From scattered experimentation to systematic AI adoption across 100 employees
Global Financial Risk Institute
Case Study | 2025
A Global Financial Risk Institute needed a clear view of its AI readiness and a strategy that would work across Legal, Finance, Certification, Marketing, and Member Services. Curiosity was high, but usage was uneven. Trust varied widely. Workflows did not yet support consistent adoption.
The Institute engaged me to assess organizational AI readiness, design a comprehensive adoption strategy, and guide implementation from initial training through sustained practice across all departments.
​
Over 16 weeks, I led a full-cycle transformation combining behavioral research, targeted capability building, and workflow integration to move 100 employees from scattered experimentation to systematic AI use in daily work, delivering measurable impact:
AT A GLANCE
-
21 leadership interviews
-
100 employees trained
-
85 ethnographic interviews
-
AI use increased from 23 percent to 67 percent
-
Cross-functional pilots: Legal, Certification, Finance, Marketing, Customer Service
67%
daily AI use
up from 23% before intervention
30%
time savings
on routine tasks in pilot departments
5 min
contract research
reduced from 30 mins with AI workflows
We observed 3 critical barriers to AI adoption
Uneven
literacy
and readiness
No shared
best practices
and frameworks
Trust and verification
concerns
Our approach
Phase 1: Diagnosed organizational readiness (Weeks 1-2)​​
-
Conducted 21 leadership and AI champion interviews
-
Mapped usage patterns, barriers, and assessed literacy, culture, and regulatory readiness
-
Identified four personas (Pioneers to Cautious Starters)
Outcome
Maturity assessment with four personas and phased roadmap tailored to departmental readiness
Phase 2: Built foundational capability (Weeks 3-6)​​
-
Delivered tiered AI literacy training to 100 employees
-
Taught prompting, intro to GenAI and tool fundamentals
-
Created safe experimentation environment with privacy guidance and engagement of AI Champions
Outcome
Organization-wide literacy foundation and early adopters activated to support peer learning
Phase 3: Understood what works in practice (Weeks 7-10)​​
-
Conducted 85 ethnographic follow-up interviews tracking post-training use, trust calibration, and workflow fit
-
Found mental models predict sustained adoption (84% collaborator vs.12% search-engine mindset)
-
Identified workflow-fit issues and tool access gaps
Outcome
Documented interaction patterns and adoption barriers, identifying high-value workflows for integration
Phase 4: Designed for workflow integration (Weeks 11-14)​​
-
Converted research into role-specific playbooks for all departments
-
Documented successful use cases, prompting patterns, and established clear guidance on AI usage
-
Designed custom GPT agents with clear verification protocols
Outcome
Practical playbooks and agents aligned to real tasks with verification protocols for safe, accurate use
Phase 5: Proved value with targeted pilots (Weeks 15-16)​​
-
Launched pilots across five departments with measurable efficiency gains: Legal (30 min → 5 min contract research), Marketing (60% faster ideation), Customer Service (reduced drafting time)
-
Deployed department-specific solutions: clause-analysis agents (Legal), exam review with bias detection (Certification), FP&A templates (Finance), standardized prompting patterns (Marketing/Service)
-
Documented what AI handles well (retrieval, pattern matching, drafting) vs. what requires human judgment (decision-making, compliance, quality assurance)
Outcome
Measurable efficiency gains with replicable patterns ready to scale
"I was resistant in the past, but once I saw how AI could help me in my job, I went all in."
– VP, Legal & Contracts
My approach to AI transformation
I start with discovery, not tools. I use behavioral research and workflow analysis to understand how people actually work with AI. I combine top-down strategy with bottom-up learning, build psychological safety for experimentation, and design the structures that sustain adoption.
A five step approach
RELATED CASE STUDIES
citizenM Hotels | 2023-2024 | AI/ML Implementation
Led cross-functional initiative applying NLP and behavioral clustering to connect guest sentiment themes with actual booking patterns.
Consolidated 12 months of feedback data, developed journey-based taxonomy, built executive dashboard linking qualitative feedback to business metrics.
​
Analysis uncovered specific correlations: sleep quality emerged as top driver of repeat bookings, while lower sentiment around room experience deterred conversion.
Surprisingly, room controls (despite significant investment) showed little influence on behavior, enabling strategic resource reallocation.
Impact
10% NPS growth · 20% increase in actionable insights · Strategic resource reallocation based on data
McKinsey & Company | 2017-2023 | Enterprise Transformation
Led year-long organizational transformation consolidating five fragmented analytics teams into unified function.
Mapped consultant workflows across data ingestion, transformation, modeling, and insight delivery.
Identified inefficiencies, facilitated leadership workshops, co-designed new operating model.
Unified teams with clear roles and governance, streamlined tooling and workflows, improved discoverability of resources and support.
Leveraged research as driver of organizational strategy.
Impact
$50M+ annual savings · 20% FTE reduction · Improved end-to-end data workflows · Enhanced tool adoption
Marianne drove impact and change professionally and creatively beyond expectations.
– Nana Maanu, Novo Holdings, ex-McKinsey
I saw her grow into an all-around transformation and change management expert in only a few months.
– Brian Bussing, Databricks, ex-McKinsey
GET IN TOUCH
Marianne van Ooij
​Core expertise
Organizational readiness assessment · Change management · Workflow integration · Human-AI collaboration · Research-driven strategy
