WBG Enterprise AI-Powered Search.
Case study
Case study
This case presents the transformation of The World Bank’s enterprise search from existing a keyword-based retrieval tool into an AI-powered knowledge discovery experience. The product introduced AI-generated, grounded summaries and structured knowledge outputs while preserving the existing search workflow and access to traditional results.
The product also serves as an entry point into the broader World Bank AI platform, connecting search-based knowledge discovery with conversational AI through AI Chat.
STRATEGY
Product Context, Business Problem
Enterprise Search serves as a long-standing, widely used entry point to The World Bank’s institutional knowledge, including projects, documents, research, employee-related data, announcements, and other internal sources.
The main business problem was not lack of information, but the effort required to turn large volumes of search results into useful knowledge.
Manual knowledge discovery.
Users had to review, compare, and synthesize results themselves.
Slow understanding.
Broad or cross-source queries could require significant time to understand.
Limited answer support.
Traditional search helped users find information, but provided limited support for understanding it.
The main business problem was not lack of information, but the effort required to turn large volumes of search results into useful knowledge.
Manual knowledge discovery.
Users had to review, compare, and synthesize results themselves.
Slow understanding.
Broad or cross-source queries could require significant time to understand.
Limited answer support.
Traditional search helped users find information, but provided limited support for understanding it.
Product Oportunity
The opportunity to introduce AI into an existing, familiar search workflow rather than create a separate experience.
- Use AI to help users synthesize and understand retrieved information.
- Introduce AI through a product employees already use in their daily work.
- Provide a trusted entry point to AI within the World Bank platform.
- Create a natural path from Search to AI Chat for deeper exploration.
AI Use Case Discovery
The discovery process focused on identifying where generative AI could add value to existing enterprise search workflows.
Several areas showed strong potential:
Several areas showed strong potential:
- synthesizing information across multiple results;
- answering broad questions using institutional knowledge;
- providing structured information for specific queries, such as people and projects;
- surfacing frequently updated operational information;
- supporting the next step in knowledge exploration.
Product Vision
To evolve enterprise search from a document retrieval experience into an answer-first knowledge discovery layer, helping employees understand relevant World Bank information faster while keeping the underlying sources visible and accessible.
Answer-First Knowledge Access
Move beyond ranked document lists by providing concise, AI-generated overviews of relevant information directly within Search.
Grounded and Traceable AI
Generate answers using relevant World Bank knowledge and provide source references, allowing users to verify the information.
Preserve the Existing Search Workflow
Introduce AI without replacing familiar search patterns. Traditional search results remain available alongside the AI experience.
Search-to-AI Chat Continuity
Allow users to move from an AI-generated search answer into AI Chat, carrying the search context forward for deeper exploration.
Answer-First Knowledge Access
Move beyond ranked document lists by providing concise, AI-generated overviews of relevant information directly within Search.
Grounded and Traceable AI
Generate answers using relevant World Bank knowledge and provide source references, allowing users to verify the information.
Preserve the Existing Search Workflow
Introduce AI without replacing familiar search patterns. Traditional search results remain available alongside the AI experience.
Search-to-AI Chat Continuity
Allow users to move from an AI-generated search answer into AI Chat, carrying the search context forward for deeper exploration.
Enterprise Readiness
Introducing AI into an established Search experience required supporting both the technology and employee readiness for AI-enabled knowledge discovery.
Data. The existing enterprise Search provided the foundation for accessing World Bank internal knowledge.
Employees. AI capabilities were introduced directly into the Search workflow, with users able to switch between AI-powered and traditional search while getting familiar with the new experience.
Platform and Governance:
The AI Knowledge Portal supported the adoption of AI Search by providing employees with learning, guidance, resources, and community support to build AI knowledge, skills, and confidence.
Data. The existing enterprise Search provided the foundation for accessing World Bank internal knowledge.
Employees. AI capabilities were introduced directly into the Search workflow, with users able to switch between AI-powered and traditional search while getting familiar with the new experience.
Platform and Governance:
The AI Knowledge Portal supported the adoption of AI Search by providing employees with learning, guidance, resources, and community support to build AI knowledge, skills, and confidence.
PRODUCT FOUNDATION
The existing Enterprise Search provided reliable access to institutional information, but its keyword-based model placed much of the burden of information discovery and synthesis on employees. The experience offered limited contextual understanding of search queries, requiring users to navigate multiple results, filters, and predefined content blocks to identify and connect relevant information.
- Limited Contextual UnderstandingSearch returned ranked results with snippets and metadata, but provided little support for understanding the meaning behind a query or connecting information across results.
- Complex Filtering and NavigationUsers relied on content tabs, filters, and pagination to narrow large result sets. Finding relevant information often required repeated refinement and navigation across multiple pages.
- Rule-Based ContextSome queries triggered predefined blocks with related announcements, terminology, or quick links. These added limited context, but only worked for specific, predefined scenarios.
- Manual Knowledge SynthesisUsers still had to open results, compare information across sources, and connect the findings themselves. Search retrieved information, but did not synthesize it into an answer.
These product limitations, combined with user pain points and business needs, shaped the evolution of Enterprise Search into an AI-powered knowledge discovery experience within the broader World Bank AI platform.
PRODUCT EVOLUTION
The existing Enterprise Search provided reliable access to institutional information, but its keyword-based model placed much of the burden of information discovery and synthesis on employees. The experience offered limited contextual understanding of search queries, requiring users to navigate multiple results, filters, and predefined content blocks to identify and connect relevant information.
Product flow transformation
Iteration 1 — Introducing AI into Search
What we introduced → what we learned → what changed
Iteration 2 — Expanding AI-Powered Knowledge Discovery
What we introduced → what we learned → what changed
AI Validation & Testing
AI products require validation beyond traditional usability testing. The product needed to be evaluated not only on whether the interface worked, but also on whether the AI produced relevant, grounded, and trustworthy responses.
Response quality: relevance and usefulness of generated answers.
Retrieval quality: whether the system surfaced appropriate internal information.
Source references: whether responses could be traced back to the retrieved sources.
User feedback: understanding where AI responses or interactions didn't meet user expectations.
Security: ensuring users could only retrieve information available to them.
Validation is treated as an ongoing part of product development rather than a one-time activity.
Response quality: relevance and usefulness of generated answers.
Retrieval quality: whether the system surfaced appropriate internal information.
Source references: whether responses could be traced back to the retrieved sources.
User feedback: understanding where AI responses or interactions didn't meet user expectations.
Security: ensuring users could only retrieve information available to them.
Validation is treated as an ongoing part of product development rather than a one-time activity.
SCALING AND IMPACT
Enterprise Al Scaling
The product evolved from a conversational AI MVP into a broader enterprise AI workspace supporting knowledge access, document-based space, internal database search, reusable internal AI applications, and multiple AI capabilities.
Scale
Scale
- Organization-wide rollout
- 20,000+ internal users
- Employees across 189 member countries
- More internal knowledge sources
- Multiple AI models and search capabilities
- Multi-document workflows
- Reusable AI applications
- Mobile access
- More accessible institutional knowledge
- AI integrated into everyday knowledge workflows
- Foundation for broader enterprise AI adoption
Hard KPIs. Workflow value
Organizational Capability Shift
~60% of users transitioned from initial exploration to repeated usage, reflecting a shift from AI experimentation toward integration into daily workflows.
AI-Assisted Workflow Efficiency
~70% faster task completion for document-based workflows, highlighting the potential of AI to improve productivity and support knowledge-intensive work.
Trusted AI Environment
~80% reduction in reliance on external AI tools, indicating a shift toward using the internal AI environment for sensitive and organization-specific work.
~60% of users transitioned from initial exploration to repeated usage, reflecting a shift from AI experimentation toward integration into daily workflows.
AI-Assisted Workflow Efficiency
~70% faster task completion for document-based workflows, highlighting the potential of AI to improve productivity and support knowledge-intensive work.
Trusted AI Environment
~80% reduction in reliance on external AI tools, indicating a shift toward using the internal AI environment for sensitive and organization-specific work.
Soft KPIs. Adoption, usage
Improved access to institutional knowledge: employees could find and work with information through natural language rather than navigating multiple systems.
Reduced friction in knowledge-intensive work: AI supported searching, summarizing, translating, drafting, and working across documents.
Higher user confidence and trust: source references and controlled access helped users understand and verify AI-generated information.
Broader AI adoption: the product moved from experimentation with chat to integration of AI into everyday workflows.
Greater discoverability of AI capabilities: reusable AI Apps/Agents made specialized AI capabilities easier to find and use.
Better continuity of work. Spaces and conversation history supported longer-running tasks rather than isolated questions.
Reduced friction in knowledge-intensive work: AI supported searching, summarizing, translating, drafting, and working across documents.
Higher user confidence and trust: source references and controlled access helped users understand and verify AI-generated information.
Broader AI adoption: the product moved from experimentation with chat to integration of AI into everyday workflows.
Greater discoverability of AI capabilities: reusable AI Apps/Agents made specialized AI capabilities easier to find and use.
Better continuity of work. Spaces and conversation history supported longer-running tasks rather than isolated questions.
Enterprise Strategic Impact
The product became more than an internal chatbot. It evolved into a reusable enterprise AI capability that connected employees with institutional knowledge, AI models, search, documents, and reusable AI applications within a trusted environment.
Enterprise AI capability
Created a shared foundation for delivering AI capabilities across the organization.
Knowledge access
Made institutional knowledge easier to discover, understand, and use in everyday work.
Platform potential
Created a foundation that could be extended with new data sources, models, AI applications, and workflows.
Enterprise AI capability
Created a shared foundation for delivering AI capabilities across the organization.
Knowledge access
Made institutional knowledge easier to discover, understand, and use in everyday work.
Platform potential
Created a foundation that could be extended with new data sources, models, AI applications, and workflows.
ROLE AND DELIVERY
My Role
My official role at The World Bank was Senior Consultant. Working within the organisation’s AI-focused technology department, I contributed to the evolution of enterprise AI products across the product lifecycle, from early exploration and MVP definition through delivery, rollout, and continuous improvement.
My work combined product strategy, AI use case discovery, hands-on prototyping, and cross-functional collaboration. I connected user and organizational needs with AI capabilities, technical constraints, and delivery requirements to shape product direction and priorities.
Product Direction and Prioritization: shaped product direction and priorities based on user needs, organizational goals, AI capabilities, and technical feasibility.
Discovery and Prototyping: explored AI use cases and product concepts through hands-on prototyping, user feedback, and iterative validation.
Cross-Functional Collaboration: worked with engineering, data architecture, and business stakeholders to define workflows, feature logic, requirements, and priorities.
Delivery and Evolution: contributed across MVP definition, delivery, rollout, and subsequent iterations, using feedback and adoption insights to guide product improvements.
My work combined product strategy, AI use case discovery, hands-on prototyping, and cross-functional collaboration. I connected user and organizational needs with AI capabilities, technical constraints, and delivery requirements to shape product direction and priorities.
Product Direction and Prioritization: shaped product direction and priorities based on user needs, organizational goals, AI capabilities, and technical feasibility.
Discovery and Prototyping: explored AI use cases and product concepts through hands-on prototyping, user feedback, and iterative validation.
Cross-Functional Collaboration: worked with engineering, data architecture, and business stakeholders to define workflows, feature logic, requirements, and priorities.
Delivery and Evolution: contributed across MVP definition, delivery, rollout, and subsequent iterations, using feedback and adoption insights to guide product improvements.
Delivery Approach
The product was delivered through an iterative Agile approach, combining structured planning, cross-functional collaboration, and continuous feedback.
Plan. Product priorities and roadmap were translated into epics, user stories, acceptance criteria, and tasks in Azure DevOps.
Explore. Product, engineering, data, and architecture teams worked together to explore solutions, clarify requirements, and align on technical feasibility.
Deliver. Work progressed through Agile/Scrum cycles, including planning, backlog refinement, daily coordination, and sprint reviews.
Learn and Iterate. User feedback, testing, and adoption insights informed product decisions and subsequent iterations.
Plan. Product priorities and roadmap were translated into epics, user stories, acceptance criteria, and tasks in Azure DevOps.
Explore. Product, engineering, data, and architecture teams worked together to explore solutions, clarify requirements, and align on technical feasibility.
Deliver. Work progressed through Agile/Scrum cycles, including planning, backlog refinement, daily coordination, and sprint reviews.
Learn and Iterate. User feedback, testing, and adoption insights informed product decisions and subsequent iterations.
🖂 hi@foxalice.com / LinkedIn
© Fox Alice
© Fox Alice
