6 steps to tackle the challenges of Microsoft Copilot adoption

We recently ran a poll with a small group of fund managers to understand how often they use a generative AI tool in their work. As you can see, 40% were not using it at all, and another 30% only used it on a monthly basis.

How frequently do you use a generative AI tool in your work? e.g. Copilot, ChatGPT (Pick one)

With all the hype around Microsoft Copilot and generative AI, why are these figures so low?

From further discussions, it became clear that many users' first interaction with the technology was underwhelming which meant that they had a level of nervousness when it came to rolling it out to the rest of the business. The reason for this trepidation is that many business leaders understand that you often have just one shot to get business users to adopt something new, and with generative AI, the expectations are very high.

It is therefore important that the rollout and adoption of Microsoft Copilot in your business needs careful planning, or the adoption rate might be low which means users will miss out on the power of this AI tool. Here are 6 challenges of Microsoft Copilot adoption and what steps you can take to tackle them.

1. It’s general purpose AI

Microsoft Copilot is a “general-purpose” generative AI technology, capable of assisting with a wide range of tasks. This means that when given to a business user, it is not immediately obvious how it can help with their day-to-day job. This could lead to the business user then playing around with it and ultimately losing interest.

When introducing any new technology, especially one as versatile as Microsoft Copilot, it's crucial to have a plan in place and introduce some training. That training should be made as interactive as possible and include some real examples of where the tool can be used to save the user time carrying out certain tasks.

2. The novelty of AI prompts

It's important to recognise that using Microsoft Copilot is something totally new for many users. Writing prompts for an AI is not like performing an internet search on Google. You need to change your approach, not necessarily putting in keywords, but writing clear instructions and providing as much context as possible. Then, have a conversation with the Copilot to further refine and get the answer you’re looking for. Prompt engineering is quickly becoming a skill in itself and takes time and practice to master.

3. Finding champions

One effective solution is to identify champions within your organisation. These are individuals who are enthusiastic about new technology and willing to explore its potential. By training these champions, you can create a group of knowledgeable users who can then help their teams discover specific use cases for Copilot. This approach can significantly improve adoption rates, as users see firsthand how the technology can benefit their daily tasks. It will also encourage you to think about “AI agents.”

4. Building specific AI agents

Once you have found the various use cases, consider leveraging Copilot Studio to build custom AI agents tailored to specific tasks within your business. By creating AI agents that focus on particular roles or functions, you can provide users with tools that directly address their needs. This targeted approach can make the benefits of Copilot more apparent, increasing user satisfaction and engagement.

5. Follow training best practices

Training is a critical component of successful technology adoption. Ensure that your training programmes are interactive and engaging, allowing users to explore Copilot's capabilities hands-on. We recommend putting users in small groups, giving them “homework,” and having a couple of sessions for them to give feedback on their experience.

6. Measuring outcomes and benefits

To truly understand the impact of Microsoft Copilot, it's essential to measure outcomes and benefits. Track metrics such as how much time Copilot saves users, improvements in productivity, and user satisfaction. By recording these metrics, you can demonstrate the tangible benefits of Copilot, making it easier to justify its continued use and investment.

Planning and training are key

In conclusion, the initial underwhelming reception of Microsoft Copilot among fund managers can be avoided with careful planning and effective training. This is a very new technology, not something anyone has seen before. While it’s extremely powerful, it’s also very general, so we should not expect business users to quickly adopt and build it into their daily workflow, they just need some encouragement and support to do so.