Microsoft is modernizing the workplace -- but where does Copilot fit in?

 Whether it’s the company’s commitment to the cloud, to developing its Office suite or being a lead player in the artificial intelligence (AI) revolution, Microsoft remains the top provider of enterprise workplace tools. Despite no shortage of competition, Microsoft passed Apple earlier this year as the largest company in the world with a market capitalization of more than three trillion dollars.

Over the last few years, much has been made of Microsoft’s investment in AI. It is the lead investor in ChatGPT and has now rolled out Copilot, its own AI offering for the workforce. Copilot is an awesome tool, which leverages and trains publicly available code to provide users with suggestions and solutions to work challenges. Used properly, Copilot offers enhanced productivity and streamlined software development processes to users.

However, AI projects still have a high rate of failure; according to industry data, failure rates range from 40 percent to 50 percent or even higher. AI is only as good as the quality of data it has access to and can train. Additionally, AI suffers from false expectations -- right now, there is so much heat in the market around AI that leadership still struggles with what AI is, how it differs from other advanced automation, and ultimately what it needs to be successful, such as resources, management and, yes, financial investment.

For companies to be successful in integrating Copilot into their Microsoft stack, they need two things:

• Understand what AI can do. There are misconceptions about what AI can do and that is no different with Copilot. It’s important to understand before implementation what Copilot looks like in Teams, Outlook, Excel, and Powerpoint and what the advantages -- and limitations -- of each is.
• Adopt a plan for data classification and management. After all, quality data is the heartbeat of any successful AI project. Doing so minimizes risks and maximizes the benefits of Copilot.
• Training and proactively championing adoption in the business. To be successful, users need to understand Copilot’s capabilities and limitations. It goes a long way to ensuring a successful rollout of the new technology.

Getting your house in order

When it launched last November, businesses had to have 300 licensed seats to adopt Copilot -- now they need as little as one. Microsoft is aggressively touting Copilot (including prime real estate advertising during the Super Bowl) and making it available to everyone.

Before turning on Copilot, make sure that there is an information governance plan in place for every piece of data that they can properly classify, identify, and understand. Before Copilot can be implemented, companies need to have their data house in order. Depending on where you are in your data journey, this could be a long road or a short one – but ultimately a critical one.

The 5 Step Approach to Copilot Mastery in Your Organization

While that may seem daunting, keep in mind that it’s critical for every AI project you take on. Clean data in - with governance to manage and security to protect -- will help you address the organization’s growing demands for AI.

As you plot your path forward, the following five steps will help you achieve success with your Copilot strategy:

Define Goals for Copilot. What are you trying to achieve with a ChatGPT implementation? How are you trying to improve operational efficiencies and cut costs? Copilot is integrated into all of Microsoft’s products, so determine what success looks like at each stage of the project and establish specific, measurables against that vision for success.

Are you a financial services organization that wants to take manual data entry out of the hands of employees? Are you in manufacturing and want to accelerate design creation? Copilot can achieve that, but only with goals and measures of success in place.

Validate Tenant and Licensing. Licensing is a headache -- and an expensive one if you don’t get it right. Microsoft has a different pricing structure based on the number of users in the organization and how they are using it. F3 workers -- like nurses -- use Copilot to consume content and not create it. F3 workers are $8 a month per license. Many companies are unaware of the difference between F3 and E3, a knowledge worker who creates content and consumes it. Knowledge workers cost $35 per month per user. Not understanding the difference can cost an organization hundreds of thousands of dollars.

Set Up Governance & Security. If data is the heartbeat that defines the success of a Copilot project, governance, and security are what keeps it beating. Governance determines who is responsible for the automation platform, who can access it to make requests, how to determine that rules are followed, and considers new rules and requirements needed to use Copilot to automate processes.

As cloud-first applications, virtual desktops (VDI), and the Internet of things (IoT) proliferate, cloud sprawl is real -- and providing cyber criminals with new, vulnerable attack surfaces to exploit. A surge in cybercrime over the past few years with cyberattack financial damage now predicted to reach $10.5 trillion annually by 2025, according to McKinsey. Apply best practices for administrative settings and secure Microsoft 365 data using Varonis for automation, analytics, and top-tier data classification. This safeguards sensitive data and enforces DLP, ensuring data integrity and security with CoPilot integration.

Create Communications Plan. Rollout can make or break an AI project. Workers are uneasy about AI and what it means for their jobs. No matter how much employers tell employees that AI will make them more efficient, it’s hard for people to get past the threat of technology replacing them.

Once you get beyond their fears, there is also uncertainty about how to use Copilot. A good strategic communications plan introduces the organization to Copilot but involves a strategic plan to introduce Copilot to the organization. That plan is not one email and done. No, it involves customized emails that breed familiarity and recorded demonstrations tailored to specific use cases so that using Copilot becomes muscle memory -- even if it doesn’t fully alleviate the fear.

Training. An extension of the comms plan that ensures customized training by role and responsibility. For knowledge workers, this may be how to build the reports that the company needs. For the governance and communications teams, it may be how to best tools like Viva Insights to track adoption rates.

Building for Continued Success

Copilot isn’t going away -- if anything, it will become a more prominent part of the Microsoft environment. Your use cases will go away or evolve -- and new use cases will emerge as companies get more familiar with the power of the technology. We are in the infancy of what ChatGPT can achieve.

There will be failures, disappointments, and successes. But, by following these steps, companies can create the foundation to drive continued success with Copilot in their workforces.

Image credit: MamunSheikh / depositphotos

Chuck Salvo is Cloud Architect, EchoStor.