People
Sanjay Verma
Partner & Managing Director
Silicon Valley
A conversation with Nabil Bukhari of Extreme Networks
Part 2 - Leveraging AI to accelerate platforms
The enterprise software industry is at an inflection point.
Artificial intelligence (AI)—once a competitive edge—is now an open-source commodity. The pace of innovation is relentless. And the barriers to entry are vanishing. In this new reality, technology alone is no longer enough to win. Profitability and sustainable differentiation hinge not on the algorithm but on how the algorithm gets used to deliver measurable value to customers. We sat down with Extreme Networks’ Nabil Bukhari, Chief Technology and Product Officer and General Manager to unpack the ways AI is reshaping the economics of software and the software-as-a-service (SaaS) industry and to learn how to unlock true enterprise value creation. This is an edited version of the conversation. Read part one of the conversation here.
The SaaS industry isn’t suffering from a lack of powerful tools; it’s overwhelmed by them. Siloed systems, fragmented architectures, and narrow point solutions are eroding productivity and inflating complexity. Extreme Platform ONE was designed to turn that fragmentation into focus by bringing networking, security, and AI together in a unified platform that simplifies operations, accelerates outcomes, and enables enterprises to scale value creation with clarity and control.
Let’s bring AI into this picture. How will AI shape the software industry, and what are the key success factors for profitability?
AI is an existential threat to incumbents in the industry. The main reason is that as a technology, AI and large language models (LLM) are undifferentiated. The technology is amazing, but everything you can do with it, anyone can do with it. You cannot create a moat around it. It’s for sale, right? You can buy from Microsoft or Amazon or someone else, but so can your competitors. In addition, the pace of the technology is moving incredibly quickly, so it’s very hard to sustain an advantage. Whatever you start to develop, by the time you finish the program or the product, a better technology will already be out there.
People talk about AI as a general-purpose technology like electricity, meaning that competitive advantage can’t come from the technology, right? You can’t be a venture capitalist and say, “Oh, these kids came out from school, and I’m going to bet $10 million on the algorithm they’re working on.” Back in the day, with Google, you could do that. Not anymore. The biggest transformative impact of AI is that it lowers barriers to entry in any market to almost nothing. Anybody can do what you’re doing—and without any of your baggage.
How will that affect profits?
In a big way. AI drives down costs, and that means it’s likely to drive down prices too. The way SaaS used to work was that you got economies of scale from delivery, and prices were relatively stable. Once you developed an application, the cost of offering it to the second, third, or hundredth customer was very low, and that was your profit. In the AI world, this is different; it’s the reverse.
There is going to be massive price erosion.
Because AI is moving so fast, everything that was differentiated today will be table stakes tomorrow, which means that the user’s expectations will continue to increase, but the price the user is willing to pay for that will continue to decrease. So, now, if you measure your success simply by number of users, might that—or might it not—translate into profitability? AI can have major impacts on your cost, but it will have an even bigger impact on your pricing.
So, if you cannot create a sustainable differentiation based on technology, what can you base it on?
You get an advantage only from using technology differently, and that comes back to customer value. It’s almost a philosophical issue. Imagine that every car is driven by AI and that all cars drive in the same, perfectly safe way. Then the only difference between you and me is where we’re going. Same with AI: The only question is what you’re trying to do with the technology to create value, and that comes down to data. Do you have a plan—like a destination for the car—and do you have access to data? Everything else is undifferentiated.
This is why the networking and security world will end up with only a few platforms. In the end, it’s not AI that matters; it’s the data that matters. The more data I have from you, the more value I can give to you. The criticality of data further increases the value of platforms over stand-alone products. If you and I have similar products, but I’m on a platform and you are not, I’m going to have data from more customers and from more different parts of each of my customers because I am dealing with them across multiple products.
Not only that: Platforms make data governance much easier. So, if all of your data is on one platform, you can apply common policies to it, I can audit it, and I can show it, which increases trust. I can work with you to analyze it to find patterns that will help you manage your revenues and costs; and the more different kinds of data I have, the more insights we can get, whereas if you have eight applications, each one of which does its data differently, then trying to secure it and manage it and prove it and analyze it becomes a hundred times more difficult.
What’s the way out of the coming profit squeeze?
Profitability, not user growth, is the name of the game. We have a framework for thinking about it, which we call ARC: three ways to deliver value.
A is acceleration: I’m going to do it faster. I’m going to do it more cheaply. I’m going to do it with fewer people. These are use cases where the bottom line is cost.
R is replacement: Replacement could be cost, or it could be risk. For instance, you used to do a job that required 10 people, and now AI can replace those people. Or you used to need a certain amount of resources to manage risk to acceptable levels, and now, with AI, you can get acceptable risk with a lower number or lower levels of capability of resources.
C is creation: Creation means doing something that helps you generate more revenue and value. The only sustainable things in the ARC framework are the use cases that fit in C, creation. If your entire value proposition is cost reduction, you’re living on borrowed time, because first, cost can go down only so far, and second, cost reduction through AI is undifferentiated.
In AI, where willingness to pay drops faster than costs, sustainable revenue comes from value creation. If your value is that I am helping you create something, that’s a sustainable value. And that’s value creation for customers.
And because of all the things I talked about earlier, I can do 400% more things and charge 40% less If you’re a stand-alone, you're done.If you’re a platform, it’s just beginning.
How do you think about value for customers?
I like to think about units of value. A unit of value is something a customer can measure, such as time saved, network uptime, error rates, costs, and insights. Thinking in terms of units of value also helps in the development of a product or an offering like Extreme Platform ONE. Rolling out a series of units of value helps bring the company's financial community over to your side. No Chief Financial Officer will agree to a plan—even a brilliant plan—that requires $40 million and three years to build and six years to break even. And it helps with your other big stakeholder, which is your sales team. You can’t expect a sales team to run with a huge new product that totally changes what the company is doing. The first reaction is fear. The team doesn’t know how to sell it. You must drip feed the product’s value to the sales force, one unit of value at a time.
The unit of value would be something you can deliver in one or two quarters, which works internally as well as with customers. If on Day One, we sat down and said we were going to change everything about the company, we would still be debating that now. When we came up with the platform idea, we didn’t explain it to the rest of the company until we were halfway there already. Sure, think through every possibility, and create this massive plan in your head—but release it per unit of value because that’s the only way. You need to think big but start small. You need to hurry up, but also wait for things to happen. That works internally—and it works with customers. One unit of value after another. Phase one has to be valuable on its own. Don’t fall victim to that engineering mindset wherein you don’t see value until phase four.
Where do you see Platform ONE going in the future?
For the foreseeable future, we will continue on the path we are on, which marries platform and AI, because, you know, we are only just introducing the platform. Now, success comes in multiple ways:
- The first level of success is the migration of existing users to it.
- The second level of success involves attracting more users.
- And the third level of success is that every user is doing more and more things on the platform.
As we go forward, as I said, the market will consist of only a few platforms. And I predict that those platforms will not stay closed and will not stay vendor specific. Platforms that stay vendor specific will die out. That tells you where my mind is: I want Platform ONE to be much more open, and I want to democratize the platform way before anybody else. I want to make it so easy for people to put their values on top of my platform that people will not want to create competing platforms, but just come to Platform ONE.
Now, this is the kind of thing that is hard for others in the company to understand. So, we’re back to the idea of units of value: One piece at a time, one unit of value at a time. But quite frankly, if you’re building a platform only for your own products and prices, you’re missing the point.
And then the last part is that in another few years, I want to get to a point where markets think about Extreme Networks, where they will not think about us as just a networking company, but a platform company. A platform with AI embedded in it that serves our customers in fulfilling their needs in a flexible way, in an open way, and with an ecosystem of apps that add value on top of Platform ONE.
Download the article here.
Disclaimer: This article is based on an interview conducted by AlixPartners. The content is intended solely for knowledge insight purposes and does not constitute an endorsement or promotion of any specific products or services.
