Service-as-a-Software: The Next AI Frontier

Artificial Intelligence (AI) has woven itself into countless industries, but many enterprises still struggle to understand how to effectively harness AI for a sustainable competitive advantage. With the right approach, enterprises can go beyond simply implementing AI to creating transformative workflows and business models. This playbook serves as a comprehensive guide for both enterprises looking to adopt AI and providers aiming to serve them effectively.

Covered in this article

Understanding the AI Advantage in the Enterprise
Part 1: The Micro Logic – Unbundling and Rebundling Work
Part 2: The Opportunity – Moving to "Service-as-a-Software"
Part 3: Enterprise Context – Performance, Not Promise
Part 4: Workflow Capture – Economics of Enterprise AI
Part 5: Business Model Advantage – Service-as-a-Software
Part 6: Challenges – Hype, Organisational Disconnect, and Lateral Attacks
Part 7: Competitive Advantage – Creating Moats and Expanding Accounts
Part 8: Defining Winners and Losers in Enterprise AI
Conclusion
FAQs

Understanding the AI Advantage in the Enterprise

Adopting AI effectively involves understanding three critical elements:

1. Micro Perspective - Unbundling tasks traditionally done by humans, and identifying how to optimise or even automate these.
2. Context - Recognising past missteps in digital transformation and avoiding pitfalls.
3. Macro View - Examining broader forces in the enterprise ecosystem that shape decision-making and power structures.

This playbook is organised into eight sections, each focusing on essential aspects of enterprise AI and offering a step-by-step approach to successful adoption.

Part 1: The Micro Logic – Unbundling and Rebundling Work

Unbundling and Rebundling Tasks: AI provides a new approach to workflow management by unbundling tasks typically performed by humans and reassigning them to software. By using AI-powered tools, enterprises can streamline workflows while maintaining the strategic value added by human decision-making.

For example, in corporate travel booking:

• LLMs can handle knowledge work, such as generating itineraries or listing preferred hotels.
• AI agents can take over managerial roles, learn organisational preferences and finalise bookings based on past interactions.

This shift enables enterprises to not only automate tasks but also redefine how tasks are interconnected and executed.

Part 2: The Opportunity – Moving to "Service-as-a-Software"

With advancements in Generative AI, the concept of "Service-as-a-Software" is emerging. This model transforms human-led workflows into AI-dominant workflows where key tasks and decisions are executed by software.

Stages of Transformation:

1. Service-Dominant Workflow: Workflows are still human-dependent.
2. Unbundling: AI starts replacing simple tasks.
3. Componentising: Tasks become software modules, enabling API integration.
4. Rebundling: Workflows are re-imagined, potentially forming new models.
5. Software-Dominant Workflow: Human tasks are minimised, creating a workflow managed predominantly by software.

Part 3: Enterprise Context – Performance, Not Promise

Unlike past technological waves, AI adoption in 2024 requires tangible performance outcomes. Enterprises are cautious, preferring solutions that promise concrete ROI over mere potential. With LLMs undergoing a transition akin to broadband's transformative effect on the internet, enterprises are more likely to invest if AI can deliver reliable, scalable solutions.

Key Advancements Driving AI Adoption:

• Token Processing Speeds: Increased speeds enable handling complex tasks in real-time.
• Context Window Size: Expanding context window size means AI can process and analyse more information in each interaction, boosting accuracy and performance.

Part 4: Workflow Capture – Economics of Enterprise AI

The true advantage of AI lies not in individual task automation but in the ability to capture and manage entire workflows. This "workflow capture" allows enterprises to adopt AI without significant disruptions to existing systems.

Steps to Capture Workflow:

1. Identify feasible workflows where AI can add value.
2. Evaluate integration opportunities in adjacent workflows to maximise AI's impact.
3. Adopt performance-based models, focusing on delivering outcomes, not software.

Part 5: Business Model Advantage – Service-as-a-Software

AI offers a unique advantage by allowing businesses to shift from subscription-based models to performance-based models. Just as IoT led to models where companies charged for outcomes (like jet engine operating time), AI can enable businesses to charge for workflow outcomes, such as leads generated or support tickets resolved.

Benefits of a Performance-Based Model:

• Predictable Outcomes: Allows clients to pay based on results, not time spent.
• Cross-Subsidisation: Providers can bundle AI-driven services, enhancing customer value.
• Bundling Opportunities: Long-term winners will combine services to drive adoption and retain clients.

Part 6: Challenges – Hype, Organisational Disconnect, and Lateral Attacks

1. Hype Management: Avoid overpromising what AI can do; focus on realistic outcomes.
2. Organisational Alignment: AI requires collaboration between model engineers and UX designers to create user-centric solutions.
3. Competition and Innovation: Due to AI’s rapid evolution, lateral attacks from competitors are common. Enterprises should focus on owning core workflow components to maintain control.

Part 7: Competitive Advantage – Creating Moats and Expanding Accounts

Establishing control points within workflows is essential to build a competitive moat. Enterprises need to focus on capturing critical decisions and actions within a workflow, ensuring AI-driven components align with enterprise goals.

Expansion Paths:

• Hub-First: Capture workflow control points first, and allow AI to replace additional tasks over time.
• Agent-First: Use AI agents to manage workflows initially, then expand coverage as trust grows.

Part 8: Defining Winners and Losers in Enterprise AI

Winning with enterprise AI requires the ability to combine vertical expertise with horizontal expansion. Successful players will:

• Establish control points by owning core tasks within workflows.
• Expand horizontally, providing bundled solutions that go beyond point services.

Conclusion

Enterprise AI has moved beyond the hype. Companies that strategically unbundle and re-bundle workflows, adopt performance-driven models and anticipate competitive dynamics are best positioned to leverage AI for lasting advantage. By following this guide, enterprises can navigate the complexities of AI adoption with confidence, transforming AI from a mere tool into a powerful enabler of growth and efficiency.

FAQs

1. What is "Service-as-a-Software" in the context of AI?

Service-as-a-Software is an AI-driven approach where complex workflows traditionally managed by humans are absorbed into software, automating knowledge and managerial tasks.

2. How is Service-as-a-Software different from Software-as-a-Service (SaaS)?

While SaaS offers software tools to perform tasks, Service-as-a-Software goes further by fully automating entire workflows, including decision-making and management, making it a more transformative model for enterprises.

3. What types of workflows are best suited for AI-driven automation?

Knowledge-intensive and repetitive managerial workflows, such as customer support, procurement, and corporate travel booking, are ideal for AI-driven workflow automation.

4. How can AI unbundle and re-bundle tasks within an enterprise?

AI can take on repetitive tasks or decisions in workflows, unbundling them from human roles and bundling them into efficient, software-driven processes that increase productivity and reduce costs.

5. What are the key benefits of adopting Service-as-a-Software?

Enterprises benefit from reduced labour costs, faster decision-making, enhanced workflow efficiency, and new opportunities for performance-based pricing models.

6. How does Service-as-a-Software provide a competitive advantage?

It allows enterprises to automate workflows, freeing up human resources for strategic tasks, and enabling businesses to offer performance-based pricing, a distinct edge in competitive markets.

7. What are the main challenges in implementing Service-as-a-Software?

Key challenges include ensuring high AI accuracy, aligning AI with user experience, managing organizational change, and defending against competitors with similar capabilities.

8. How can businesses protect their AI workflows from lateral attacks?

Businesses should establish control over critical workflow components and integrate deeply with client operations to build a moat, making it harder for competitors to take over.

9. What role do CXOs play in successful AI adoption in enterprises?

CXOs should champion AI adoption, align AI strategies with business goals, and foster cross-functional teams to ensure the AI solution meets both technical and user needs.

10. How will AI-driven workflows impact traditional business models?

AI-driven workflows can shift enterprises from subscription-based to performance-based pricing, creating new revenue models and fostering long-term client relationships through efficiency and measurable outcomes.