An AI tool that helps an employee finish a task faster is useful. An AI-enabled process that changes how you build, sell and support your software can change your entire business.
That distinction was my biggest takeaway from Greg Buzek’s presentation at BlueStar’s VARTECH event last week at Mohegan Sun in Uncasville, Connecticut. The IHL Group president challenged attendees to look beyond individual productivity and consider how their organizations would operate with AI at the core of the work.
For independent software vendors (ISVs), that challenge has two parts. You need to improve your own operations, then carry what you learn into your product. Neither happens by buying licenses or getting better at prompting. Both require deciding what should change and proving the change creates value.
Start with the workflow, not the tool
Buzek described a familiar pattern. Organizations bolt AI onto existing processes without reconsidering the processes themselves. Employees draft emails and summarize meetings faster, but the underlying handoffs and decisions stay the same.
McKinsey research he cited shows the cost of stopping there. Only 11% of leaders surveyed said their organizations had redesigned roles, workflows and operating models around AI. Of that group, 48% report realizing enterprise value, compared with 13% of organizations that simply hand employees AI tools.
His starting point was to examine any activity that consumes more than four hours a month and ask how it would work with AI at the center and all the relevant knowledge available. “Take the time to rethink,” he said.
For an ISV, security questionnaires and request for proposal (RFP) responses make a good test case. Having AI polish an answer addresses one step. Connecting your product documentation, prior responses, compliance evidence and approval rules could take you from intake to a review-ready response. Measure the whole process: How long does a response take, how much staff time does it consume and how often does someone correct a technical or compliance claim?
Give AI the context to do useful work
Buzek called a second brain probably the single biggest unlock in AI. He described it as one cloud-drive directory that agents can access, with a roadmap to where everything lives. Because his system knows his organization’s offerings and discount schedules, he said he can go from a client call to a finished proposal in about 15 minutes, though he usually still adjusts the result.
That’s his example, not a guarantee, but the lesson travels. AI gets more useful when it can draw on what makes your business different. For you, that means product documentation, application programming interface (API) specs, release notes, known issues and support history, kept accurate, accessible and appropriately restricted. If your senior engineers routinely fill gaps your documentation doesn’t cover, automation will hit those same gaps. Knowledge management belongs at the start of an AI initiative, not the end.
Expand autonomy with accountability
Buzek then showed his own agent team of 15 specialized agents covering research, writing, marketing, reporting and security, coordinated by a routing agent that serves as his chief of staff and supports IHL’s five full-time employees. Across 3,600 agent runs since April 1, he said, the agents have averaged a 1,500% improvement over human performance on the same work. That’s IHL’s internal measure, not an independent benchmark, but it shows how far redesigned work can move beyond assisted work.
The practical lesson for you is to start with a bounded responsibility. A support agent might first triage incoming tickets and suggest known-issue fixes before it ever responds to a customer. More authority should follow demonstrated reliability, clear permissions and a defined escalation path.
Buzek’s governance model made the split explicit. Humans set goals, approve tools and permissions, define risk tiers and keep final escalation authority. Agents execute within those boundaries and hand exceptions to a human owner. This matters even more for ISVs, since agents that touch customer data fall squarely within your System and Organization Controls 2 (SOC 2) scope. You should be able to show an auditor, or a customer’s security team, what each agent can access, what it can change and where its actions are logged. Everyday habits count too. Buzek warned against sharing AI conversations through public links, a risk that became real in 2025 when thousands of shared ChatGPT chats surfaced in Google search results.
Measure the business result
Unclear value sinks projects. In a June analysis, Buzek highlighted Gartner’s prediction that over 40% of agentic AI projects will be canceled by the end of 2027 due to escalating costs, unclear business value or inadequate risk controls.
Faster ticket triage is promising but incomplete without resolution quality, reopen rates and customer satisfaction. Establish a baseline, pick a limited workflow and compare results after the change, counting the time people spend reviewing outputs along with tool costs. Hours saved only become value when you redirect that capacity deliberately, toward your roadmap, your backlog or growth.
Turn internal experience into product advantage
Buzek’s sharpest point was a warning. What happens, he asked, when a customer that uses three functions of its point-of-sale (POS) system decides to vibe-code its own? For ISVs, that isn’t hypothetical. Any customer relying on a narrow slice of your product can now ask whether AI lets it build that slice itself.
Your defense and your opportunity are the same. The internal work described above teaches you what AI needs to perform reliably, including clean data, integrations, permissions, monitoring and constant refinement. Those are exactly the things a customer’s homegrown tool will lack. Build that knowledge into your product through embedded agents, reporting and analytics, and you’ll deliver value customers can’t easily replicate.
I left with a practical challenge for ISV leaders. Pick one recurring workflow, define the result that matters and give someone ownership of improving it. Measure the full outcome, then apply what you learn to the next process and to your product. That’s how AI adoption becomes an operating advantage and a reason customers keep paying you.
