Chief Consumer & AI Acceleration Officer — How to structure a department that reinvents Consumer Intelligence with AI while superpowering the entire organization.
+7.8% like-for-like. Fragrance & Fashion 72%, Makeup 13%, Skincare 11%. IPO May 2024 at €14B valuation.
EMEA 55% (€2.75B) · Americas 35% (€1.76B) · APAC 11% (€530M, +21.7% — fastest growing)
Beauty & Fashion (Albesa, Deputy CEO) · Charlotte Tilbury (Pinsent, CEO) · Derma (Toulemonde, President)
150+ countries, 32 direct offices. Family-owned (~80%), post-IPO governance with 11/13 independent directors.
The current scope document defines a Chief Consumer & AI Acceleration Officer role that inherits two very different mandates: (1) a Consumer department that needs reinvention, and (2) an AI Acceleration agenda that must transform the entire company. These are not the same thing, and pretending they are is the first structural mistake to avoid.
You're being asked to simultaneously run a department (consumer intelligence, media, CRM) AND change the company (AI transformation). These require different skills, different speeds, different metrics, and different political capital. The department architecture must acknowledge this duality, not paper over it.
Placing the consumer as the organizing principle for AI transformation is the correct strategic instinct. AI without consumer centricity becomes technology for technology's sake.
Deploy → Redesign → Reinvent is a sound maturity model. Not every function needs to be "reinvented" — some just need better tools (deploy), others need process redesign.
The ambition to superpower brands AND markets AND operations is correct for a C-suite role. Without enterprise scope, this becomes another marketing analytics team.
The document lists ~15 responsibilities without clear sequencing. When everything is a priority, nothing is. Day 1 looks identical to Month 18 — this guarantees dilution of focus and political vulnerability.
Who are the customers of this department? Brand GMs? Regional directors? The CEO? Without defined internal clients, the team becomes a service desk that responds to the loudest voice.
The document merges "consumer understanding" with "AI tool deployment" as if they're the same capability. A consumer insights expert and an ML engineer are fundamentally different profiles. The org must accommodate both.
No clarity on reporting lines, team size, budget authority, or how decisions get made. Who owns the data? Who decides which AI project to prioritize? Who resolves conflicts between brands?
Ongoing consumer insights, media performance, CRM execution, dashboards, weekly reports. Brands depend on this daily. If it breaks, trust evaporates immediately.
~70% of team capacity needed here
AI transformation, process redesign, new tools, capability building. This is where the future lives — but it doesn't produce immediate ROI and competes for the same people.
~30% of team capacity — but where the strategic value is
One team, one methodology, one data platform. Ensures consistency across brands. Risk: becomes a bottleneck. Brands feel "served" not "empowered."
Each brand has embedded analysts. Faster, more contextual. Risk: fragmentation, data silos, inconsistent quality. What works for Rabanne may not work for Byredo.
Own consumer intelligence, media, CRM. Clear scope. Team knows what they do. Risk: AI Acceleration becomes a side project, not an enterprise transformation.
Transform the company with AI. Superpower every function. Risk: consumer intelligence becomes a "use case" rather than the core, and you lose the heartbeat of the role.
The answer is not to pick one side. It's to design an architecture where consumer intelligence IS the engine that powers AI transformation. Consumer understanding is not separate from AI — it's the data, the context, and the "so what" that makes AI valuable. The department must be structured so that every AI initiative starts and ends with the consumer.
This is NOT a P&L-owning business unit. It is a Global Support Function — like Finance, HR, or Legal — that exists to make Brands and Markets smarter, faster, and more effective. The department does not own media budgets, does not own IT infrastructure, and does not own the AI platform. It owns intelligence, frameworks, playbooks, and the strategic demand for AI that the rest of the organization executes.
| ✅ YOU OWN | 🤝 YOU ENABLE (but don't own) |
|---|---|
| Consumer intelligence & insights — the "single source of truth" for all consumer understanding across the group | Media budgets — owned by brands and markets. You define HOW to allocate, WHAT the KPIs are, and WHICH playbooks to follow |
| Frameworks & methodologies — how we measure media, how we segment consumers, how we attribute ROI | IT/Data infrastructure — the platform, the data lake, the CDP. IT builds it. You define what it must contain and how it serves business needs |
| Agency governance — how we manage the relationship with media and creative agencies. Performance standards, contract frameworks, review cadence | Brand-level execution — brands run their campaigns. You provide the intelligence, the archetype playbooks, and the performance benchmarks |
| Consumer activation playbooks — the brand × archetype × channel frameworks that turn insight into action | AI/ML engineering — IT/Data builds the models. You define the business requirements, the use cases, and what "success" looks like |
| AI acceleration strategy — the WHAT and WHY of AI across the enterprise. Which processes to redesign, which business models to reinvent | Market-level decisions — regional teams decide local execution. You provide the global standards, tools, and insights they use |
| KPIs & measurement — defining what success looks like for consumer engagement, media ROI, and AI adoption | Product development — R&D owns formulation. You provide consumer insights, trend intelligence, and whitespace analysis that guides their decisions |
A Global Support Function's power comes from being indispensable, not from controlling budgets. If brands can't make a good media decision without your insights, if markets can't plan a quarter without your forecasting, if the CEO can't brief the board without your intelligence — you have more real power than any budget controller. The strategic goal is to become the nervous system of the organization: not the muscles (brands execute) and not the bones (IT builds), but the intelligence that connects everything and makes it work.
IT, Data & Analytics reports to the COO (Javier Bach). You do not own this team. But you are their #1 internal client for consumer-related AI and data initiatives. This means you must define your relationship as a "Demand → Supply" partnership: you define WHAT is needed (the business requirements), they define HOW to build it (the technical implementation). The risk is that without governance, IT builds things nobody uses, or you wait 6 months for a dashboard that should take 2 weeks.
Don't structure around "Consumer" and "AI" as separate things. Structure around three operational engines that together create a virtuous cycle: KNOW (understand the consumer) → ACT (activate with media & CRM) → TRANSFORM (reinvent how the company works). AI is not a separate pillar — it's the connective tissue that accelerates all three.
The current scope document proposes three AI pillars: Deploy, Redesign, Reinvent. The problem is these are maturity levels, not operational units. You can't hire a "VP of Reinvent" — it's not a job. Instead, structure around what the team actually does every day, and let the maturity model (deploy → redesign → reinvent) be the ambition framework applied to each engine.
Mission: Be the single source of consumer truth for the entire organization.
Mission: Turn intelligence into action through media, CRM, and personalized experiences.
Mission: Superpower the organization with AI, starting from consumer understanding outward.
Engine 3 (TRANSFORM) does not operate in isolation. Every AI initiative must be co-owned by either Engine 1 or Engine 2. The AI team builds the tools — the Consumer and Activation teams define the requirements and measure the outcomes. This prevents the classic failure mode where AI teams build impressive technology that nobody uses.
| Function | What It Delivers | Internal Clients | AI Maturity |
|---|---|---|---|
| Generative Intelligence Engine | The L1-L6 framework: from automated insight orchestration to executive-level proactive intelligence | C-suite, Brand GMs, Regional MDs | Reinvent |
| Market Intelligence | Market sizing, share tracking, competitive positioning (Circana, Euromonitor, Mintel) | Brand Strategy, Finance, M&A | Deploy |
| Social & Cultural Listening | YouScan-powered brand health, sentiment, emerging cultural signals | Brand Comms, PR, Product Dev | Redesign |
| Behavioral Archetypes | 8 consumer archetypes with brand-specific activation playbooks | Marketing, Media, CRM | Redesign |
| Trend Forecasting | Predictive models for ingredient trends, category shifts, whitespace detection | Product Dev, Innovation, Strategy | Reinvent |
| Retailer Data Hub | Harmonized sell-out data from 80% wholesale channel (the Excel chaos → structured) | Sales, Trade Marketing, Finance | Redesign |
| Competitive & Indie Radar | Competitive ad spend tracking, indie brand monitoring, M&A intelligence | Strategy, M&A, Brand GMs | Deploy |
| Price Intelligence | Dynamic pricing insights across channels/geographies (building on China project) | Revenue Mgmt, Sales, Finance | Redesign |
| Function | What It Delivers | Internal Clients | AI Maturity |
|---|---|---|---|
| Paid Media Strategy | Cross-channel media planning, optimization, budget allocation across brands | Brand Marketing, Finance | Redesign |
| Earned Media Intelligence | PR value tracking, influencer impact measurement, editorial coverage analysis | Brand Comms, PR agencies | Deploy |
| Owned Media & D2C | Website optimization, email/SMS performance, app engagement (20% D2C channel) | E-commerce, Brand teams | Redesign |
| Shared Media & Social | Social content strategy, UGC leveraging, community management intelligence | Social teams, Brand Comms | Deploy |
| CRM & First-Party Data | Bloomreach orchestration, audience segmentation, lifecycle marketing (27K+ Rabanne buyers as pilot) | E-commerce, Retail, Loyalty | Redesign |
| Personalization Engine | Audience × Brand × Channel personalization using archetype playbooks | All Marketing teams | Reinvent |
| Attribution & Measurement | Multi-touch attribution, marketing mix modeling, incrementality testing | CMO, Finance, Brand GMs | Reinvent |
| Function | What It Delivers | Internal Clients | AI Maturity |
|---|---|---|---|
| AI Strategy & Roadmap | Enterprise AI vision, prioritization framework, quarterly OKRs, board-level reporting | CEO, ExCom, Board | — |
| DEPLOY Track | Roll out AI tools to existing workflows: copilots, assistants, automation of repetitive tasks | All departments | Deploy |
| REDESIGN Track | Re-engineer processes with AI at the core: insight pipelines, media optimization, supply chain intelligence | Operations, Marketing, Supply | Redesign |
| REINVENT Track | New business models, revenue streams, or capabilities that didn't exist before AI | Strategy, Innovation, M&A | Reinvent |
| Custom GPT Lab | Brand-specific GPTs, consumer-facing AI experiences, internal productivity tools | Brands, HR, Legal, Finance | Redesign |
| AI Platform & Data | Data infrastructure, governance, CDP, AI/ML platform, vendor management | IT, all departments | — |
| Capability Building | AI literacy programs, upskilling paths, change management, AI champions network | HR, all departments | — |
If you can't tell the organization something about the consumer they didn't already know, you have no right to exist.
Why this is #1: Without intelligence, everything else is empty. Media frameworks without consumer insights are just process. AI acceleration without data-driven use cases is just technology theater. The intelligence engine is the department's raison d'être — it's what no one else at Puig does at this level, and it's what makes everything else possible.
What this means concretely:
First, you must unify the fragmented data landscape. Today, consumer data at Puig exists in silos — Circana market data in one team, YouScan social listening in another, CRM data in Bloomreach, retailer sell-out data in 50 different Excel formats. Your first deliverable is a single, harmonized consumer intelligence platform that anyone in the organization can query. This is the L1 (Insight Orchestrator) of the Generative Intelligence framework we defined.
Second, you must produce proactive intelligence, not just reports. The shift from BI to Generative Intelligence means the data talks to you — not just when asked, but proactively. Example: "Rabanne's 18-25 male segment in Germany is showing a -7% decline in repeat purchase rate. This correlates with a competitor launch. Recommended action: increase paid social budget targeting this archetype by 15% for 4 weeks." That's not a dashboard. That's an intelligence engine.
Third, behavioral archetypes must become operational. The 8 consumer archetypes (Status Signaler, Authenticity Seeker, Self-Expressionist, etc.) need to move from theory to activation. Each brand should know which archetypes over-index for them, what content resonates with each, and which media channels reach them most efficiently. This becomes the foundation for every playbook in Engine 2.
Fourth, retailer data harmonization. This is the ugly, unglamorous work that nobody wants to do but everyone needs. 80% of Puig's business is wholesale. That means the most important performance data comes from retailers — in inconsistent formats, with different update cadences, and often 4-8 weeks delayed. Building an automated pipeline that harmonizes top-10 retailer data into a single sell-out view is a Phase 1 deliverable. It won't be sexy but it'll make you indispensable to Sales and Trade Marketing overnight.
Brands own the budget. You own the intelligence that decides how to spend it.
Why this is #2: Media is where the money is. Puig spends hundreds of millions on media annually across brands and markets. Even a 5% improvement in media efficiency — driven by better intelligence, better targeting, better measurement — represents tens of millions in value. This is where you prove ROI fastest.
What this means concretely:
First, build the Hybrid Playbook Matrix. This is the Brand × Archetype × Channel framework where consumer intelligence meets media activation. For each brand, you define: which archetypes matter most, what content resonates with each, which channels are most efficient, and what the optimal media mix looks like. These playbooks become the "operating system" that brands use to plan campaigns. They don't replace creative judgment — they inform it with data.
Second, standardize measurement and attribution. Today, every brand likely measures media differently. Some use last-click, some use agency-provided reports, some don't measure at all. Your mandate is to define a unified measurement framework: what KPIs we track (across all media types: paid, earned, owned, shared), how we attribute value, and how we compare performance across brands and markets. Without this, you can't prove that your intelligence actually improves outcomes.
Third, agency governance. Puig works with multiple agencies across brands and markets. Your role is to define the framework: how do we select agencies? How do we evaluate their performance? What are the contract standards? How do we ensure that insights from Engine 1 actually reach agency teams and inform their work? This is where many companies leak value — the consumer intelligence team and the agency team never talk to each other.
Fourth, CRM & first-party data activation. With 80% wholesale, first-party data is precious and scarce. The Bloomreach implementation, starting with Rabanne's ~27K buyers, is a strategic asset. Your job is to define the CRM strategy: how do we grow the database? How do we segment and personalize? What's the lifecycle marketing framework? This becomes increasingly important as third-party cookies disappear and first-party data becomes the competitive moat.
You don't build the AI. You tell the company where to point it and why.
Why this is #3 (not #1): Because AI acceleration without consumer intelligence is technology without purpose. You need Engine 1 producing intelligence before Engine 3 can define which AI use cases are worth building. That said, this is where the long-term strategic value of the role lives — and it's what makes this a C-suite position rather than a senior director job.
What this means concretely:
First, map the AI opportunity landscape. Across all of Puig — not just consumer/media, but supply chain, product development, finance, HR — identify where AI can create the most value. Use the Deploy → Redesign → Reinvent maturity framework as the lens: some functions just need better tools (Deploy), some need fundamentally reimagined processes (Redesign), and some represent entirely new business capabilities (Reinvent). Prioritize ruthlessly using a value vs. feasibility matrix.
Second, define the "demand backlog" for IT/Data. Your role is to be the voice of the business when IT asks "what should we build next?" Today, without a Chief Consumer & AI Acceleration Officer, IT either builds what the loudest executive asks for, or what the tech team thinks is interesting. You create a prioritized backlog of AI use cases ranked by business impact, consumer value, and feasibility — and you govern the pipeline with IT through the joint AI Steering Committee.
Third, drive adoption and change management. The biggest failure mode in enterprise AI is building tools nobody uses. 95% of AI pilots fail to reach production (MIT study). Your job is to ensure adoption — through training programs, AI champions in each brand/market, clear communication of "what's in it for me" for every stakeholder. The Custom GPT Lab is an excellent vehicle for this: each brand gets AI tools tailored to their needs, branded for their team, built on the consumer intelligence from Engine 1.
Fourth, measure AI impact in business terms. Not in "models deployed" or "accuracy scores" — in revenue, cost savings, speed improvements, and decision quality. Every AI initiative needs a business owner (from a brand or market), a success metric, and a kill-or-scale decision point at 90 days.
A Global Support Function lives or dies by its reputation. No one is obligated to use you — you have to earn it.
Why this is #4 but the most time-sensitive: This isn't a strategic priority — it's a survival priority. In the first 90 days, every Brand GM and Regional VP is asking: "What does this new department do for me?" If the answer is "we're building a 18-month roadmap," they'll check out and find other ways to get what they need. You need 3-5 tangible wins in the first quarter that make skeptics into believers.
What "quick wins" look like in practice:
Win 1 — The Competitive Flash Report (Week 4): Produce a monthly competitive intelligence report that no one at Puig currently has. Combine ad spend data (MediaRadar/Pathmatics), social buzz (YouScan), market share shifts (Circana), and indie brand emerging signals into a single 5-page briefing. Send it to every Brand GM and Regional VP. Make it so good they can't imagine not having it. Cost: almost zero (you're synthesizing existing data). Impact: you become the "intelligence person" in every executive's mind.
Win 2 — One Brand's Media Diagnostic (Week 6-8): Pick the brand with the most receptive GM — probably Rabanne or Carolina Herrera given existing relationships. Do a full media diagnostic: where is money being spent vs. where should it be based on consumer archetype targeting? Find €1-2M in wasted spend or reallocation opportunity. Present it as an offer, not a mandate. If the brand acts on it and sees results, you have your proof point.
Win 3 — One AI Tool Everyone Can Use (Week 8-10): Deploy a consumer intelligence chatbot (using the Custom GPT Lab framework) that lets anyone at Puig query consumer insights in natural language. "What are the top fragrance trends in Gen Z women in the US?" — answered in 30 seconds instead of 3 days. This is the most visible way to demonstrate what AI acceleration means.
Win 4 — Retailer Dashboard for Top 3 Accounts (Week 10-12): Harmonize sell-out data from the top 3 retailers (whoever is easiest to standardize) into a single dashboard. Give Sales teams visibility they've never had. This creates immediate allies in the commercial organization.
You can't do all of this alone. The first 5 hires define the next 5 years.
Why this is #5 but runs in parallel: Hiring takes 3-6 months. You need to start immediately, but the actual impact is delayed. While you're hiring, you execute quick wins with existing team members, consultants (Convexify), and borrowed resources.
The hiring sequence that matters:
Hire 1 — Chief of Staff (Month 1): This is your operational brain. Manages priorities across 3 engines, runs governance cadence, tracks OKRs, shields you from low-value meetings. Profile: ex-strategy consulting (McKinsey/BCG/Bain) with 4-6 years experience, or a high-performing Puig internal who understands the political landscape. This person multiplies your capacity by 3x.
Hire 2 — Head of Consumer Intelligence (Month 1-2): The most critical functional hire. This person builds Engine 1. Profile: 10+ years in consumer insights/analytics in beauty or FMCG. Must understand both qualitative research and data science. Must be comfortable presenting to C-suite. Must have worked with market data providers (Circana, Euromonitor) and social listening tools. Does NOT need to be technical — needs to be strategic and deeply consumer-obsessed.
Hire 3 — Head of Media & Activation Frameworks (Month 2-3): Builds Engine 2. Profile: 10+ years in media strategy, ideally beauty or luxury. Deep understanding of paid/earned/owned/shared. Experience with attribution modeling and agency management. Comfortable defining frameworks that others execute (this is the hardest skill to find — most media people want to run campaigns, not write playbooks).
Hire 4 — AI Acceleration Lead (Month 2-3): Builds Engine 3's demand side. Profile: hybrid profile — understands AI/ML enough to have a credible conversation with IT engineers, but is fundamentally a business strategist. Has led AI transformation programs (not just built models). Comfortable with change management, stakeholder influence, and ROI measurement. Does NOT need to code — needs to translate business needs into technical requirements.
Hire 5 — Senior Data Analyst / Intelligence Analyst (Month 2): Your first "doer" hire while the VPs are still being recruited. This person starts producing the quick wins: competitive reports, media diagnostics, retailer dashboards. Profile: 5-7 years analytics experience, SQL proficiency, visualization skills (Looker/Tableau), beauty/FMCG context preferred but not required.
This org is designed for ~30-35 people in Year 1 growing to ~40-45 by Year 2. It uses a hub-and-spoke model: the hub (your direct reports) sets strategy and builds platforms; the spokes (embedded in brands/functions) translate that into local execution. The AI team is not a separate silo — it's a horizontal accelerator that co-owns projects with the other two engines.
| Role | Why It's Critical | Profile | When |
|---|---|---|---|
| Chief of Staff | You can't run 3 engines, manage politics, and attend every meeting. This person is your operational brain — manages priorities, resolves cross-engine conflicts, tracks OKRs, and protects your calendar. | Ex-consultant (McKinsey/BCG) with 3-5 yrs experience, or internal Puig high-potential with strong analytical and political skills. NOT a pure administrator. | Day 1 |
| VP Consumer Intelligence | This person defines the "brain" of the department. Must translate raw data from 22+ sources into actionable intelligence that changes how brands make decisions. Without this role working, Engine 2 and 3 have nothing to act on. | 10+ yrs in consumer insights/analytics in beauty/luxury/FMCG. Comfortable with data science but primarily a strategist. Understands Circana, Euromonitor, social listening. Has presented to C-suite. | Month 1 |
| VP AI Acceleration | This person is the technical/strategic lead for the AI transformation. Must be credible with IT (speaks their language) AND with business (speaks ROI). Bridges the two worlds. | Background in AI/ML + business transformation. Has built and scaled AI teams (not just models). Understanding of LLMs, agentic AI. Prior experience with enterprise AI programs. Does NOT need to be from beauty. | Month 1-2 |
| VP Consumer Activation & Media | Owns the "action" layer. Must understand paid/earned/owned/shared media across beauty channels, and be comfortable with attribution, personalization, and CRM orchestration. | 10+ yrs in digital marketing/media in beauty or luxury. Hands-on with programmatic, social, CRM platforms (Bloomreach a plus). Data-driven mindset. Agency or brand-side leadership. | Month 2-3 |
| Generative Intelligence Lead | The technical architect of the L1-L6 intelligence engine. This is the person who makes the Analyst → Strategist → Predictor → Playbook Builder pipeline a reality, not a PowerPoint slide. | Senior data scientist / ML engineer with NLP/LLM experience. Has built multi-agent systems or similar AI pipelines. Understands retrieval-augmented generation, fine-tuning, prompt engineering at production scale. | Month 2-3 |
| Internal Client | What You Deliver | Cadence | SLA |
|---|---|---|---|
| Beauty & Fashion Division Rabanne, CH, JPG, Byredo, Penhaligon's, DVN |
Consumer insights, archetype playbooks, media frameworks, competitive intelligence, trend forecasts, retailer data | Ongoing + quarterly deep dives | Ad-hoc requests: 48h · Monthly reports: by 5th · Playbook updates: quarterly |
| Charlotte Tilbury Operates semi-autonomously (Demetra Pinsent, CEO) |
Global consumer intelligence, cross-brand benchmarking, AI tools from Custom GPT Lab, media best practices | Monthly sync + quarterly review | CT has own analytics team — you complement, not duplicate. Focus on cross-brand insights CT can't generate alone. |
| Derma Division Toulemonde — pharmacy channel, different dynamics |
Consumer archetypes for dermo-beauty, pharmacy channel intelligence, ingredient trend forecasting | Monthly sync | Lighter touch — Derma has different dynamics (pharmacy vs. selective). Focus where cross-pollination adds value. |
| Regional Markets EMEA (Pilar Trabal), Americas, APAC |
Market-level consumer intelligence, local competitive analysis, localized playbook adaptations, regional media benchmarks | Bi-monthly regional reviews | Global frameworks, locally adapted. Markets can request custom analysis with 1-week lead time. |
| CEO / ExCom Marc Puig, José Manuel Albesa |
Executive Intelligence briefings, AI progress reports, strategic consumer signals, M&A consumer lens | Quarterly QBR + monthly 1-page flash | Board-ready quality. No more than 5 pages. Insight → implication → recommended action. |
| IT/Data & Analytics Under Javier Bach (COO) |
Prioritized AI use case backlog, business requirements for data products, testing & validation of AI tools, adoption feedback | Weekly priority sync + monthly pipeline review | Requirements delivered in structured brief format. No ambiguous requests. Clear success criteria for every project. |
| Decision | Decides | Consulted | Informed |
|---|---|---|---|
| AI project prioritization (which projects to fund) | Vanita + VP AI Acc. | VP Consumer Intel, VP Activation, IT, Finance | Brand GMs, ExCom |
| Consumer data access & governance | VP Consumer Intel + Legal | VP AI Acc., IT, DPO | All data users |
| Media budget allocation across brands | VP Activation + Brand GMs | VP Consumer Intel, Finance | Vanita, Agency partners |
| New tool/vendor procurement (>€50K) | Vanita | Relevant VP, IT, Procurement | Finance, Legal |
| Brand-specific insight requests | VP Consumer Intel | Brand Connector, Brand GM | Relevant analysts |
| AI deployment to production (any function) | VP AI Acc. + IT CTO | VP owning the use case, Legal | Vanita, affected teams |
Who: 3 VPs + Chief of Staff + Vanita
When: Monday 9:00 AM, 60 min
Purpose: Cross-engine dependencies, resource allocation, blockers, decisions needed. CoS prepares agenda and tracks actions.
Who: Vanita + IT CTO + CFO + CLO + Brand GMs
When: First Thursday of month, 90 min
Purpose: Pipeline review, pilot results, budget decisions, enterprise AI roadmap, risk/compliance, data governance.
Who: Vanita → CEO + ExCom
When: End of quarter, 2 hours
Purpose: OKR results, business impact metrics, strategic course corrections, budget reallocation, next quarter priorities.
70% of capacity goes to "Run the Business" (Engines 1 & 2 core operations). 30% goes to "Change the Business" (Engine 3 + transformation projects in E1 & E2). This ratio is reviewed monthly. The CoS tracks it. If brands are screaming for reports and dashboards, you may temporarily shift to 80/20 — but never below 20% transformation capacity, or the department loses its reason to exist as a C-suite function.
Hire Chief of Staff. Audit inherited team (skills, gaps, morale). Map all existing data sources and tools. Define internal clients and SLAs. Secure budget authority. Meet every Brand GM 1:1. Produce a "State of Consumer Intelligence" diagnostic for the CEO — this is your political license to build.
Hire VP Consumer Intelligence + VP AI Acceleration. Stabilize existing operations (nothing breaks). Launch "quick win" AI deployments (2-3 tools that save time immediately). Harmonize retailer data from top 5 retailers. Begin Generative Intelligence Lead search. Deploy first consumer archetype framework to one brand (Rabanne) as pilot.
Hire VP Consumer Activation & Media. Full teams at ~70% capacity. Launch Generative Intelligence Engine MVP (L1-L3). Implement behavioral archetypes across 3 brands. Begin M&A radar and indie monitoring. First AI Steering Committee meeting. Data governance framework established. Brand connectors embedded in top 3 brands.
Full operational capacity (~36 people). Generative Intelligence Engine at L4-L5. AI-powered media optimization live. Personalization engine deployed. Attribution model operational. First enterprise AI use cases beyond consumer (supply chain, finance, product dev). First Quarterly Business Review with hard ROI numbers.
Full L1-L6 intelligence engine live. Executive Intelligence (permission-based agents for C-suite) operational. Enterprise AI program touching 5+ departments. Custom GPT Lab with 50+ internal tools. First "Reinvent" initiative: entirely new capability that didn't exist before (e.g., real-time dynamic pricing engine, AI-generated consumer research). Department established as indispensable.
| Company | Structure | Key Lesson | Puig Advantage |
|---|---|---|---|
| L'Oréal | Chief Digital & Marketing Officer (Asmita Dubey). 8,000+ digital talents. Centralized GenAI Task Force. 42K employees trained in AI. | Scale requires a massive upskilling program alongside the technical build. "GenAI for All" course was key to adoption. | Puig is smaller → faster to move. Can achieve 100% AI literacy in months, not years. |
| Estée Lauder | Jane Lauder as EVP Enterprise Marketing + CDO. ConsumerIQ (AI agent). 240+ custom GPTs. Trend Studio for forecasting. | Combining consumer intelligence with data officer role creates power. 31% media ROI increase proved the model. | Puig can build ConsumerIQ equivalent from Day 1 (the Generative Intelligence Engine) — EL took 3 years. |
| P&G | Embedded model: central AI Factory + data scientists placed inside business units. 10PB data lake. AI in 80% of business. | Embed, don't centralize. Data scientists inside consumer teams produce 3x more relevant insights than centralized CoE alone. | The brand connector model in our org explicitly borrows this. Each brand gets embedded intelligence. |
| LVMH | Centralized data platform, decentralized brand control. "Quiet Tech" philosophy. MaIA internal chatbot (2M+ monthly requests). | Technology should be invisible to the consumer. Each maison maintains uniqueness while sharing AI capabilities. | Same multi-brand challenge. Puig can adopt "Quiet Tech" — the platform is shared, the expression is brand-unique. |
| Unilever | 5 business groups with centralized data (8PB, 25K pipelines daily). 23K employees AI-trained. 500+ AI capabilities. | Restructuring from matrix to BU model enabled faster AI adoption. Clear accountability = faster implementation. | Puig's brand-centric structure already aligns with this. The 3-engine model adds the cross-cutting intelligence layer Unilever built later. |
Every competitor built their AI and consumer intelligence capabilities sequentially — first data, then analytics, then AI, then integration. Puig has the advantage of building all three engines simultaneously because you're starting from a near-blank slate. This is not a weakness — it's an opportunity to skip two generations of legacy architecture and go directly to the AI-native operating model that competitors are spending billions to migrate toward.
| Risk | Severity | What Happens | Mitigation |
|---|---|---|---|
| Political resistance from brands | HIGH | Brand GMs see centralized intelligence as loss of control. They resist sharing data, refuse to use archetype frameworks, go around the department to agencies. | Brand connectors embedded in their teams (not parachuted in). Quick wins in first 90 days. Let brands co-own playbook design. Make the value undeniable before asking for compliance. |
| IT territory conflict | HIGH | CTO sees AI Acceleration as encroachment. Blocks data access, delays infrastructure requests, insists all AI goes through IT pipeline (6-month approval cycles). | Early alignment with CTO on "decision rights" (you own strategy + use cases, IT owns production infrastructure). Joint AI Steering Committee. Make IT a hero, not a victim. |
| IT/Data becomes a bottleneck | HIGH | They don't prioritize your requests, AI projects stall | Weekly sync + Joint AI Steering Committee with COO's authority. Escalate early. Build enough internal analytics capability to not be 100% dependent. |
| Talent scarcity | MEDIUM | Can't hire VP AI Acceleration or Gen Intelligence Lead fast enough. Department stalls at Phase 1. AI ambitions remain PowerPoint. | Use Convexify/consultants to bridge gap. Interim fractional leaders. Hire for potential + upskill. Don't wait for perfect candidates — hire 80% match and complement with external expertise. |
| Data quality crisis | HIGH | Retailer data is 80% wholesale in Excel. CRM has 27K records. Market data is subscription-dependent. AI models garbage-in, garbage-out. | Retailer Data Hub is Phase 1 priority. Accept "good enough" data initially. Use AI to clean and harmonize (one of Engine 3's first projects). Don't wait for perfect data to start. |
| Scope creep / "department for everything" | MEDIUM | Every department wants AI help. You become the AI help desk. Consumer intelligence work gets crowded out by "can you build me a chatbot?" | The 70/30 rule. CoS enforces it. Clear intake process for AI requests (triage → pilot → scale). "No" is a complete sentence. Use the AI Steering Committee as political cover. |
| EU AI Act compliance | MEDIUM | Consumer-facing AI (personalization, pricing) triggers regulatory requirements. Fines, reputation risk, project delays. | Legal on Data Governance Committee from Day 1. Privacy-by-design in every AI project. AI ethics framework before deployment, not after. |
| Brands resist the "Global Support" model | HIGH | They see you as overhead, not value | Quick wins in first 90 days. Make it pull (they come to you) not push (you impose on them). Let brands co-design playbooks. Measure and publish the ROI you generate. |
| Charlotte Tilbury opt-out | MEDIUM | CT operates semi-autonomously and may not engage | Offer CT unique value they can't generate alone: cross-brand benchmarking, competitive intelligence beyond beauty, AI tools from Custom GPT Lab. Don't try to own their analytics — complement it. |
| Political conflict with Deputy CEO's scope | MEDIUM | Albesa (B&F President) may see consumer intelligence as "his" territory | Position as enabling his brands, not competing. Early 1:1 alignment. Invite him to co-sponsor the first archetype pilot. Make him look good with intelligence he couldn't access before. |
This department exists to answer one question better than anyone in the beauty industry: "What does the consumer want, and how do we deliver it — faster, smarter, and more personally than any competitor?"
AI is not the mission. Consumer understanding is the mission. AI is the engine that makes it possible at a speed and scale that was previously unimaginable. Every hire, every tool, every meeting, every budget decision should be tested against this question. If it doesn't serve the consumer mission, it doesn't belong in this department.