Wolf&Grayson EditorialMany companies still struggle to translate AI experiments into meaningful revenue growth, even as AI becomes a centrepiece of corporate innovation strategies. For Gus Byleveld, Fractional Chief Revenue Officer at Accrete Consulting, the issue is the operating model surrounding the technology. “Revenue is a system, not a department,” Byleveld says. “Growth becomes predictable when you tighten the loop between customer behavior, what you learn, and what you change.” AI has the potential to compress that loop dramatically, but only if organizations redesign how decisions are made.
His view reflects more than two decades of experience scaling B2B technology businesses. Over 25 years working at the intersection of strategy and execution, Byleveld has focused on building repeatable growth engines that connect product, sales, and customer success into a single revenue system. AI, he says, becomes transformative only when it strengthens that system rather than sitting beside it.
AI Only Works When the Business Model Is Clear
“Ambiguity is the silent killer,” he says. “When teams avoid clear choices about who they serve, what they will not do, and what outcome they stand behind, execution becomes thrash. AI does not fix that. It accelerates it.” In practice, unclear strategy tends to surface quickly once AI tools are introduced. Data models can highlight patterns, but they cannot compensate for unclear priorities or conflicting incentives across teams. When those issues remain unresolved, automation simply magnifies existing confusion. Byleveld’s perspective reframes AI as a forcing function for clarity. Businesses must define the decisions they want AI to improve and the outcomes those decisions should drive. Without that foundation, the technology becomes ornamental rather than operational.
From Feature to Decision Engine
One of the most common traps Byleveld observes is treating AI as a product enhancement rather than as part of the decision-making infrastructure. “Teams build demos and copilots, but they do not change how pricing gets set, how pipeline gets qualified, how churn gets predicted, or how product priorities get chosen,” he says. “If decisions stay the same, outcomes stay the same.” To generate real impact, AI must sit inside the mechanisms that shape revenue outcomes. That might include identifying which deals deserve sales focus, predicting expansion opportunities within enterprise accounts, or signaling when customers are at risk of churn.
Equally important is ownership. If AI initiatives remain inside labs, centers of excellence, or temporary innovation teams, they rarely reshape the business. “Real change requires a leader who owns a business outcome and has permission to redesign the process around it,” Byleveld says. Data quality also determines whether these systems succeed. When companies rely on summarized reports or incomplete information, AI produces confident but misleading outputs. The most effective implementations draw from raw behavioral signals across customer interactions, product usage, and commercial activity.
The 90-Day Model for AI-Driven Revenue
For organizations eager to move from experimentation to results, Byleveld recommends a focused redesign rather than a sweeping transformation. If tasked with making AI visibly improve a company’s business model within three months, he would start with a single decision that directly influences revenue. “Not ‘use AI in sales’ or ‘use AI in support,’” he says. “Choose one money-making decision such as qualification, pricing, renewal risk, or upsell targeting. Define what ‘better’ means, then instrument the signals that drive it. If you cannot measure the decision, you cannot improve it.” The next step is to turn both the product and go-to-market (GTM) engine into a learning system. A small capability should be shipped quickly to generate new behavioral data. Teams then operate on a rapid feedback cadence: what was learned, what changed, and what should be tested next.
Byleveld favors a “startup within the startup” structure. A small cross-functional team drawn from product, data, revenue operations, and GTM owns the outcome and has the authority to change workflows. Just as importantly, that team must have permission to say no to initiatives that cannot demonstrate measurable impact within the 90-day window. “You are not trying to ‘do AI’ in 90 days,” he says. “You are proving a new way to make money and building the muscle to repeat it.”
The Rise of Adaptive Business Models
Looking ahead, Byleveld expects AI-native companies to evolve beyond static software products toward continuously learning operating systems. Pricing, packaging, and expansion strategies will increasingly respond to real-time usage patterns and measurable outcomes. Revenue operations will also become more predictive. AI systems will surface next-best actions, highlight expansion triggers, and detect friction across the customer lifecycle before it affects retention. At the product level, AI-driven onboarding and personalization will shorten the time it takes customers to reach meaningful outcomes.
Ultimately, AI-enabled innovation is less about new tools and more about how quickly a business can learn and adapt. The organizations that succeed will be those whose business models evolve faster than the markets they serve. “The companies that win will treat AI like a new team member,” Byleveld says. “It has defined inputs, defined decisions, guardrails, and a feedback loop that helps it improve.”
Follow Gus Byleveld on LinkedIn for more insights.
