The ETR Insights hosted a panel discussion between cybersecurity executives discussing Palo Alto Networks, a broader transition from hardware to software-based firewalls, and the increased demand for AI-driven security measures. Prisma Access and Cloud simplify infrastructure monitoring and offer superior network segmentation, with panelists pointing to healthcare and IoT-intensive environments. Competitively, Netskope and Cloudflare offer specialized capabilities in client-less integration and web application firewalls. These executives also voice concerns over pricing and vendor lock-in—noting Cisco’s competitive moves on cost—and emphasize the importance of flexible subscription models. Generative AI presents data privacy and compliance challenges, currently managed independent of cybersecurity, though panelists see an opportunity for Palo Alto to expand its offerings to include AI oversight, governance, and data classification solutions. 

During the panel, the discussion turns to pricing. All agree that Cisco, Check Point, and Palo Alto are expensive, though on that front, Cisco does appear to be making a play for Palo Alto’s business. “Generally not something you see in relation to Cisco, but with where Palo has priced themselves out, to get everything we need from a full ingress/egress perspective, it is much more advantageous from a cost perspective to go with Cisco. Not saying we’re going to, but they’re making it much more compelling.” For our guests, the long-term equation involves balancing upfront spending with lower labor costs and hardware maintenance expenses over time. “We had a separate initiative to show return on investment on the software product we’re working on, and it shows that even though there is an uplift—an uptrend of spend due to engineering effort at the beginning—eventually you will reduce FTEs, because you would not need as many people to monitor things.”

The seamless integration and predictable product roadmaps can justify bigger budgets; large enterprises are looking at “efficiency, bringing value, and the time to which consultants can implement it.” One company, for instance, relies heavily on Splunk, backed by a team of specialized consultants and custom-coded enhancements. While cost and vendor lock-in are of concern, this panelist finds 

that Splunk’s framework, with its established network of integrations, delivers enough value to justify continued investment. “Any area of efficiency that Splunk is missing, they hire external developers to build out a process such that we can still continue utilizing Splunk as opposed to transferring to something else.”

Our guests have begun to demand more flexible software subscriptions, and are pushing their vendors to accommodate overlapping contracts without additional charges. “We have been building that into any contract, that there’s going to be a period where we’re not paying you for your service, because we know that we’re not going to be utilizing it, either partially or fully, until X amount of date. Then once we hit that date—which usually from a project planning perspective coincides with or around the termination of the other contract that we’re getting rid of—that has been how we’ve approached it.” They find vendors usually accept grace periods of three to six months as standard practice. “I think most of them understand that that’s how budgets work, right? That you can’t go to the executive team and say, for six months, we will be learning a new product and a new way of doing things, so you need to pay for two different products.”

With generative AI, these executives continue to grapple with data privacy and regulatory compliance. “I can’t imagine being in healthcare and trying to use gen AI solutions these days.” While they might rely on Palo Alto for cybersecurity support, most generative AI initiatives in these companies are presently managed through other specialized tools or cloud providers’ controls. “We have had to put some guardrails in, but it’s also been done more on either the Microsoft Azure side of the world, not on the Palo side of things.” Initial enthusiasm around LLMs has given way to measured implementation, with a focus on safeguarding against potential data breaches; to that end, some panelists see opportunity for cybersecurity vendors like Palo Alto to expand into AI oversight, especially as state and international regulations continue to evolve. “The data classification piece, I think, is probably the most important right now. A year ago, people were using any data sets to just try things out. Today, there’s a lot more guardrails and general governance around that.”