EY Software / SaaS global

Microsoft

Enterprise Knowledge Search

Microsoft's finance organization wanted to improve data literacy and speed product-launch work, but critical knowledge was spread across tens of thousands of historical financial documents and disparate databases, making it slow to retrieve requirements, compare prior launches and turn finance data into actionable insight.

Read the original story Explore all use cases

50% reduction in the time required to gather finance requirements for new product introductions

75% reduction in time for MVP development

Replay AI reduces AI application MVP development from four months to six weeks

over 100 employees are now utilizing Finance Launch AI

Reported impact

Business outcome

Microsoft's finance team gained reusable AI applications that cut launch and analytics cycle times, improve access to institutional knowledge and let more employees retrieve trusted finance information without waiting on specialists or rebuilding tooling from scratch.

Solution

Microsoft and EY built two connected generative AI applications to modernize finance operations. Replay AI became a centralized ecosystem of AI-based applications linked to different finance databases so teams could reuse summarization, extraction and analysis capabilities instead of rebuilding them project by project. Natural language processing converts business questions into SQL queries, making reports, charts and trend analysis available to nontechnical users. In parallel, Finance Launch AI uses a large language model and a knowledge library of historical launch artifacts - including Excel, Word, PDF, PowerPoint, email, JPEG, OneNote and Visio files - to answer launch questions, surface comparable prior launches, identify stakeholders and draft initial finance requirements. Retrieval-augmented generation keeps both systems grounded in trusted internal sources, giving Microsoft's finance team a scalable knowledge layer for launches, analytics and faster MVP development.

Data flow

Historical finance documents and launch artifacts are collected into a knowledge library connected to Replay AI and Finance Launch AI. Users ask questions in natural language, NLP turns queries into database instructions and analytical outputs, RAG retrieves trusted source material, and the assistants return summaries, requirements, comparable launches and reports for finance teams.

Solution architecture

5 components · 4 layers

Compute
- Natural Language Processing (NLP) Transforms finance users' natural-language questions into SQL queries and structured analysis outputs.
Storage
- knowledge library Stores historical launch and finance documents in multiple formats so the AI applications can query trusted internal information.
Serving
- Replay AI Hosts reusable AI applications connected to finance databases for summarization, extraction and analysis.
- Finance Launch AI Provides a chat-based assistant that retrieves launch knowledge and drafts finance requirements for new product introductions.
Orchestration
- Retrieval-augmented generation (RAG) Grounds both AI systems in trusted source material to improve response quality and reliability.

Architecture clues

Finance Launch AI uses a large language model over a knowledge library containing Office documents, images and email.
NLP translates user questions into SQL queries and report outputs.
RAG was implemented across both systems to keep responses grounded in trusted information.
Replay AI is a centralized ecosystem of AI applications linked to different finance databases.

Evidence from the source

"Finance Launch AI enabled a 50% reduction in the time required to gather finance requirements for new product introductions."

"Finance Launch AI is a chat-based application powered by a large language model designed to simplify essential financial processes."

"Microsoft set out to enhance data literacy and harness institutional expertise by consolidating tens of thousands of detailed historical financial documents into a single knowledge library that can be queried in natural language."

"Microsoft's work to bring AI-enabled efficiency to the finance function began with two parallel efforts, Replay AI and Finance Launch AI."

"Retrieval-augmented generation (RAG) was implemented to enhance the performance of both Replay AI and Finance Launch AI by confirming that the systems reference trusted sources of information."

Transparency Evidence & methodology 100/100 reference value · 85% confidence · 1 source

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

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 value100/100

Problem clarity20/20
Measures whether the customer's operational problem is specific enough to compare.
Implementation depth25/25
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 scale10/10
Rewards evidence of a live or scaled deployment and meaningful integration scope.

Evidence confidence85%

Source authority20/25
The main evidence is an official vendor-published customer story.
Claim traceability25/25
12 of 12 key fields link to a named source.
Direct evidence20/20
5 direct source quotes retained.
Date provenance0/10
The publication date has no documented provenance.
Source transparency10/10
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.

Traceability

Sources & provenance

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

EY
How AI innovation powers Microsoft's finance journey | EY - Global

First-party
https://www.ey.com/en_gl/insights/consulting/how-ai-innovation-powers-microsoft-s-finance-journey

Source type
Vendor-published customer story
Supports
- Customer
- Industry
- Region
- Company size
- Business problem
- Solution
- Products
- Solution components
- Data flow
- Integrations
- Business outcome
- Metrics
- Architecture
- evidence quotes

Date evidence

First archived Nov 14, 2025 · Internet Archive · Independent First Wayback Machine capture; confirms the page existed by this date, not when it was published.
Open date source

Similar use cases

See all like this

MongoDB Sep 9, 2026 Squary AI Software / SaaS United States Squary AI needed to eliminate slow, outage-prone legal search infrastructure that combined MongoDB with OpenSearch and hurt document review speed, cost, and trust. At Quatrini Law Group, Squary AI analyzed more than 460,000 pages and saved over 400 attorney and paralegal hours. Document ProcessingEnterprise Knowledge Search Amazon Bedrock · Claude · MongoDB Atlas

In-house Sep 1, 2026 Thomson Reuters Software / SaaS global General-purpose LLMs can produce subtle but consequential errors in legal and compliance work, where professionals need trustworthy, citation-backed analysis across very large document sets. As of Q1 2026, CoCounsel has crossed one million users Document ProcessingEnterprise Knowledge Search CoCounsel · Practical Law · Tabular Analysis

Anthropic Archived · Aug 30, 2026 Atlassian Software / SaaS Australia Atlassian wanted agents to do consequential work inside Jira, Confluence, and related products, but enterprise customers would only trust long-running automation if the models could follow complex instructions, use… 5 million+ agents executed in customer business workflows each month Enterprise Knowledge SearchIT Operations Automation Claude Code · Claude Platform · Rovo Chat