How Enterprises Strip Away GTM Complexity and Open the Door for GTM Engineers

As CMOs plan massive generative AI investments, much of that money is flowing into new software layered directly on top of old workflows. But buying point solutions is not the same as modernizing a go-to-market motion. Data suggests that companies outpacing their peers are redesigning their operating models first, rather than just plugging algorithms into broken systems.

Sathish Muthuswamy is the Senior Vice President of Strategy and Operations at the global enterprise cloud leader Infor and a veteran operations leader who has helped scale SaaS businesses to billion-dollar revenues at SAP. He oversees global go-to-market planning, P&L, and data strategy while leading an enterprise-wide AI transformation. His experience dictates that the path to unlocking growth starts with rethinking processes before buying tools.

"The mistake that many companies make is that they are tool-focused. The more you focus on the tool, the less successful you are. I've seen that companies and teams that focus on the processes and reengineer those processes using technology as an enabler have been far more successful," says Muthuswamy. That philosophy aligns with how top firms approach AI transformation. Under the 10-20-70 rule, only 10% of the work is about algorithms and 20% is data and technology, leaving 70% for business and people transformation. Yet the vendor environment makes it tempting to focus on the tech.

• Addition by subtraction: Walking through a recent Gartner marketing symposium, Muthuswamy noticed a floor completely flooded with AI vendors. He left the event more distracted than when he arrived, struck by how quickly native LLM platforms might subsume niche tools. "The more you add tools, the more complex the process becomes," he says. "We find in retrospect that deletion of tools and of process steps is sometimes far more effective than the addition of tools and of steps."

• Re-engineer and reintegrate: Muthuswamy offers a practical example about lead qualifying. Traditionally, a system routes a marketing-qualified lead to a sales development representative, who manually qualifies it before handing it to an account executive. Instead of automating that handoff, Muthuswamy’s team asks a more basic question. "What we realized was that a better problem to solve would be thinking, 'Do we even need lead routing in the first place?'" Viewed through an AI-native lens, some of those steps disappear. Rather than routing every lead to a BDR, a system can score, prioritize, and trigger direct outreach with personalized content. Such aggressive deletion maximizes generative AI potential by cutting out the human bottlenecks entirely. "So when we think about the underlying process and see from a customer point of view, we think about how we can rethink the entire process through an AI lens. We think about how we're able to fundamentally reengineer and reintegrate the entire process better."

That kind of simplification is upending traditional SaaS playbooks. In his earlier work running account-based marketing, Muthuswamy notes that teams often built seven or eight variants of the same campaign to serve different verticals. Today, research on upgrading software business models points to a different approach: building a single, generalized engine at the core.

For some organizations, that centralized engine serves as an AI-native moat. It handles the heavy lifting in the background while feeding a highly tailored experience on the front end.

• General engine, custom front: "The engine that we build is getting heavily generalized," Muthuswamy says. "The dichotomy is that the centralized engine we are building can serve a highly personalized front end. As you run it, the system is able to auto-generate and auto-heal to make sure it is precise to the customers it addresses." On that front end, AI allows for an extreme level of granularity. Recent analysis on AI-powered personalization and the future of personalization shows that top performers use data to tailor experiences down to the individual level. Muthuswamy notes that his team operates one generalized engine that still produces highly specific content for a given account, geography, or role.

• Hunting for unicorns: Executing a decentralized, AI-native motion often demands a hybrid skill set. Traditional silos struggle here. Pure technologists rarely understand the nuances of enterprise sales processes, while experienced go-to-market operators often lack the technical depth to build workflows on top of LLMs. Some operators are responding to this gap by defining a new hybrid role. "One role I see that is evolving quite significantly is the role of the GTM engineer," he says. "This is a combination of a traditional GTM person with a strong technology and an engineering background who understands technology natively as much as they understand the business process."

The cumulative effect of these changes—simplified workflows, centralized engines, hybrid talent, and decentralized micro-sprints—forces a complete rethink of corporate structure. Muthuswamy recalls a famous rule from Bill McDermott, the former SAP chief executive. McDermott once stated that only two functions truly mattered in a software company: engineering to build the product, and sales to sell it. Everything else was overhead. Muthuswamy believes AI flips that paradigm entirely.

"The two roles I think are going to be super crucial going forward are not engineering and sales," he says. "Engineering is highly automated through AI, and sales can be automated as well. It's actually product and operations. Product, because you need a clear value proposition for your customers. And operations, because the better you can leverage AI to make things more efficient, that will translate to better value."