Each colour maps a published value to the exact source passage used to support it. Only bounded excerpts are public; administrators can inspect the complete captured source.
Title
Roland Berger cuts knowledge-search response times to 8 seconds with graph-based AI agents on Azure
…sources—regardless of whether the knowledge is in-house or from third parties. “By using the graph-based AI agents, we were able to reduce response times to an average of eight seconds. The fact that answers come with sources and citations also increases employees’…
Description
Roland Berger redesigned its internal Knowledge Hub using graph-based, reasoning-acting AI agents built on Azure AI Search, Azure OpenAI, Azure AI Document Intelligence and Azure Databricks. The multi-agent system reduced average response times for complex, multi-source consulting knowledge queries to 8 seconds, with citations included, and is now used by around 70 percent of the firm's workforce.
…sources—regardless of whether the knowledge is in-house or from third parties. “By using the graph-based AI agents, we were able to reduce response times to an average of eight seconds. The fact that answers come with sources and citations also increases employees’…
…port provided by the AI agents is very popular with Roland Berger’s employees: “The AI agents are already being used by around 70 percent of our workforce. This clearly indicates the advantages of the solution and shows that it’s being…
…tely available. However, the search for it took a lot of time. For this reason, Roland Berger decided to implement a central knowledge hub based on Microsoft Azure. The company has continued to develop this hub ever since. Today, specialized AI…
Company
Roland Berger
…ess to knowledge, and make processes even more efficient. Discover more details CUSTOMER Roland Berger PRODUCTS Azure OpenAI Service Azure SharePoint Azure Database for PostgreSQL Az…
Country
Germany
…rosoft 365 Copilot SERVICES AND SUPPORT ORGANIZATION SIZE 1,000-9,999 employees COUNTRY/REGION Germany BUSINESS NEED Artificial Intelligence INDUSTRY Professional and Business Servic…
Industry
Professional Services
…00-9,999 employees COUNTRY/REGION Germany BUSINESS NEED Artificial Intelligence INDUSTRY Professional and Business Services Share the story Explore related stories /h2> "> Explore related stories Stories…
Problem
For management consultants at Roland Berger, having the right information in the right place at the right time is essential, but searching through documents in huge databases and knowledge repositories for non-customer-specific information was at times laborious and time-consuming. Even its first-generation Knowledge Hub reached its limits on complex issues that involved bringing knowledge together from different sources: consultants sometimes had to ask up to three separate questions to get filtered knowledge from three different studies, and each answer came back with only its own source displayed, making it time-consuming to refer back to the documents for review.
…s method and industry knowledge, employee expertise, or specific study results, was at times laborious and time-consuming. “Generative artificial intelligence can provide support here by automatically…
…om a single study, it was able to give them an answer without any problems. But if they wanted to obtain filtered knowledge from three different studies, they sometimes had to ask up to three separate questions. In addition, each answer would come back with only its respective source infor…
…h answer would come back with only its respective source information displayed, making it very time-consuming to refer back to the documents for review purposes. But users also made it clear what enormous potential they felt an AI-based kno…
Solution
Roland Berger redesigned its Knowledge Hub with graph-based 'reasoning acting' (ReAct) AI agents, each an expert in a different area or process, connected via a graph and sharing a process memory. The agents break natural-language user requests into individual steps, retrieve knowledge from Azure AI Search, and hand results to the agent that generates the final response, drawing on internal SharePoint pages or, depending on data classification, external sources via the Microsoft Bing API. Azure AI Document Intelligence extracts information from the various document formats feeding the data corpus, with processing parallelized through Azure Databricks; Azure Database for PostgreSQL stores user-specific information, Azure OpenAI provides the underlying AI models, users interact through an in-house interface built on Azure Static Web Apps, and quality is assured through evaluation in Azure AI Foundry.
…s assigned to a single agent, which works iteratively with the other AI agents. The agents are connected to each other by a graph and share a process memory. To complete their task, the AI agents pinpoint and retrieve knowledge from Azur…
…tinuous feedback from users ensures that the high-quality requirements are met. Quality assurance is also carried out by means of evaluation in Azure AI Foundry . For the AI agents to work effectively, it’s essential that the data corpus in A…
…icks . Processing is parallelized to make the large volumes of data manageable. Azure Database for PostgreSQL is used to store and manage user-specific information, and Azure OpenAI provides access to pre-trained AI models for analysis. Now, when consultants are looking for information contained in three different…
Business value
Using graph-based AI agents, Roland Berger reduced average knowledge-search response times to eight seconds, with answers that include sources and citations, increasing employees' trust in generative AI; consultants now invest the hours they used to spend searching through documents in higher-value, value-adding activities, and the AI agents are already used by around 70 percent of the firm's workforce.
…would make the Knowledge Hub even smarter. “Now, thanks to automated processes, our consultants can take the hours they used to spend searching through documents and invest that time in activities that add value.” Michael Schich, Vice President of Tech & AI Innovation, Roland Berger The solu…
…port provided by the AI agents is very popular with Roland Berger’s employees: “The AI agents are already being used by around 70 percent of our workforce. This clearly indicates the advantages of the solution and shows that it’s being…
AI capabilities
Agentic AI, Document Intelligence
…g from classified databases and those that have been explicitly made available. Azure AI Document Intelligence makes this knowledge usable by extracting all the information from the various formats. The process is coordinated in Azure Databricks . Processing is parallelized to…
…ever, that wasn’t really ideal for our use cases—which is why we decided to use graph-based agents, each of which is an expert in a different area or process,” Schich says. These agents take user requests using natural language and break…
Technology
Azure OpenAI Service, Azure SharePoint, Azure Database for PostgreSQL, Azure AI Document Intelligence, Azure AI Search, Azure Databricks, Azure Static Web Apps, Azure AI Foundry, Microsoft 365 Copilot
…ake processes even more efficient. Discover more details CUSTOMER Roland Berger PRODUCTS Azure OpenAI Service Azure SharePoint Azure Database for PostgreSQL Azure AI Document Intelligence Azure AI Search Azure Databricks Azure Static Web Apps Azure AI Foundry Microsoft 365 Copilot SERVICES AND SUPPORT ORGANIZATION SIZE 1,000-9,999 employees COUNTRY/REGION Ger…
Deployment model
cloud
…tely available. However, the search for it took a lot of time. For this reason, Roland Berger decided to implement a central knowledge hub based on Microsoft Azure. The company has continued to develop this hub ever since. Today, specialized AI…
Deployment options
cloud
…tely available. However, the search for it took a lot of time. For this reason, Roland Berger decided to implement a central knowledge hub based on Microsoft Azure. The company has continued to develop this hub ever since. Today, specialized AI…
Headline outcome
…sources—regardless of whether the knowledge is in-house or from third parties. “By using the graph-based AI agents, we were able to reduce response times to an average of eight seconds. The fact that answers come with sources and citations also increases employees’…
Use case type
Knowledge search
…ub ever since. Today, specialized AI agents provide support in day-to-day work. Employees access all of Roland Berger's approved, non-client-specific knowledge through the AI agents, receive support with lengthy work processes, and gain more time for value-addi…
