I am so excited to talk to you today about Microsoft 365 Copilot and the new pricing and what it means for organizations. There are three things I want to talk about today. The first item that I want to talk about is why is Microsoft 365 Copilot the cost that it is.
The second one that I want to talk about is what's the value of Microsoft 365 Copilot. Can we quantify the ROI? I can we understand whether this is a worthwhile investment for many organizations and employees? The third item that I want to talk about is how to deploy something like Microsoft Three Six Five Copilot and how that impacts this evaluation of the price of the value and the return on investment that something like Microsoft Three Six Five Copilot provides. Let's start with item number one.
Why does Microsoft 365 Copilot cost $30 per user per month in US? Dollars today? The truth is Microsoft three six five copilot isn't just giving you access like Bing for business or Bing Chat Enterprise does to an LLM model in a responsible, safe and secure way. What it's actually doing is it's giving you this incredible capability to reason over your organizational data, the relationships between that data and the relationships between people and that data.
And because it provides this extra capability on top of the baseline capabilities of the LLM or these large linked models or generative AI, it's why it costs Microsoft dollars for usage, for powering. It for providing a relatively quick response rate to make sure that you're getting the information you're requesting quickly and reasonably. Especially because so many of these types of uses for things like generative AI are more natural.
So it leads to much higher usage rates, which means Microsoft needs to provide the right scaling price so that they can make sure they provide the right performance for that impact. Now, I do think that these things will improve over time both in terms of capabilities and performance. But I want to note I actually think it's very reasonable and it's actually much better pricing than I thought it would be.
And I was worried about it being for the public market. Now, I want to acknowledge that $30 per user and US dollars per month is a lot for a lot of customers. That could be a two times increase in cost, more or less from your E Five suite or something that you have like that.
It could be almost a three X cost if you were talking about business premium. So at those types of cost impacts, is it worth it? So item two that I wanted to talk about was what is the ROI of something like Microsoft 365 Copilot? And so I've wanted to frame this in a few ways. The first one is this is not a new area of academic research.
The good news is there's been a lot of work, really hard work from lending many academics around the cost modeling as well as the productivity gains that you should expect from something like ChatGPT or these generative AI tools. And the good news is, I think we actually have a growing consensus that there is a very tangible and notable impact around these things. Some of that research is more advanced when it talks about things like GitHub Copilot and the way people use it for coding, where we see it's not uncommon to see like two times increase in productivity and much greater quality levels on the other side.
But we also see that even with general employee management scenarios. So what you do is you essentially grade the quality result on the one end and you look at the productivity outcome and how much more efficient people are on the other end. And what we see in that early stage research, which we'll put some links, maybe even throw up a diagram for you for reference here, is that the treatment effects of productivity when you use these types of tools like ChatGPT or tools of that nature, is around a 40% overall productivity increase.
And we're seeing around 18% to 20% in terms of, like, the grading quality, where people say it's higher quality. And so those are pretty substantial findings already in this early stage. Now, it's important to note what we're talking about when we talk about Microsoft 365 Copilot is this $30 per user per month price, which is actually for something beyond something like ChatGPT.
If you want to responsibly and securely use a tool like ChatGPT, you can use Bing Chat for business, which is also now publicly disclosed. So you can use this amazing tool that allows you to connect with your generative AI, to add images, to do text, to do all the things you want to do for conversation support, et cetera. You can use that with ChatGPT through Bing Chat.
Right? So Bing Chat Enterprise is the way the placeholder starting point there. Now that's different than what actually Microsoft 365 Copilot provides. Because if this is where that experience of responsible and secure AI collaboration starts, what Copilot does is it reasons and understands the connectivity between your data and the people in your organization.
So there's this concept of the graph and this graph is what adds a ton of value both for Microsoft 365, but also for Microsoft 365 Copilot. So think of it like this. We start at this baseline of 40%.
We're already at 40% productivity gain and probably around 20% for quality. And that's just with Bing Chat for business. So please go use Bing Chat Enterprise.
It's amazing. The second thing that you want to do is then add Copilot. Now, you're not going to do this for all users, we'll talk about that, but you want to start with some users that are typically super users super collaborators or we often describe them as people who require a lot of digital fitness.
So if you think of your organization, there are some people who being really digitally fit, adds a lot of value. It tangibly works for them because they work on a lot of digital assets, they work on a lot of digital documents and document processing and collaboration, high velocity digital collaboration. And those people are almost always over leveraged, which means in an organizational context, if we can scale those employees, the return is not a linear line of value.
But it's also really important because if we can improve productivity of those employees, we also know, because they're super collaborators, highly networked and over leveraged, that they are almost certainly going to improve those around them in terms of quality and objective outcomes. So what I'm saying here is there's this really fun way that we often talk about the value of collaboration and we often describe it as one plus one equals three. And the reason we often describe it as one plus one equals three is because if you take two people and their experiences and their knowledge and their expertise and their skills and you put them together, what they produce isn't as good as what each of them could do on their own, right? And then you combine, that what it is.
Is it's something that's better because of the collaboration itself? So this is sort of the idea that the sum is greater than the whole sorry, is greater than the sum of the parts, right? So that's the idea of collaboration. And we use this term a lot actually when we talk about communication outcomes and collaboration outcomes within organizations and enterprises. Now, here's the really interesting thing.
If you take that funny math of one plus one equals three and you instead think about how AI impacts, that, the way it actually works is it's one times AI plus one times AI equals exponentially more than three in our example. And the reason for this is because each person uses AI in different ways. And remember, AI is basically a technology that gives us access to skills that we might not have had before.
It gives us more accessibility to more advanced skills, whether they're technological, whether they're literary, science and educational ones and so on and so forth. So it gives us those capabilities and allows us to do that while reasoning over the organizational data, its connectivity and more. And so what this means is when you think of two people collaborating, it's really exponentially more because each one uses AI in different ways and via collaborating with each other and AI.
Because each one is using Copilot in that scenario of collaboration, they're actually going to be learning from one another and finding new ways to use AI and new ways to benefit from it. So this is already what we're seeing with Microsoft 365 Copilot in usage with these private preview customers. There's so much exponential value that starts to happen through this collaboration.
So what do we know? We know that if you take the users that are over leveraged, highly collaborative, highly digital today, and you make them more digitally fit, right, and give them capabilities like Copilot, these new capabilities, you're not going to see something like a 40% increase in productivity. You're going to see a two or three times type productivity increase. Now, here's the other interesting thing about this.
What we're also seeing is that the way collaboration works is there's a lot of overhead. One of the reasons we don't go one plus one equals three. Like we don't do that with every scenario is because there's costs, right? There's miscommunication, there's communication, overhead costs, there's time and availability issues and things like that.
So we know that there's a cost set that comes with collaboration. Well, here's the interesting pattern. When you use Copilot, we actually see two things occur at the same time.
We see a reduction in collaboration signals for a lot of scenarios, right? If you take user stories or scenarios and journeys that used to be collaborative, they don't require collaboration with another human, which actually saves a lot of money. They're able to get their responses, their answers, et cetera, through the AI agency. This Copilot, they get what they need without having to bother another human.
That's a huge ROI. Think of the scenario where I'm looking for a particular asset or resource. I can do that much faster with something like Copilot and I don't need to bother someone when I want to do that exercise.
Those are substantial, pure play benefits, right? Boom, boom, boom. Lots of benefits that we're getting from that. Now, the other side of it is it's not like collaboration goes away.
In fact, from a pure data perspective, we still find that people collaborate as much, if not more, when they use something like Copilot. The difference is the form of the collaboration is much more focused and much more valuable and oriented around adding that extra layer of value versus drafting versus those early stage discussions that often the AI tool can do a better job of. So this means that not only is it one times AI plus one times AI for the exponential benefit, but we're actually seeing that there's both a cost reduction in baseline collaboration that's really not highly valuable.
So low value collaboration is basically shifted out, which creates a huge benefit. And then the second one is the high value collaboration becomes even more valuable because of the nature of what people are able to do with these tools. Again, reducing potentially the number of people that you need in a meeting.
That's really important as well because that, again, is a lot of hidden costs, right, with larger meetings, longer meetings, if you can have shorter, more focused meetings, those are with less people. Those are huge benefits. We have lots of tangible evidence that shows that with Viva insights, research and other things like that.
Okay. What I'm basically saying here, and this is, based on our numbers, based on everything I've just described, is it's very reasonable to see a three x or a four x productivity gain in terms of when you break down? All these types of math scenarios on an employee that uses something like copilot, which makes the price increase, if it's two times, a very reasonable consideration. Now, I want to make one other comment, so I'm hoping that that little section, which was a lot, helps explain a little bit of the ROI.
But the point number three was, how would you roll something like this out? And I've already hinted at the fact that on point number three, you don't want to roll it out to every employee. You want to target specific individuals. Well, of course that helps manage the costs, but it's also important to understand how you get started.
So let me frame something for you. When you're learning AI and you're trying to address it within an organization, we have lots of experience here with helping customers, whether it's with Azure OpenAI stuff, or whether it's using the previews of Copilot, Microsoft 365 Copilot, or some other things that we do, we know that you need to start with personal productivity. So the way the journey works is you learn how to use AI better yourself.
And as I learn how to use AI myself, I can then work with my team who might also be on that journey of learning how to use AI. And now we can get to the team level. You don't just jump to the team level.
You have to start with personal productivity, understand how to use that for AI, then work to the team level. And then eventually, once you have a few teams that understand how to use it, then you go to the departmental level. And then as you understand how a few departments work, you get the idea.
It goes eventually to the organizational level. You can't just jump to the whole organization, everyone being licensed for Microsoft 365 Copilot and expect success. Please don't do that.
It would be a terrible waste of money. You want a rolling pilot system because the way it's going to work is, again, you're going to start with something like Bing Chat for Enterprise. Right? This capability, which is a responsible and secure way to use large language models today within your organization, that's your baseline.
Give that to everyone, that's fine. Right? It's probably included in your Microsoft 365 licensing because it's included in some suites already. So give them that.
And then when you look at the data for that, there are some people that are more actively using it. Those are the people you might want to consider adopting into the Microsoft 365 Copilot suite. These other people within that subset might also be super collaborators.
The people you might use Microsoft Teams might license for Microsoft viva insights. Premium. So when I say teams I meant Teams Premium like the premium features.
So those people often is a no brainer to give them better digital tools because they're super collaborators. They're over leveraged those people. If we can again make them more successful, we're going to get tangible benefit out of it.
So the same people you license for those tools may be now licensed for Microsoft 365 Copilot, which again in that grander scheme, when these people have many, many different licenses assigned to them, is not a very big uptake in cost and quite substantial in terms of its improvement. So that's the high level there. Now the second piece to this is to think about, well, what does that mean from a change management perspective as an organization? And so I've already highlighted that there's this rolling pilot that's going on and I think that the way we found most success is organizations look at it as sort of a digital fitness initiative.
There are some people in your organization that need to be really digitally fit or that we would like to be really digitally fit so we can benefit from that right. As an organization. There are some that don't need to be that digitally fit.
And there are some people in this Ven diagram who really aren't going to benefit for quite a while on the broader Microsoft 365 Copilot experiences because they just don't have a lot of work documents or business documents or business processes that they're going to work on, where these reasoning capabilities that it provides are going to be highly valuable. Instead they're the people who are probably going to stay on something like Bing Chat Enterprise. It's still vastly useful for coordination, emailing scenarios like all sorts of things.
So you could still get a lot of benefit there. But they may never reach that point until again, Microsoft evolves what Microsoft Copilot represents today. That was a lot of content in a very short amount of time, but I hope it gives you a bit of an idea on how to think about Microsoft 365 Copilot.
Don't just look at it and go, ah, $30 per user per month and either whoa, that's a lot of money and so you slow it down or hey that's nothing, let's go license everyone, be targeted, be considerate of how to think about this and try and measure the kind of tangible impact it's having. Both with qualitative things like surveys as an example as well as some baseline measurements on how long certain tasks take today and how long they're going to take a bit later. Again, you can do that through surveys or preferably through other measurement tools.
So I hope this has helped you maybe convince a manager or a leadership team you always can reach out to us if you have more questions. I just wanted to share why I'm so excited about Microsoft 365 Copilot licensing now being public, as well as some of the things that we've already been talking to customers about when it comes to planning for this particular investment.
ā¶ļø Watch Is It Worth It + How Much Will Microsoft Copilot Cost Now
