When your company is investing in AI, it can be hard to gauge ROI. What use cases will generate the most value? How will you measure ROI? With these big questions looming, it’s understandable that leaders might feel uncertain about answering one of the most basic questions: how much should I spend?
Adding to the complexity is the fact that we’re operating in a fast-changing AI market, as innovations constantly disrupt the status quo and models become commoditized. Last month it was DeepSeek—what will tomorrow bring? In the midst of so much volatility, it’s natural to wonder how this will affect the AI market in general, and prices specifically.
But here’s the thing: No matter how uncertain and fast-moving the environment is, you still need to drive value from your AI investment. So don’t get distracted by factors you can’t control.
At Microsoft, the core of my job is helping customers get the most value out of their AI investments. As part of that effort, we’ve been evolving our approach to pricing AI. We believe the best way to optimize your organization’s AI investment is to focus on three key elements:
Flexibility: Give yourself the freedom to learn how AI drives the most value for your organization—and to adjust accordingly.
Scalability: Set yourself up to expand as rapidly as possible as high-potential areas of ROI become clear.
Cost efficiency: Match what you need to what you buy as closely as possible.
Here’s the thinking behind our approach and how it plays out in our pricing model.
The Price That’s Right
Microsoft recently released a new offering called Microsoft 365 Copilot Chat. It’s the simplest way to illustrate what I’m talking about because it’s the simplest price: free. Copilot Chat is AI chat for our Microsoft 365 commercial customers, available at no charge. It’s powered by GPT-4o, one of the most sophisticated mainstream models available.
I bring this up not to pitch you our product, but to show how it addresses those key elements that every organization needs to consider when investing in AI.
Flexibility: A free offering offers a low-stakes way for everyone in an organization to learn how one of the most foundational AI use cases—chat—can help them become more productive in their day-to-day work. Organizations can then see how those collective productivity gains start to generate ROI.
Scalability: As the paths to ROI start to become clearer, organizations can see where to focus additional AI investment.
Cost efficiency: This one may seem a little obvious, but it’s worth noting that what you’re getting for free is an on-ramp to discovering the value of AI at scale. You’re not getting access to maximum-strength AI—that would be our flagship Microsoft 365 Copilot offering—but it’s enough to start identifying where potential value lies.
At the next level of AI capability are agents, and here we took a page from the cloud-services pricing playbook. Along with the free Copilot Chat, we’re also offering metered access to AI agents—you pay as you go based on the volume and complexity of tasks the agent carries out. We’re aiming to lower the barrier to entry for one of AI’s most exciting uses by once again addressing those three core needs.
Flexibility: With pay as you go, you’re not making a big financial commitment before you know how valuable agents will be for transforming your organization’s business processes. Employees have a low-risk way to experiment with agents and learn what they’re capable of.
Scalability: A metered approach lets you expand quickly as value becomes apparent. The access is already there, and different teams can respond with agility as their AI needs emerge.
Cost efficiency: You pay only for what you use.
The third leg of the AI-pricing stool is probably the most familiar: subscriptions. You pay a fixed fee for unlimited access to AI. It’s simple and straightforward. This is how we price Microsoft 365 Copilot. Once again, it’s an approach that addresses the three core needs of organizations investing in AI—in this case, most likely the ones furthest along the AI adoption curve.
Flexibility: With a flat-rate subscription, you don’t have to worry about how much AI you’re using. The faucet is always on.
Scalability: In any part of your organization where ROI is becoming clear, you just add another seat.
Cost efficiency: Pay-as-you-go agents make sense up to a point, and that point is where you’re spending more on metered use than you would for a subscription. It’s easy to see when you’re hitting that tipping point, and it’s easy to switch to a flat-fee subscription.
Finding Your Mix
How your organization invests in AI probably depends on how far along you are on the AI adoption curve, but for many companies it will involve a mix of these three approaches.
Whenever a technology is new, some companies want pay-as-you-go pricing. They’re excited to get moving quickly but don’t want a big commitment or a lot of risk. Typically, once they start using it and seeing the value, they come back and say, “Actually, I don’t want consumption-based pricing any longer. Will you give me a flat rate instead?” I expect the same thing will happen with AI.
There’s another approach that’s starting to get some attention. Outcome-based pricing ties the cost of AI services directly to specific results—the fee could be based on, say, the number of resolved conversations in customer support. This approach builds ROI right into the payment structure. But for it to work, both the provider and customer need to be perfectly aligned on what constitutes results. As these issues get smoothed out, we expect this model to work for some applications.
Summing It Up
There’s a lot going on right now with AI (and the world in general). Don’t let that paralyze you.
The big picture is that AI is essential to your company’s future, whatever industry you’re in. Every employee will have an AI assistant, and every business process will be transformed by agents. Right now, it’s essential for your employees to explore and innovate with AI. By staying focused on flexibility, scalability, and cost efficiency, you’ll put your organization in the best position to most effectively invest in AI. It will help you tune out the noise and concentrate on your path to maximizing the value of AI, regardless of the news cycle.
For more insights on AI and the future of work, subscribe to this newsletter.
