AI has created a vast new revenue opportunity for software firms. We estimate that agentic AI adds $3 trillion to the global addressable market. But seizing this opportunity requires a fundamental reinvention of both product strategy and company operating model.
This reinvention is a significant—and urgent—challenge. Although very few organizations are “vibe coding” to fully replace core platforms, many will reduce their software spending to fund AI initiatives. For many software providers, the result will be stagnation and compressed multiples.
In this new era, winning is about risk, experimentation, and being product-obsessed—behaviors more typical of a company’s founding stage. The prize for this transformation is significant: a combination of AI-fueled growth and productivity improvements that make “Rule of 60” performance (annual revenue growth plus a margin exceeding 60%) possible.
Stay ahead with BCG insights on artificial intelligence
An Internal AI Startup
To kickstart the AI transformation, CEOs must think of the business as two connected engines.
A ruthlessly efficient SaaS business. AI must be a “force multiplier” within the existing business, driving a higher run rate per employee. This is not about purchasing and deploying licenses for AI tools and hoping they get used: roles, workflows, and the organization chart must be reshaped, starting with the processes with the biggest impact. For most organizations, product and engineering are the right places to start.
The goal: An increase in operating margin of 20 percentage points, with much of the benefit delivered through more-efficient growth instead of cuts.
A product-obsessed AI startup. The enhanced cash flow from the existing business should be reinvested in a new ring-fenced innovation unit. This must develop a new customer value proposition, typically high-risk/high-reward and vertical- or end-market-specific.
The revenue impact from a customer moving from the core SaaS product to the AI-enhanced offering should be a factor of two. If it is just 10% to 20%, then the product is either incremental or underpriced.
The goal: Seizing a share of that $3 trillion-plus growth opportunity.
Managing the AI “startup” requires adopting or returning to a founder’s mentality. This switch is vital. It’s no coincidence that the few SaaS companies that have built strong AI businesses are often founder-led, including Intercom and Notion. CEOs who are not founders must embrace the mindset and start building. All CEOs must adjust their style as they move among the parts of the business.
It is also critical that the AI unit is judged by different metrics, or it will be suffocated by traditional SaaS optimization and cost discipline. It should have its own P&L. The split into two businesses helps create a clearer narrative for investors.
A Head Start... for Now
This optimistic strategy may seem disconnected from the current gloom around incumbent SaaS, particularly among investors. The pessimism is misplaced. Incumbent software companies have important advantages:
- Deep domain expertise, access to customer data, and intimate knowledge of customer workflow.
- Established, trusted customer relationships.
- Embedded systems and connectivity.
However, these advantages are inherently transitory: if incumbents don't build great AI products, their customers will quickly shift loyalty—and migrate data and workflows—to more innovative startups. That is why SaaS players must move forward with this twin-engine AI strategy.
The AI playbook still has many blank pages. The SaaS CEOs who seize the AI opportunity will not only write this new playbook but also capture an important share of the huge value AI is creating.
