Case Study 2: AI-Powered Search
This case presents the transformation of The World Bank’s Enterprise Search from an existing 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 workflow and access to traditional results.
AI-powered Search was part of the broader World Bank AI platform, introducing AI through a familiar enterprise workflow. It also connected Search with AI Chat, allowing employees to continue exploring knowledge through conversation.
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.
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 and compare multiple search results to identify relevant information and build an understanding of the topic.
- Time-Consuming Search. Broad or cross-source queries could take significant time to answer, reducing the time available for employees core responsibilities.
- Lack of Knowledge Synthesis. Traditional Search returned relevant links and results, but did not transform them into contextual, synthesized knowledge.
AI Use Case Discovery
The discovery process focused on identifying where generative AI could create meaningful value within existing enterprise Search workflows. Use cases were evaluated based on user value, availability of relevant enterprise data, and feasibility of delivering grounded results.
Several areas showed strong potential:
Several areas showed strong potential:
- synthesizing information across multiple results;
- answering broad questions using institutional knowledge;
- structure information around key entities,
- surfacing frequently updated operational information;
- support deeper knowledge exploration.
Product Vision
To evolve Enterprise Search from document retrieval into an answer-first knowledge discovery experience, helping employees find, understand, and explore institutional knowledge faster while keeping information grounded and accessible.
AI Summary Experience
Move beyond ranked document lists by providing concise, AI-generated overviews of relevant information directly within Search.
Grounded and Traceable AI
Use relevant World Bank knowledge and visible sources to make AI-generated information traceable.
Connected AI Experience
Allow users to move from an AI-generated search answer into AI Chat, carrying the search context forward for deeper exploration.
AI Summary Experience
Move beyond ranked document lists by providing concise, AI-generated overviews of relevant information directly within Search.
Grounded and Traceable AI
Use relevant World Bank knowledge and visible sources to make AI-generated information traceable.
Connected AI Experience
Allow users to move from an AI-generated search answer into AI Chat, carrying the search context forward for deeper exploration.
Product Trade-offs
AI Innovation / Familiarity
AI was introduced as an additional layer rather than replacing the existing Search workflow, allowing employees to benefit from AI-generated answers while retaining traditional results.
AI Answers / Enterprise Trust
AI-generated responses needed to simplify knowledge discovery without obscuring the underlying information. Grounded answers, source references, and existing access controls were preserved.
AI was introduced as an additional layer rather than replacing the existing Search workflow, allowing employees to benefit from AI-generated answers while retaining traditional results.
AI Answers / Enterprise Trust
AI-generated responses needed to simplify knowledge discovery without obscuring the underlying information. Grounded answers, source references, and existing access controls were preserved.
Enterprise Readiness
Introducing AI into an established enterprise Search experience required alignment across technology, employees, and the organisation.
Data. The existing Enterprise Search provided the foundation for accessing World Bank internal knowledge.
Employees. During the transition to AI-powered Search, employees could use both the traditional and AI-powered experiences, allowing them to gradually adopt the new workflow.
Platform and Governance. AI-powered Search was introduced within the broader World Bank AI platform, aligned with organisational AI priorities and supported by shared governance, guidance, and employee learning resources through the AI Knowledge Portal.
Data. The existing Enterprise Search provided the foundation for accessing World Bank internal knowledge.
Employees. During the transition to AI-powered Search, employees could use both the traditional and AI-powered experiences, allowing them to gradually adopt the new workflow.
Platform and Governance. AI-powered Search was introduced within the broader World Bank AI platform, aligned with organisational AI priorities and supported by shared governance, guidance, and employee learning resources through the AI Knowledge Portal.
PRODUCT FOUNDATION
The existing Enterprise Search provided reliable access to institutional information, but its keyword-based workflow placed much of the burden of information discovery and synthesis on employees. As the volume and diversity of institutional knowledge grew, this workflow became less effective for broader and more complex information needs. The experience offered limited contextual understanding of search queries, requiring users to navigate multiple results 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 evolution of Enterprise Search focused on expanding how employees could find and understand institutional information. AI was introduced into the existing Search experience, allowing employees to ask questions in natural language and receive AI-generated summaries and structured answers, with the option to continue exploring the topic through AI Chat while preserving traditional search results and familiar filters.
Product flow transformation
Beta Validation
Before the full rollout, AI-powered Search was introduced through a Beta experience that allowed employees to switch between traditional and AI-enhanced Search. This enabled the team to validate the new experience with real users, collect feedback, and refine the product before broader adoption.
Full rollout
Key Product Changes
Key Product Changes
AI-Generated Summaries
AI-generated overviews synthesize relevant World Bank knowledge into concise, grounded answers directly within Search, while traditional results remain available below.
Natural-Language Search
Employees can ask questions in natural language rather than relying on keyword matching, enabling broader and more complex information needs.
Structured Contextual Answers
AI responses can adapt to the type of information being requested, presenting structured details for entities such as Projects and People, as well as direct answers to operational questions.
Guided Exploration
The new Search experience helps employees explore knowledge beyond the initial query through suggested prompts and context-aware related searches.
Operational Answers
AI Search can surface specific operational information directly, such as guest Wi-Fi instructions, while keeping supporting sources accessible.
Search-to-Chat Exploration
Users can continue exploring a topic in AI Chat from an AI-generated Search answer, extending the experience into conversational exploration.
AI-generated overviews synthesize relevant World Bank knowledge into concise, grounded answers directly within Search, while traditional results remain available below.
Natural-Language Search
Employees can ask questions in natural language rather than relying on keyword matching, enabling broader and more complex information needs.
Structured Contextual Answers
AI responses can adapt to the type of information being requested, presenting structured details for entities such as Projects and People, as well as direct answers to operational questions.
Guided Exploration
The new Search experience helps employees explore knowledge beyond the initial query through suggested prompts and context-aware related searches.
Operational Answers
AI Search can surface specific operational information directly, such as guest Wi-Fi instructions, while keeping supporting sources accessible.
Search-to-Chat Exploration
Users can continue exploring a topic in AI Chat from an AI-generated Search answer, extending the experience into conversational exploration.
Trust, Responsible AI
AI-generated answers were presented with visible source references, allowing employees to verify the information behind the summary. Existing access controls and responsible-use guidance were maintained as part of the AI-powered experience.
Continuous Feedback
In-app feedback was introduced to give employees an easy way to share their experience with AI-generated answers directly within Search. This feedback created a continuous learning loop, helping the team identify issues, understand user needs, and guide further product improvements.
Outcome
Enterprise Search evolved from a keyword-based retrieval tool into an AI-powered knowledge discovery experience that could understand natural-language questions, synthesize information, and provide concise, structured answers while preserving access to traditional search. This shifted Search from a place to find information into a place to understand and explore it, creating a more direct path from a question to institutional knowledge.
SCALING AND IMPACT
The impact of AI-enhanced Search was assessed through usage patterns, employee feedback, and observed changes in knowledge discovery workflows. As an internal enterprise product, its value was reflected primarily in adoption, accessibility, efficiency, and the way employees interacted with institutional knowledge.
Enterprise Al Scaling
Scale:
Organizational Impact:
Faster Access to Institutional Knowledge
AI-generated summaries reduced the effort required to identify and understand relevant information.
Reduced Knowledge Discovery Friction
Employees could move from a question to a synthesized answer without manually reviewing multiple result sets.
AI Adoption Through Familiar Workflows
AI capabilities were introduced through an established enterprise Search experience rather than requiring employees to adopt a separate tool.
Foundation for Broader Enterprise AI Adoption
Search provided a familiar entry point for introducing AI-powered knowledge discovery across the organization.
- Organization-wide rollout
- 20,000+ internal users
- Employees across 189 member countries
Organizational Impact:
Faster Access to Institutional Knowledge
AI-generated summaries reduced the effort required to identify and understand relevant information.
Reduced Knowledge Discovery Friction
Employees could move from a question to a synthesized answer without manually reviewing multiple result sets.
AI Adoption Through Familiar Workflows
AI capabilities were introduced through an established enterprise Search experience rather than requiring employees to adopt a separate tool.
Foundation for Broader Enterprise AI Adoption
Search provided a familiar entry point for introducing AI-powered knowledge discovery across the organization.
Supporting Evidence
Repeated AI Usage
~60% of users transitioned from initial exploration to repeated usage, indicating movement from experimentation toward regular AI use.
Workflow Efficiency
~50% faster task completion was observed for document-based workflows, demonstrating the potential of AI to improve knowledge-intensive work.
Reduced Reliance on External AI
~80% reduction in reliance on external AI tools indicated growing use of the internal AI environment for sensitive and organization-specific work.
~60% of users transitioned from initial exploration to repeated usage, indicating movement from experimentation toward regular AI use.
Workflow Efficiency
~50% faster task completion was observed for document-based workflows, demonstrating the potential of AI to improve knowledge-intensive work.
Reduced Reliance on External AI
~80% reduction in reliance on external AI tools indicated growing use of the internal AI environment for sensitive and organization-specific work.
Enterprise Strategic Impact
Enterprise AI Capability
AI Search introduced AI-powered knowledge discovery into an existing organization-wide workflow, expanding access to AI beyond standalone AI tools.
Knowledge Access
The product made institutional knowledge easier to find and understand, reducing friction across distributed knowledge sources.
Trusted AI Adoption
Grounded responses, visible sources, existing access controls, and a familiar Search experience helped introduce AI within a trusted enterprise environment.
Platform Impact
Search connected established knowledge discovery with conversational AI, creating a natural entry point into the broader World Bank AI platform.
AI Search introduced AI-powered knowledge discovery into an existing organization-wide workflow, expanding access to AI beyond standalone AI tools.
Knowledge Access
The product made institutional knowledge easier to find and understand, reducing friction across distributed knowledge sources.
Trusted AI Adoption
Grounded responses, visible sources, existing access controls, and a familiar Search experience helped introduce AI within a trusted enterprise environment.
Platform Impact
Search connected established knowledge discovery with conversational AI, creating a natural entry point into the broader World Bank AI platform.
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 transformation of Enterprise Search into an AI-powered knowledge discovery experience, working across product strategy, discovery, prototyping, delivery, rollout, and continuous improvement.
My work focused on connecting existing Search capabilities and workflows with emerging AI capabilities, balancing user needs, organisational priorities, technical constraints, and enterprise requirements to shape the product’s evolution.
Product Strategy and Evolution: shaped the evolution of Enterprise Search from keyword-based retrieval toward AI-powered knowledge discovery, defining product priorities based on user needs, organisational goals, and AI opportunities.
AI Use Case Discovery: identified high-value enterprise Search use cases and explored how AI could improve knowledge discovery through hands-on prototyping and iterative validation.
Cross-Functional Collaboration: worked with engineering, data architecture, and business stakeholders to define AI-powered workflows, response logic, requirements, and priorities.
Delivery and Rollout: contributed to beta validation, product refinement, full rollout, and subsequent improvements, using user feedback, usage patterns, and adoption insights to guide product evolution
My work focused on connecting existing Search capabilities and workflows with emerging AI capabilities, balancing user needs, organisational priorities, technical constraints, and enterprise requirements to shape the product’s evolution.
Product Strategy and Evolution: shaped the evolution of Enterprise Search from keyword-based retrieval toward AI-powered knowledge discovery, defining product priorities based on user needs, organisational goals, and AI opportunities.
AI Use Case Discovery: identified high-value enterprise Search use cases and explored how AI could improve knowledge discovery through hands-on prototyping and iterative validation.
Cross-Functional Collaboration: worked with engineering, data architecture, and business stakeholders to define AI-powered workflows, response logic, requirements, and priorities.
Delivery and Rollout: contributed to beta validation, product refinement, full rollout, and subsequent improvements, using user feedback, usage patterns, and adoption insights to guide product evolution
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.
Let's connect
Mail