Normalized claim
Agent response speed: 2 x increase
"enabling it to respond to users’ questions up to two times faster"
Verdantas builds agents for proposal development and contract review with Copilot Studio
Use case typeDocument managementUpdated Jun 13, 2026
Verdantas is an integrated environmental science, engineering, and consulting firm that grew rapidly through acquisitions, creating disconnected systems and data. The company designed an agent in Copilot Studio and used Microsoft Fabric to unify data across the organization. The agent supports employees with proposal development, technical resource search, project and client intelligence, and resource optimization. Verdantas also built a contract management agent that reviews uploaded contracts against legal policy stored in SharePoint to identify escalation clauses and negotiation points. Multi-agent orchestration in Copilot Studio routes requests to sub-agents by data source and improves response speed.
- Organization
- Verdantas
- Industry
- Professional Services
- Location
- United States
- Published
- June 2026
Reported outcomes
150 requests
additional agent requestsOther quantified impact
2xagent response speed
Strategic outcomes
New product / capabilityBuilt proposal and contract review agentsBetter decisions & insightUnified access to internal data sourcesSpeed & agilityFaster employee responses to questionsRisk & complianceImproved contract risk review process
Why do we believe this?Outcome claims, sources, and evidence checks
Normalized claim
Additional agent requests: 150 requests increase
"received over 150 requests for additional agents"
Why do we believe this deployment?Customer identity, provider attribution, maturity, and source checks
- Customer
- Verdantas
- Provider
- Microsoft
- Maturity
- Production
- Linked source
- Microsoft Customer Stories
Used Dataverse as the operational data layer and Power Automate cloud flows to pull data from file shares and systems like ServiceNow
Customer identity supportedSource describes one deploymentMaturity supported
Primary read
Use case focus
Showing 3 of 6
- 1Knowledge management
- 2Document review
- 3Proposal generation
- Rapid growth via acquisitions created disconnected systems and complex, varied datasets.
- Employees struggled to find information across siloed systems, slowing proposal and contract work.
- Built a technical resource agent in Copilot Studio using retrieval-augmented generation.
- Unified data in Microsoft Fabric from sources such as Workday, Deltek, and Harbor Compliance.
- Used Dataverse as the operational data layer and Power Automate cloud flows to pull data from file shares and systems like ServiceNow.
- Applied Azure AI Search for vector-based retrieval and semantic reranking, with custom APIs and tailored prompts to align with internal terminology.
- Implemented contract review capabilities using SharePoint policies and multi-agent orchestration in Copilot Studio to activate sub-agents by data source.
- The agent responds to users’ questions up to two times faster.
- Contract risk tasks that once took hours now take minutes.
- Verdantas received over 150 requests for additional agents across departments.
Architecture
Copilot Studio agents with multi-agent orchestration connected to a Microsoft Fabric unified data layer, Dataverse operational data, Power Automate cloud flows, Azure AI Search vector retrieval and semantic reranking, Microsoft Purview governance, and SharePoint policy content for contract review.
Sources & evidence1
Evidence: Medium65/100Evidence strength
- Customer explicitly identified
- Deployment status explicitly supported
- Primary source available
- Quantified outcome available
- Technical implementation details available
Type: Customer StoryPublished: Jun 13, 2026Publisher: MicrosoftEvidence: PrimaryConfidence: High
AI-generated summary. Verify important details with the linked sources before relying on this case.
Explore related AI use cases
Customer evidence
Provider evidence
Market context
AI patterns
Browse the catalog
Was this useful?
Community
Comments
No published comments yet.