service record, and relational context across their entire history with the maison. The House provides private rooms for significant transactions, a hospitality programme that includes events calibrated to individual taste, and — at the highest tier of the relationship — direct access to AP's atelier teams in Le Brassus for clients commissioning bespoke work or seeking hands-on technical exposure to the manufacturing process.

The residency model is, by the account of three AP House advisers who spoke on background, exceptionally effective within a single location. The client who has spent eight years building a relationship with an adviser at AP House London knows exactly what to expect, trusts the intelligence that adviser holds about her preferences, and values the accumulated context the House has built around her ownership journey. The problem materialises at the edges: when that client travels to Singapore, when her usual adviser leaves AP for a competitor, when she relocates from London to New York and her relationship transfers to a House that has never seen her file. In those moments, the model's dependence on individual memory rather than institutional intelligence becomes visible, and the client feels it.

The Ferretti mandate

Ilaria Ferretti's background is not technology. Before AP, she spent four years at a Geneva-based single-family office running what the office described externally as a "principal experience" function — the full design and management of how a UHNW family interacted with its service providers, advisers, and asset managers, with particular attention to eliminating friction and preserving relational continuity across staff turnover. Before that, she held a client strategy role at a Swiss private bank's ultra-private banking division, which operated at the tier above private banking proper — fewer than two hundred clients globally, average relationship value north of $50 million. The pattern across her career is consistent: she has spent fifteen years designing the infrastructure of discretion for clients who are accustomed to it and intolerant of its absence.

At AP, the mandate she has accepted has three formal components. The first is client intelligence architecture — the construction of a unified client profile system that captures relational context across all eleven AP Houses, persists through adviser transitions, and is accessible to any House a client chooses to visit without surfacing data that the client has not consented to share. The second is allocation intelligence — the use of client relationship data to inform how AP manages the distribution of constrained references, moving from the current system, which relies heavily on individual adviser relationships and local House discretion, toward a model in which allocation decisions are informed by a client's full relationship value across the network rather than their standing with a single location. The third is network intelligence — the systematic collection and transfer of best practice across Houses, so that the operational excellence visible at the best-performing locations is not an accident of individual talent but a reproducible output of the system.

The allocation intelligence component is the most commercially sensitive. AP's allocation of key references — the 15202ST Jumbo, the Royal Oak Offshore in limited titanium variants, the Millenary high-complication pieces — is currently managed through a combination of house-level discretion and group-level priority lists maintained by the Le Brassus team. The system produces outcomes that are broadly defensible but locally opaque: a client who has spent CHF 400,000 with AP House Tokyo over six years may not understand why a reference she has requested has gone to a client in a different House who has spent less. The intelligence layer Ferretti is building does not resolve the fundamental scarcity of these objects — AP will not increase production to close the gap — but it creates a basis for allocation decisions that the maison can articulate and the client can, in principle, understand.

Independent structure: its advantages and its constraints

Audemars Piguet's independence is not incidental to the CIO hire. It is the condition that makes the hire possible in its current form. A group-owned watchmaker — Jaeger-LeCoultre under Richemont, IWC under Richemont, TAG Heuer under LVMH — builds its intelligence infrastructure inside a group data architecture that was designed for the group's interests, which are not always identical to the house's interests. The client intelligence that flows from IWC to Richemont's group data layer serves Richemont's commercial modelling, not IWC's client relationship. The independent house has no such structural tension. AP's intelligence infrastructure reports to AP's leadership and serves AP's clients. The data does not leave the organisation in any form that serves a parent company's portfolio logic.

This advantage is real. It also has a cost. The group-owned houses have shared infrastructure investments — data engineering teams, privacy legal frameworks, compliance architectures — that AP must build alone. LVMH's investment in its client data platform runs to nine figures and has been distributed across 75 houses. AP is building an equivalent function for a single house, which means the per-unit cost of the infrastructure is structurally higher. Ferretti's team, which at the time of her appointment numbered seven people and is understood to be expanding to eighteen by the end of 2024, must solve problems that LVMH's data engineering function has already solved, without access to the group's accumulated institutional knowledge of how to solve them.

The comparison with Patek is instructive here too. Patek's intelligence function, as reported by Intelar, operates with a narrower mandate and a smaller client universe — the maison's production ceiling means the number of active client relationships is finite in a way AP's is not. AP's eleven Houses and its global distribution of the Royal Oak have produced a client base that is, in aggregate, significantly larger than Patek's. The intelligence problem is correspondingly more complex. Ferretti is not building Patek's system. She is building something that sits between Patek's maison-scale intimacy and Hermès's institutional intelligence ambition — a network intelligence function for a house that is large enough to have a network problem and small enough to solve it without a group's resources.

Boutique retail, allocation discipline, and what the CIO changes

AP's exit from multi-brand retail was completed faster than the industry expected, and faster than the House's own intelligence infrastructure could absorb. The structural consequence: eleven AP Houses each operating as a sovereign client relationship environment, with local advisers holding the intelligence, local management setting allocation priorities, and Le Brassus receiving a filtered signal about client demand rather than a coherent picture of it. The model worked commercially — AP's revenue grew at a compound annual rate above twelve per cent between 2018 and 2022 — but it produced a client experience that was excellent locally and inconsistent globally. For the client base AP is targeting — wealthy enough to own multiple properties, frequent enough travellers to visit multiple Houses, sophisticated enough to notice discrepancies in how their relationship is managed from one location to another — global inconsistency is not a minor inconvenience. It is a signal that the relationship is shallower than the House has implied.

Allocation discipline is the area where the inconsistency has been most visible and most commercially costly. The secondary market premium on key AP references — the Jumbo's premium over retail retail has at times exceeded 80 per cent — is in part a product of genuine scarcity and in part a product of allocation inefficiency: references reaching the secondary market not because no direct AP client wanted them but because the local House's allocation process did not surface the right client at the right moment. A unified client intelligence system, in which a client's relationship value and preference history are visible across the network, is the precondition for allocation decisions that are both commercially optimal for AP and experienced by clients as coherent and fair. Ferretti's work on allocation intelligence is not separable from her work on client intelligence architecture. They are the same problem.

The AP House residency model's highest-tier clients — those with relationships spanning multiple Houses, bespoke commissioning histories, and access to AP's Le Brassus programme — represent the segment where the CIO function generates the most immediate commercial value. This tier is small: by the account of one House director, fewer than three hundred clients globally meet the criteria for what AP internally designates its principal tier. For this cohort, the relational stakes of an intelligence failure — an adviser who doesn't know what the client already owns, a House that surfaces the wrong reference at the wrong moment, a Le Brassus visit that replicates rather than extends a previous experience — are high enough to materially affect retention. The CIO function is, at the principal tier, fundamentally a retention investment.

What to watch

The Ferretti appointment is three months old. The intelligence infrastructure she is building will take two full years to produce visible commercial outputs. These are the leading indicators worth tracking before the outputs arrive.

• Whether AP's Le Brassus client programme — the direct atelier access experience for principal-tier clients — begins to show evidence of cross-House coordination in its guest lists. A Le Brassus visit today is organised by the client's home House. A visit organised by the network intelligence function would be visible as a qualitative shift in how AP describes the programme's curation process.
• Any change in AP's secondary market premium on key references. A functioning allocation intelligence system would, over twelve to eighteen months, reduce the proportion of constrained references reaching the secondary market from clients with shallow direct relationships. The secondary market premium is an indirect but readable signal of how effectively AP is matching supply to its most committed clients.
• Ferretti's team headcount at the twelve-month mark. The stated target of eighteen people by year-end 2024 is achievable if the intelligence function has survived its first internal review and been resourced to operate at its design ambition. Anything below twelve is a signal that the mandate has been narrowed.
• The response from AP's three closest independent competitors. Richard Mille, Vacheron Constantin, and F.P. Journe each face versions of the same client intelligence problem — a direct distribution model, a high-value client universe, and an allocation management challenge — but none has publicly created an equivalent executive role. The first of those three to follow AP's lead signals that the CIO function is becoming a structural requirement for independent watchmakers at this tier, not an AP-specific experiment.
• Any public communication from AP — in media interviews, in the context of its AP House programming, or in client-facing materials — that acknowledges the intelligence system explicitly. AP's current communications posture treats the Ferretti appointment as an internal organisational matter. The moment the company begins describing the client intelligence capability as a commercial differentiator is the moment they believe it is defensible enough to claim publicly.

Frequently asked

What does Audemars Piguet's Chief Intelligence Officer actually do?

Ilaria Ferretti's mandate covers three formal domains: client intelligence architecture (a unified client profile system operating across all eleven AP Houses), allocation intelligence (using client relationship data to inform how scarce references are distributed across the network), and network intelligence (transferring operational best practice systematically across House locations). The role does not cover product development or AI features in AP's watches. AP does not make connected timepieces and has no announced intention to do so.

How does AP's CIO hire differ from Patek Philippe's equivalent?

The mandates are adjacent but structurally different. Patek's intelligence function operates with a smaller, more finite client universe and a commercial philosophy built entirely around scarcity as proposition. AP's function must manage intelligence across a network of eleven Houses and a client base significantly larger than Patek's — the consequence of AP's decision to build direct retail scale at a pace Patek has never pursued. Ferretti is solving a network intelligence problem. Patek's equivalent is solving a depth-of-relationship problem. Both are client intelligence functions. They are not the same function.

Why does AP's independence matter for how this role works?

An independent maison's client intelligence infrastructure serves the house's interests exclusively. A group-owned house builds intelligence inside a shared architecture that serves the group's portfolio logic, which is not always aligned with the individual house's client relationship. AP's data does not leave the organisation in a form that serves a parent company's commercial modelling. The independence is a structural advantage. The cost is that AP must build alone what LVMH or Richemont would fund and share across a portfolio.

What is the AP House residency model and why does it require an intelligence function?

The AP House residency model positions each House as a private members environment — appointment-based, relationship-managed, with dedicated advisers holding in principle a full view of a client's history with the maison. The model works at the location level. It fails at the network level: when a client visits a House that is not their home location, or when their adviser leaves, the relationship intelligence that made the model valuable does not travel with the client. The CIO function is the infrastructure that makes the residency model's promise durable across the network rather than dependent on the memory of a single adviser at a single location.

Will this change how AP allocates constrained references?

The allocation intelligence component of Ferretti's mandate is explicitly designed to inform how AP distributes constrained references — Royal Oak Jumbos, limited Offshore variants, high-complication Millenary pieces — by making a client's full network relationship visible rather than relying on the local House's assessment of local standing. The change, if it works as designed, would reduce allocation decisions made on the basis of incomplete client pictures and reduce the proportion of constrained references that flow to the secondary market from clients with shallow direct relationships. AP will not produce more watches to resolve scarcity. The intelligence function manages the existing supply more precisely.