AI at Work: What Microsoft’s Research Really Tells Us

Artificial intelligence is attracting its fair share of headlines, often accompanied by sweeping claims about the future of work. Given much of what is written remains speculative, any evidence-based insight into how AI is being used in real workplaces is welcome.

Microsoft’s latest research (here) analysed over 200,000 anonymised interactions with Microsoft Copilot, the company’s AI assistant built into Microsoft 365, to offer a detailed view of how professionals are using AI.

While we’re big users of Copilot ourselves and have a deep relationship as a Microsoft partner, it’s also fair to say the report is somewhat self-selecting – given how Copilot licencing works and that to date is typically being adopted mostly by higher wage knowledge workers.

However there are still some great insights for businesses of all sizes – so read on 😊.

 

What the Research Looked At

The study tracked how users interact with Copilot across familiar tools such as Word, Outlook, Excel and Teams. Researchers linked this usage data to specific job roles to assess how applicable AI is to each one. In other words, they were not predicting what AI might do in future but measuring how it’s already being used today.

This is important as it gives a clear picture of where AI is delivering value now and how it is reshaping day work today.

 

What the Study Says

  1. AI is Already Supporting Everyday Tasks

AI is proving most useful in roles where information is created, processed or communicated in digital form. Typical tasks include:

  • Drafting written content
  • Summarising meetings or documents
  • Researching and interpreting information
  • Creating reports and presentations
  • Responding to emails or customer enquiries

We think this makes sense and they apply to a wide range of jobs including marketing, sales, admin, operations, finance and customer service.

 

  1. Some Roles Have Adopted AI Faster Than Others

The research highlights a marked difference in AI usage between roles. For example:

  • Writers and editors are using AI to draft content more quickly, repurpose materials and improve consistency.
  • Data analysts are applying AI to interpret spreadsheets, visualise trends and automate reporting.
  • Customer support teams are handling routine enquiries faster with suggested responses and automatic summaries.
  • Sales professionals are preparing more tailored proposals and doing research on prospects in a fraction of the time.
  • Roles with Physical presence or hands-on activity, such as technicians, facilities managers or tradespeople are unsurprisingly less impacted so far

Our note of caution here, especially for the writers and editors, is don’t allow Copilot to make you too generic. AI it’s great at doing stuff but you’re still responsible for knowing what you’re asking it to do and why you’re asking it to do that.

We also expect AI to spread quickly across support, admin, and comms teams, where it can be highly effective behind the scenes. It becomes even more powerful when paired with other AI and automation tools. (Since we’re on the Microsoft track, we’re big fans of Power Automate too).

 

  1. AI is Allegedly Assisting, Not Replacing Human Work

One of Microsoft’s core claims from the study is that AI is being used as a tool, not a substitute. It helps people do their jobs more efficiently by reducing manual effort and improving accuracy. They say in most cases it is saving time on routine tasks so that staff can focus on work that requires judgment, creativity or collaboration.

Really? It seems many big complains feel it’s politically correct to say this is not about removing people from the process and that it is about giving them better tools to succeed.

In our experience, businesses are absolutely looking to harness the clear productivity gains from AI to drive growth. This is often focused on improving customer and employee experiences, but cost efficiency, including the potential to reduce headcount, is also a key part of the conversation.

 

Why This Matters for UK SME & Businesses of all sizes

For business leaders and decision-makers, the research offers three key takeaways.

  1. AI Can Add Value Immediately

You do not need to wait for some future version of the technology. Many AI tools are already built into the software your team uses every day. With the right approach, small steps can deliver quick wins.

  1. The Main Barrier is Adoption, Not Access

In many businesses, the technology is already in place but not being used effectively. Encouraging your team to try AI for common tasks like writing, editing or summarising is often enough to spark momentum.

  1. Leadership Makes a Difference

Teams are far more likely to embrace AI when senior staff lead by example. If you use AI to plan a meeting agenda or summarise a project report, others will follow.

 

Copilot is probably one of the lower-risk AI tools for businesses to adopt, as it operates within their existing Microsoft tenant, which already handles sensitive data. While it may be lower risk than allowing people to copy and paste your data into a free online AI, there are still real world issues around identity and access management to consider and talk to your IT service provider about.

 

Beyond Copilot: Natural langue voice agents and assistants are a great starting place

While Microsoft Copilot is often the first step into AI for many SMEs, it is far from the only tool making a difference in the workplace.

The next step is AI Agents: voice-powered assistants that can answer questions, schedule appointments, qualify leads or route calls, all without requiring a member of staff to pick up the phone. These agents are available around the clock, respond immediately, and connect directly to your CRM or helpdesk systems. By taking care of repetitive tasks such as checking order statuses or booking a demo, they free up your team to focus on more valuable and complex work.

Voice AI tools, such as those in Windsor AI Assist, automatically transcribe and summarise phone and video conversations. This helps teams follow up more effectively, spot customer issues earlier, and improve performance across sales, service and contact centre roles. Accurate records and quick access to insights are becoming essential, not optional.

Together, these tools turn everyday conversations into useful data and actionable insights, while also providing always-on support. It is a practical way to improve customer service, reduce costs, and scale without adding headcount.

 

Get in Touch

If you are considering how AI could support your business, we would be happy to help. Whether you are starting to explore Microsoft Copilot or want to understand the benefits of Voice AI and summarised call recordings, we can show you what works in practice.

Book a short call with our team to see what is possible. No jargon, no pressure, just a clear and honest conversation about how to make AI work for you.

Pete Tomlinson
CEO – Windsor Telecom