CRESCO adopts Microsoft 365 Copilot as a transformative time-saving investment

Generative AI Business Transformation Lab:
Transforming as a company, creating value for customers

Generative AI (hereafter, GenAI) has the potential to be a game-changer for all types of production. Regardless of industry, many companies are feeling the same sense of urgency as they try to come to terms with this momentous revolution. While CRESCO, founded in 1988, has played a role in a digitizing society as an independent systems integrator, they are no exception to this need for change. What sets CRESCO apart from other corporations is that their own growth fuels their customers' growth.

In 2022, ChatGPT and other GenAI took the world by storm, and the concept of natural language interface caught global attention. AI evangelists within CRESCO took notice of this trend. These forward-thinking individuals came forward and urged the president to join in this revolution by saying, “With the normalization of GenAI, what customers need from our system integration and system development process will change drastically. We need to consider the future. If we continue software development in line with traditional methods, we will not survive in the IT field.”

Convinced by these visionaries, the president initiated the use of GenAI. At the core of this decision is the CRESCO charter, created at the company's founding, which continues to breathe life into its core to this day:

• CRESCO is a people-centered, merit-based company.

• CRESCO is a company characterized by freedom, youth, and dreams.

• CRESCO is a company that demonstrates the most sophisticated technology.

• CRESCO is a company where each person participates in running the business.

• CRESCO is a company that is a citizen of the world.

Thus, the challenges of GenAI are in line with CRESCO’s way of thinking, as outlined in their charter.

In 2023, CRESCO developed and initiated use of their in-house conversational GenAI service, CrePT. In July 2024, they created the cross-functional Generative AI Business Transformation Lab under the president’s direct control, giving form to the vision of these pioneers within the organization.

This lab is comprised of two teams: the Basic Research Team which focuses on expanding technology and digital solutions, and the Application Promotion Team, which focuses on bringing about digital transformation. Their main targets are (1) creating new value for customers; (2) digital transformation within the company using GenAI; and (3) human resource development for the GenAI generation.

Takayuki Teramura, Director Managing Executive Officer at CRESCO, talks about one of the most critical themes of their research lab, human resource development:

“As a systems integrator, one of our main assets is our staff. GenAI allows each employee to expand their potential and strengthens our human capital management.”

In CRESCO Group’s long-term vision, CRESCO Group Ambition 2030, they state their aim to create competitive superiority for their customers through IT and technology while aspiring to become a company that creates digital value to advance society. Their initiative, centered on the Generative AI Business Transformation Lab, will provide the impetus necessary to realize their vision, and CRESCO selected Microsoft GenAI to create this strategic infrastructure.

Copilot trial convinces CRESCO  the way they work will change dramatically 

In April 2024, CRESCO began a trial of Microsoft 365 Copilot (hereafter Copilot), the Microsoft 365 AI assistant. Makoto Suzuki, Quality & Process Management Unit General Manager at CRESCO, reflects on the introduction of Copilot; “We use Microsoft Teams for internal communication, and Word and Excel for regular tasks, so integration of an AI assistant with office software is an extremely attractive concept. With Copilot, using GenAI is simple, without switching between screens or worrying about system integration. It’s simple for the staff to use, and it’s easy for the IT system team to manage and control, which is a major advantage.”

Suzuki also praised the security aspect of this system; “Microsoft provides a responsible AI that is fair, reliable, safe, and values privacy and security. It employs measures that ensure that any data collected is not used as training data and cannot be viewed by Microsoft. Designed from the ground up to be used by major enterprises, I find it to be extremely satisfactory.”

The Development Promotion Office of the Quality & Process Management Unit was selected for the Copilot trial. “Staff who participated in the trial recommended it to me, saying it would be useful for my daily work, so I tried it, too,” says Teramura.

“As a top management member, I handle the final interviews for potential staff. Previously, I had to enter a wide variety of post-interview data for each interviewee by myself, which took about thirty minutes. But by using the Teams Copilot summary function, that was reduced to only five minutes,” explains Teramura about the impact of Copilot on his work. “Consequently, we extended the license to thirty other executives, who all commented on its convenience, and currently all executives use it. Also, Copilot can give feedback on promotion exam reports by summarizing key points and addressing a staff member’s strengths and weaknesses. As I am the one who has to perform a final evaluation of this data, viewing a summary really helps me utilize my time more efficiently.”

In July 2024, they reviewed the results of the Copilot trial. They discovered that it saved the average user approximately ten hours of work per month, and thus a proposal was submitted to the top management to make Copilot available to all employees. “The initial trial was for only thirty licenses, so we had some debate before deciding to increase that to one thousand and use it company-wide.”

“Personally, I have found Copilot to have a drastic impact on my work style. This transformation benefits not only my own company but also our customers, creating additional value for all. I firmly believe we can provide stronger solutions this way. I explained to everyone that this upfront investment is very much in the spirit of CRESCO’s charter, which states that we aim to provide the most sophisticated technology. My fellow executives using it for potential employee interviews also found it to be extremely beneficial, and after receiving the agreement of the president and the entire board, we introduced company-wide use of Copilot in January 2025.” 

Asking Copilot first for any situation

Tomoyuki Ojin of CRESCO’s Development Promotion Office talks about educating staff in Copilot use. “In 2023, we taught all staff the basic rules for dealing with GenAI as well as things to be careful of, such as AI hallucinations. Microsoft provided e-learning courses to help with the company-wide initiation of Copilot. We also provided a Teams channel dedicated to AI-related topics so that users could share insights and information about Copilot. Additionally, we provided hands-on training for various topics for departments such as HR and the legal department, which were not as familiar with the use of IT.”

Six months after introducing Copilot, usage exceeded 90%. According to the user metrics, Copilot was most often called upon to summarize web meetings and create minutes, as well as for document searches and report creation. “Using Copilot has become the norm. Previously, our production staff sent many inquiries to the legal team requesting checks for errors in contracts. Now, the production teams use Copilot to check for any contract errors before forwarding them to the legal department as a part of the workflow. Copilot has helped to reduce the number of inquiries to and corrections by the legal department, reducing the workload on both sides,” Suzuki says. 

“I also use it to review proposals to customers. I’ll ask Copilot, ‘If you were the customer, how would you respond to this?’ The resulting comments really help me to see things from a different point of view.”

Suzuki talks about using Copilot with citizen developers. “We revamped the core system using the standard package  and thus decided not to develop add-ons. When add-ons are absolutely necessary, we utilize citizen developers. After learning how to use Copilot, even management staff with no coding experience can assist in development. We have a staff support center, but, first, people ask Copilot, ‘If I want to do XYZ, how could I achieve it using Excel?’ That’s the first step. Also, younger staff members are getting used to using Copilot to handle issues on their own.”

CRESCO also uses Azure OpenAI, which leverages GenAI models, and GitHub Copilot for coding support in their core operations area of system development. “We use different GenAI models in different ways, depending on what the target is,” Ojin says.

CRESCO monitors Copilot usage and quantifies its effectiveness, with metrics such as the reduction in work hours as a result of its adoption, and its ROI (return on investment) has almost reached a satisfactory level. Ojin adds, “What we look for in the qualitative results is how the way we work has changed. For example, I can have Copilot sit in on a web meeting where my attendance is not essential, and check the minutes afterwards, which allows me to increase my productivity.”

CRESCO has determined their basic policy for use of GenAI, which they will apply in all aspects of their business to maximize productivity and competitiveness and produce new value.

Usage, prediction models, rules, and education ― all in one package

Ojin says of future development, “Our company has implemented AI-driven development rooted in GenAI. Using GenAI in the development process allows for higher quality and productivity with fewer staff.” Teramura explains how AI-driven development is key from a human resource management standpoint. “By leaving downstream tasks to AI, engineers can focus on upstream tasks and display greater creativity. As the workforce shrinks, this will allow workers to be more effective, help us find ways to increase motivation, and raise our capacity for competitive development.”

Teramura discusses how customer requests for proposals have changed. “Major auto makers use GenAI but don’t really know how to maximize its use and effectiveness in their business. As a collaborative partner, we work with customers and use trial and error to figure out the best way to use GenAI for solving relevant issues. Not just in this case, more and more customers have been asking us to propose solutions that include GenAI. By strengthening the collaboration with Microsoft, we feel that we can better fulfill our customers’ wants and needs.”

CRESCO provides feedback on the introduction of Copilot and resulting developments and insights to the Generative AI Business Transformation Lab, which helps create greater creative value for customers. “One major benefit of this system is that we can assist with not only system introduction, but also practical applications, demonstrations of success and failure prediction models, rules, training, and overall insight as a set,” says Ojin. Teramura says of Copilot’s essential value, “What is vitally important is that this is not simply about streamlining the process. Copilot gives us the gift of time, which allows us to be a true collaborative partner, using that time to ponder our customers’ needs more deeply, and create a greater understanding.”

CRESCO’s group ambition is to “create the future, beyond the horizon.” Copilot and Microsoft GenAI models help CRESCO to contribute to the sustainable growth of corporations and society by realizing their visions together.