Graebel drives growth and automation through AI innovation on Dynamics 365, Power Platform, and Copilot Studio

Graebel is a global leader in workforce mobility and managed services, partnering with many of the world’s most recognized organizations to support their people and plan for what’s next. As the company expanded in scale and geographic reach, growing data volumes began to strain its legacy systems. Many critical processes relied on manual, time-consuming workflows that limited efficiency and scalability. To address these challenges, Graebel chose to combine human expertise with AI agents by adopting Microsoft Dynamics 365 Finance, Power Platform, and Microsoft Copilot Studio. The company created AI agents to automate key processes such as knowledge retrieval, invoice processing, and action-based workflows across its core business applications. With these new technologies, Graebel is driving automation and enabling future business growth.

Growth constrained by legacy systems

For more than 75 years, Graebel has helped organizations move people across cities, countries, and continents—often during some of the most stressful moments in an employee’s life. Graebel has evolved into a global workforce mobility and managed services provider, supporting complex relocation, compensation, immigration, payroll, and compliance needs for enterprises worldwide.

In recent years, Graebel reached an inflection point common to many long-established global organizations. Multiple lines of business, regional platforms, and specialized systems have evolved over time. Many of these systems were heavily manual and loosely integrated, even as digital tools advanced. Critical workflows continued to depend on spreadsheets, email, and hand-keyed data.

When Graebel implemented Microsoft Dynamics 365 Finance in April 2019, it marked a significant step toward modernization. However, the surrounding ecosystem still reflected years of accumulated workarounds and manual processes. “We were operating effectively despite an ecosystem shaped by years of workarounds and disparate customizations,”  says Shaun Eades, Senior Director of Process Improvement at Graebel. Legacy integrations relied heavily on manual Data Management Framework (DMF) imports, while operational scale continued to grow. 

Invoice processing became a daily bottleneck. Hundreds of emails containing invoices arrived each day, each requiring manual review, data entry into Dynamics 365, and document routing across multiple systems. A team of 16 people managed the workflow, yet errors and delays were inevitable. “Our leaders across the world want efficiency, automation, and intelligence, not forgetting an outstanding customer experience, and we needed to find a way to meet those needs,” says Eades.

Becoming a “Frontier Firm” centered on AI innovation 

Graebel’s technology leadership saw a strong alignment between its own ambitions and evolving Microsoft AI roadmap. The organization committed to becoming a Frontier Firm, a term Microsoft uses to describe organizations that combine human expertise with AI agents, redesign enterprise-wide workflows around AI-first collaboration, treat AI as a digital workforce, and continuously adapt and scale through data, platforms, and AI reasoning.

These ambitions were part of a broader enterprise strategy and culture of innovation, embodied in Graebel’s Be Bold initiative led by CEO Ron Dunlap. 

“That strategy came out of a challenge from our CEO, who asked, 'How do we become that frontier firm? How do we revolutionize this business? How do we do things differently than we've done them before? How do we incorporate AI into everything we do?' That vision has now been set at the highest level,” says Steve Goodwin, Senior Director of Enterprise Applications, Graebel.

Be Bold is based on four pillars: enrich employee experiences, reshape business processes, reinvent customer engagement, and bend the curve on innovation. By 2025, Graebel had laid much of the foundation for the initiative, positioning the company to accelerate transformation in 2026 through structured, cross-functional working groups. Rather than limiting AI to IT, Graebel aimed to extend it across the organization, empowering both professional developers and citizen developers.

A modern data and application platform

Dynamics 365 Finance sits at the center of Graebel’s modern Microsoft architecture. Over time, the platform was extended through Azure Data Factory, Azure Synapse Analytics, Databricks, and a lakehouse architecture that enables fast, SQL-based access to trusted operational and financial data. These services allow data to be cleansed, structured, and prepared for analytics in Power BI.

Power Platform became the connective tissue around Dynamics 365. Power Apps, Power Pages, and Dataverse support customer and client portals, while Power Automate enables end-to-end workflow automation. Building on this foundation, Graebel is actively experimenting with emerging standards such as Model Context Protocol (MCP) to explore new ways of connecting Dynamics 365 with external data sources and operational systems. While this work remains iterative, the intent is clear: reduce reliance on brittle, point‑to‑point integrations and create a more flexible, AI‑ready integration layer. Over time, these patterns are expected to support AI‑assisted experiences where agents can securely retrieve context, create or update records, and initiate workflows. The long‑term outcome is a shift away from employees navigating complex application screens toward more natural, AI‑driven interactions that reduce friction and free capacity for higher‑value work.

Automating key finance processes

One of Graebel’s most tangible examples of AI in action is its Accounts Payable invoice automation solution. Built with Power Automate and AI Builder, the solution ingests incoming vendor invoice emails, extracts data using machine learning models, creates records directly in Dynamics 365, and automatically routes documents to the appropriate repositories. 

What once required a large team to manage invoice intake and manual entry is now handled by a much leaner group focused on exceptions, inquiries, and reconciliations rather than routine processing. The system handles complex variations in vendor data formats, including more than a dozen date formats, and incorporates error handling logic based on documented resolution processes. Since its initial go live in early 2022, the solution has processed hundreds of thousands of document lines year to date, with success rates approaching three-quarters, depending on document quality and vendor consistency.

Plans are underway to evolve this automation into vendor-facing portals and conversational experiences, enabling future AI agents to answer queries, support onboarding, and handle maintenance tasks directly through Dynamics 365. “The future of Dynamics 365 is intelligent and human-centric. Copilot Studio is how we will make that vision real,” says Eades. 

Introducing AI agents into everyday work

As Graebel’s confidence grew, attention shifted to AI agents built with Microsoft Copilot Studio and Power Platform. Early use cases focused on knowledge retrieval, enabling employees to access policy documents and operational artifacts through chat instead of manually searching large PDFs. 

This approach changed how teams interact with institutional knowledge. “The retrieval agent we built on Copilot Studio automatically retrieves information from a number of areas within the organization,” says Goodwin. “This will help our consultants greatly reduce the amount of time they spend searching through shared folders and files to find what they need.”

From there, Graebel began developing agents that could act as well as respond, supporting scenarios such as recruitment screening, employee onboarding, and structured multistep workflows.

One of the most innovative use cases emerged in environments where modern APIs didn’t exist. In collaboration with Microsoft partner Get AI, Graebel drove early adoption of Computer Use Agents (CUAs) to automate interactions with legacy systems. CUAs are a class of agents that can use software the same way a human does—by seeing the screen and interacting with the user interface; it's like a virtual keyboard and mouse rather than running scripts. CUAs can dynamically adapt to UI changes and make it easy to automate old legacy systems, third-party apps, or antiquated websites. Early testing on initial CUAs has shown extremely strong performance, with plans to move workloads into production as human-in-the-loop controls become available. “The goal is not to replace people, but to remove unnecessary administrative friction so teams can focus even more on delivering outstanding experiences,” says Eades.

Governance before speed

The focus on rapid experimentation was matched by a robust commitment to governance. To achieve this governance, Graebel partnered with Get AI once again to establish a robust Center of Excellence (CoE) to govern, standardize, and scale AI innovation. This work spans Power Platform and Copilot Studio and includes standardizing environments, tenant-wide policies, automating pipelines, and strict separation of development, testing, and production workloads.

The results were measurable. Following the CoE implementation, Graebel significantly improved its security posture, disabled more than a thousand unused connectors, and reduced administrative sprawl across environments. Governance frameworks ensured that AI innovation could scale safely, without compromising compliance or data integrity.

A human-led, AI-operated future

Graebel’s agentic journey is still unfolding, with multiple AI-driven solutions currently being piloted across the business. One example is an agent that analyzes publicly available information from top client websites to detect signals such as expansions or organizational changes. The solution is designed to help Graebel identify potential indicators of future service delivery needs and support more proactive client engagement.

Additional pilots span both client-facing and operational functions. These include an agent supporting Request for Proposal (RFP) evaluation, as well as solutions designed to automate invoice auditing and reconciliation. Graebel is also expanding internal knowledge agents that allow employees to retrieve IT documentation, client data, and HR and corporate policies through SharePoint and Copilot Studio.

Other initiatives focus on audit activities and data-population workflows, combining Copilot Studio, Azure Content Understanding, and Power Automate to reduce manual effort and improve data quality.

Collectively, these efforts reflect the scale and diversity of Graebel’s agentic experimentation. The company remains committed to leveraging the Microsoft AI ecosystem to transform how it serves clients, empowers employees, and drives innovation at scale.

As Dunlap explains, “Becoming a frontier, AI-enabled organization requires courage to reimagine long-standing processes and embrace new ways of working. Our Be Bold strategy reflects Graebel’s commitment to lead that transformation while continuing to deliver trusted, people-focused services.”

“We’ve built a strong foundation that allows us to turn these technologies into measurable business impact,” says Goodwin. “While the transformation continues, we’re moving forward with clear momentum and confidence.”