Use Case Intelligence RSS Search use cases Search
Microsoft Commercial and Professional Services Australia

KPMG

Enterprise Knowledge Search

Early in 2023, KPMG Australia deployed the first version of its KymChat conversational AI assistant. Two months later, the company added Azure Cosmos DB for MongoDB vCore to the solution, substantially increasing the quality and speed of its query responses. Centered on ease of use, the solution has also been made available to KPMG clients looking to improve their own generative AI adoption.

Read the original story Explore all use cases
"Search quality jumped from 50 to 91 percent, and our results were being delivered in under a second." KymChat also makes use of Azure App Service and Azure Container Registry, which helps the solution deliver faster results at scale.
Even more impressive, the solution has led to as much as a 50 percent jump in productivity across workloads like document drafting, communications, sales proposal development, and research.
KPMG Australia, on the other hand, has already developed and deployed KymChat, a generative AI agent that could revolutionize how the firm's 10,000 employees communicate and collaborate.
Search quality jumped from 50 to 91 percent, and our results were being delivered in under a second." Robert Finlayson, Senior Product Manager, KPMG Australia Many companies are still grappling with the prospect of introducing large language models to their workstreams.

Reported impact

Business outcome

Search quality jumped from 50 to 91 percent, and our results were being delivered in under a second." Robert Finlayson, Senior Product Manager, KPMG Australia Many companies are still grappling with the prospect of introducing large language models to their workstreams.

  • Customer Experience
  • Productivity Improvement
  • Revenue Growth
  • Risk Reduction

Solution

KPMG built its Microsoft solution to improve knowledge retrieval and expert-assistance workflows. Early in 2023, KPMG Australia deployed the first version of its KymChat conversational AI assistant. Two months later, the company added Azure Cosmos DB for MongoDB vCore to the solution, substantially increasing the quality and speed of its query responses. Centered on ease of use, the solution has also been made available to KPMG clients looking to improve their own generative AI adoption. Key components included Azure OpenAI Service ran model inference or heavy processing, and Azure App Service delivered dashboards, apps, or copilots to users. User prompts and domain content are orchestrated through Azure OpenAI Service, processed by Azure OpenAI Service, and then delivered through Azure App Service so KPMG can return context-aware answers or automation in the flow of work. The stack stayed governed and extensible on Microsoft.

Data flow

User prompts and domain content are orchestrated through Azure OpenAI Service, processed by Azure OpenAI Service, and then delivered through Azure App Service so KPMG can return context-aware answers or automation in the flow of work.

Solution architecture

2 components · 2 layers
  1. Compute
    • Azure OpenAI Service Runs large language models that generate grounded responses, summaries, and copilots.
  2. Serving
    • Azure App Service Hosts customer-facing or internal web applications on managed Azure infrastructure.

Architecture clues

  • "Search quality jumped from 50 to 91 percent, and our results were being delivered in under a second." KymChat also makes use of Azure App Service and Azure Container Registry, which helps the solution deliver faster results at scale.
  • Backend code responsible for data preparation, prompt construction, and AI integration with Azure Cosmos DB for MongoDB vCore is stored in an Azure Container Registry docker image, and KymChat uses Azure App Service to deploy a containerized web app based on that docker image.
  • KymChat helps expand these client benefits by enhancing the discoverability of the firm's large stores of unstructured data, automating mundane tasks, and offering rapid responses to common queries.

Evidence from the source

"Search quality jumped from 50 to 91 percent, and our results were being delivered in under a second." KymChat also makes use of Azure App Service and Azure Container Registry, which helps the solution deliver faster results at scale.
Early in 2023, KPMG Australia deployed the first version of its KymChat conversational AI assistant. Two months later, the company added Azure Cosmos DB for MongoDB vCore to the solution, substantially increasing the quality and speed of its query responses. Centered on ease of use, the solution has also been made available to KPMG clients looking to improve their own generative AI adoption.
Even more impressive, the solution has led to as much as a 50 percent jump in productivity across workloads like document drafting, communications, sales proposal development, and research.
KPMG Australia, on the other hand, has already developed and deployed KymChat, a generative AI agent that could revolutionize how the firm's 10,000 employees communicate and collaborate.
Search quality jumped from 50 to 91 percent, and our results were being delivered in under a second." Robert Finlayson, Senior Product Manager, KPMG Australia Many companies are still grappling with the prospect of introducing large language models to their workstreams.
So far, 70 percent of the firm's employees have adopted KymChat, making over 120,000 requests since its release.
Transparency Evidence & methodology 90/100 reference value · 89% confidence · 1 source

Review how this reference was scored and trace each normalized claim back to its documented source.

01

Scoring methodology

How scores are calculated

Customer Reference Value measures usefulness as a customer reference. Evidence Confidence measures the quality and traceability of available evidence, not independent verification.

Customer reference value90/100
  • Problem clarity20/20
    Measures whether the customer's operational problem is specific enough to compare.
  • Implementation depth20/25
    Rewards named components, data flow, integrations, and architecture detail.
  • Outcome specificity20/20
    Measures whether the reported benefit is concrete rather than a generic claim.
  • Quantified impact25/25
    3 metrics describe measurable impact.
  • Deployment scale5/10
    Rewards evidence of a live or scaled deployment and meaningful integration scope.
Evidence confidence89%
  • Source authority20/25
    The main evidence is an official vendor-published customer story.
  • Claim traceability21/25
    10 of 12 key fields link to a named source.
  • Direct evidence20/20
    6 direct source quotes retained.
  • Date provenance10/10
    The publication date comes from the source page or publisher.
  • Source transparency8/10
    1 source reference document publisher, URL, authority, retrieval, and supported fields.
  • Implementation specificity10/10
    Rewards source-backed components, data flow, quantified impact, and a specific outcome.
02

Traceability

Sources & provenance

Field coverage shows which facts each source supports. Dates are shown only when documented by the source.

  1. Microsoft

    KPMG customer story

    First-party
    https://www.microsoft.com/en/customers/story/1700854587537724864-kpmg-australia-azure-ai-professional-services
    Source type
    Vendor-published customer story
    Published
    Nov 8, 2023
    Supports
    • Customer
    • Business problem
    • Solution
    • Products
    • Solution components
    • Data flow
    • Integrations
    • technical area
    • use case category
    • Business outcome
    • outcome category
    • Metrics
    • Architecture
    • evidence quotes
    • Publication date

Date evidence

  • Published on source page Nov 8, 2023 · Microsoft · First-party
    Open date source
Not separately sourced Industry, Region, Company size. Treat these normalized attributes as lower-confidence until a dedicated source is added.

Related

Similar use cases

See all like this
Microsoft May 19, 2025 Accenture Commercial and Professional Services Global Accenture's enterprise clients wanted to move past generative AI proofs of concept to production-grade, scalable, compliant applications, but ad hoc development stitched together a dozen tools and could not answer how… 16+ solutions in full production Enterprise Knowledge SearchIT Operations Automation Azure AI Content Safety · Azure AI Foundry · Azure AI Search Microsoft Jan 27, 2025 mci group Commercial and Professional Services Switzerland mci group plans to expand the use of innovative technologies to deliver better data-driven solutions, improve collaboration, and boost operational effectiveness among employees, thereby better supporting them in their… Enterprise Knowledge SearchIT Operations Automation Microsoft 365 Copilot Microsoft Jan 20, 2025 Gojob Commercial and Professional Services France It also offers an environment which is consistent with our needs in terms of data sovereignty, confidentiality, and security," Pascal Lorne adds. Gojob is a 100% digital employment agency, which was created in 2015 out of the conviction that access to… Customer Service AutomationEnterprise Knowledge Search Azure OpenAI Service · Power Platform