How we measured the business impact of Copilot on Sales at Microsoft

Microsoft

Apr 11, 2025

Measuring the business impact of your AI investments is a critical step in understanding how best to deploy and leverage AI for business transformation. With Copilot Analytics, you can customize reports in Viva Insights to measure the business impact of Copilot usage against your business metrics.

Our Microsoft Customers and Partners (MCAPS) team used Microsoft 365 Copilot to improve seller efficiency and performance. The team provided training and enablement programs to all sellers to understand the technology and how to leverage it in their daily workflow. As early adopters of Copilot internally at Microsoft, the MCAPS leadership team was eager to understand the ROI of their efforts and how sellers were using Copilot to further improve sales performance.

By tracking Copilot usage and key sales metrics with Copilot Analytics, MCAPS gained insights into how Copilot improved pipeline hygiene, accelerated deal velocity, and increased revenue per seller across the sales org.

How, exactly, did they get there? Below are a series of Pro Tips from an MCAPS expert for getting the most out of your Copilot Analytics experience.

1. Identify and utilize key experts

It’s an old cliché but appropriate in this instance: you don’t know what you don’t know. When driving Copilot adoption and utilization, it’s crucial to identify key experts and stakeholders on the teams themselves. These experts can help you identify the right metrics to track and data sets to use to measure Copilot business impact.

For MCAPS, that meant keeping an open line of communication with sellers on the ground, the ones who stood to tangibly benefit most. In what areas did they hope to improve? Where did they expect Copilot to have the greatest impact? Once identified, these experts and champions helped identify the key metrics and secure the data needed for analysis.

Business knowledge and seller cohort knowledge are invaluable when coaching customers through deployment. This knowledge helps in understanding how different teams and organizations can think through their metrics and improve their processes.

2. Reach a critical mass

Before experimenting with different metrics, focus on driving adoption. The report helps organizations understand how higher users of Copilot compare to lower users of Copilot with regard to a specific business metric. Achieving a critical mass of users is essential for conducting meaningful analysis and identifying impactful metrics. As a general rule of thumb, organizations need at least 50 people to add to the report for a given function, ideally starting with more than 100. This provides enough distribution of Copilot usage to see meaningful correlations between usage and business impact.

Before making any meaningful conclusions on Copilot impact, MCAPS found that at least 50% of users should be using copilot daily. As adoption and usage grew within MCAPS, they found that the definition of high usage group should increase overtime, e.g. 60% by year 2.

Additionally, it can take time to see changes in business metrics because of upstream changes. For example, sales cycles can take weeks or months in some industries. If you measure Copilot business impact too early before the sales cycles play out, you won’t be able to identify changes to the actual time to close a sale.

3. Start wide then narrow down on the most meaningful metrics

After chatting with sales experts and stakeholders (see: Pro Tip #1), MCAPS focused on a few key performance metrics to track: revenue per seller, for example, percentage of deals won, cross-checking those with daily and monthly usage of Copilot. MCAPS is very focused on revenue generation, so focusing on individual metrics that account for revenue generation was a good place to start for this use case.

By the end of the trial period, MCAPS had seen a 9.4% increase in revenue per seller and a 20% increase in won deals for a cohort of sellers. Daily usage of Copilot by over 50% of sales professionals (30k users) has resulted in a 5% increase in the overall pipeline¹.

But an MCAPS leader stressed that these KPIs can fluctuate over time, and that while it’s useful to have a few foundational metrics to keep a close eye on, others can surprise you later in the process. Pipeline generation, for instance, ended up spiking more dramatically that they’d expected at the outset of the process.

We suggest you look at a variety of business metrics in your initial analysis to see which metrics are impacted the most with high Copilot usage.

4. Regularly refresh your data and deeper analyses

With new technology in a quickly shifting business landscape, it’s critical to fine-tune your process and refresh the data on a regular basis.

On the MCAPS team, their analysis is refreshed every month to ensure that the stats still hold true. This involves cleaning up the data and verifying the claims. Deeper analysis is conducted on features and adoption -- this includes exploring which features are driving adoption and how to improve usage among laggards.

Utilizing Microsoft Copilot Analytics for continuous monitoring and trend identification is essential for successful adoption and usage.

5. Use early adopter champions to drive further change

Modern Work sellers – as a cohort that’s keenly aware of the need to sell products, meet their targets, and stay ahead of the curve -- were already eager to adopt M365 Copilot. But even on this competitive team, there were a few members that stood out from early in the trial run.

High usage of Copilot among these sellers correlated with increased deal closures and revenue – a proof of concept that quickly spread through the rest of the group. Supporting early adopters and using them as your initial champions can drive broader adoption across your team and the organization as a whole.

Beyond these Pro Tips, we have a host of resources to get you take the next steps on your AI measurement journey.   