The Path to a $100M AI Agency: Scaling Agentic Workflows and Moving Up-Market
YouTube
This video features Nate Herk, founder of Appit AI, in a deep-dive podcast with Devin Kearns, co-founder of CustomAI Studio. Devin shares his extensive experience building custom AI systems for clients ranging from solo founders to large enterprises. He discusses the rapid evolution of the AI landscape, particularly the shift from simple prompt-based chat tools to complex agentic systems that integrate deeply into business operations. Devin outlines his strategic approach to building a scaleable AI agency, focusing on moving up-market to serve mid-market companies where the return on investment for AI integration is significantly higher and more defensible.
The conversation covers critical aspects of the current AI-first business model, including the collapsing value of pure development and the rising importance of strategic consulting. Devin explains how CustomAI Studio uses a phased engagement funnel, starting with workshops and blueprints before moving to full-scale development and ongoing technology partnerships. By focusing on measurable business outcomes like reducing refund rates or automating complex underwriting processes, the agency aims to achieve an enterprise valuation that goes beyond mere services income. The podcast provides a roadmap for aspiring AI operators to move beyond basic automations and build robust firms with long-term enterprise value.
This video covers the strategic roadmap for building and scaling a high-value AI consulting agency in the era of agentic workflows. Devin Kearns explains the transition from simple AI automations to integrated production systems that provide measurable ROI, focusing on serving the mid-market to maximize enterprise value and prepare for a significant business exit.
Key Takeaways
Pure development value is trending toward zero as AI makes coding more accessible. The real value now lies in strategic business logic and system architecture.
The mid-market (companies with $10M to $250M revenue) is currently the prime opportunity for AI transformation because they have repeatable systems but significant labor costs.
High-value AI agencies should move away from time-and-materials billing toward value-based or performance-based models tied to specific business outcomes.
A successful AI engagement funnel starts with low-friction workshops to build trust, followed by detailed blueprints, before moving into development and ongoing partnerships.
Agentic AI is not just about chat. It is about building event-driven, deterministic systems that can replace or augment entire labor-intensive business processes.
Diagram
Loading diagram...
Timestamps
00:52
IntroductionMeet Nate Herk and Devin Kearns from CustomAI Studio.
05:45
The Inflection PointWhy mid-market companies are now ready for AI transformation.
09:06
Labor Replacement vs. AugmentationHow AI is reshaping organizational charts and labor value.
17:36
Business Economics of AIMoving from time-and-materials to value-based ROI projects.
22:24
The Different PlaybooksComparing solo freelancing, product development, and full-stack consulting.
26:47
The Future of AI WorkWhy thin wrappers and simple automations won't survive long-term.
44:28
Valuation and MultiplesThe math behind reaching a $100M enterprise valuation.
49:50
The Engagement FunnelStructuring workshops and blueprints to close large contracts.
60:48
Target Audience
AI agency founders, software developers looking to transition into AI services, business consultants, and tech entrepreneurs interested in the AI services market.
Use Cases
-Designing an engagement funnel for an AI consulting business
-Identifying high-value AI automation opportunities in the mid-market
-Understanding how to value an AI services firm for eventual exit
-Moving from simple chatbot implementations to complex agentic systems
-Developing a framework-based approach to AI business transformation
The AI market has moved past the initial hype phase where simple chatbots and thin wrappers around large language models were enough to command a premium. Today, even non-technical business owners are beginning to ship AI-driven solutions to production. As a result, the service provider's role must shift from being a basic builder to being an architect of business transformation. Devin highlights that the early majority of companies are now entering the market, creating a massive opening for operators who have been in the space since the beginning to build firms that truly change how a company functions. The traditional linear labor model of services firms is being replaced by compounding leverage, where a small team of experts can manage a flurry of AI agents that produce outsized results.
Why the Mid-Market is the Sweet Spot
While solo founders and small businesses are easy to acquire, they often lack the scale to provide significant ROI for custom AI systems. Conversely, massive enterprises often move too slowly and have fragmented internal politics. Mid-market companies typically have enough annual revenue to make a five or ten percent efficiency gain worth millions of dollars. They already have established standard operating procedures (SOPs) and repeatable workflows that can be codified into agentic systems. By focusing on these clients, an agency can build high-contract-size relationships that are defensible and more likely to lead to long-term technology partnerships.
The Productized Service Funnel
To manage a client from a position of authority, Devin suggests a four-stage engagement funnel. It begins with a two-session AI Workshop designed to align the client with the current state of the technology and identify pain points. The second stage is the Custom AI Blueprint, where the agency performs a deep diagnostic of business logic and workflows to create a business case for the board. The third stage is the actual build of the custom AI project. Finally, the engagement matures into an AI Technology Partnership, where the agency is embedded in the client's operations as a fractional CTO or AI-native engineering team. This model ensures that each stage earns the next, building trust and depth with every step.
Economics and Valuation of AI Agencies
One of the most compelling parts of the discussion is the math behind a $100M exit. Devin explains that services firms re-rate their valuation multiples as they scale. A small firm making less than $3M in EBITDA might sell for a 1x or 2x multiple. However, once a firm crosses the $5M to $10M EBITDA mark with recurring revenue, proprietary IP, and deep mid-market contracts, those multiples can jump to 8x, 15x, or higher. To reach these heights, firms must move beyond one-off projects and focus on building an "AI Operating System" that becomes a fundamental part of the client's infrastructure. By capturing the "AI-native premium," agencies can build businesses that are far more valuable than traditional labor-based consulting shops.
Practical Applications
AI agency owners can apply these insights by auditing their current service offerings. Instead of offering general automation, they should identify a specific vertical or high-cost labor process and develop a proprietary framework for automating it. For example, focusing on a specific outcome like reducing e-commerce refund rates by five percent can be worth millions to the right client. Additionally, operators should start documenting their internal processes to transition from being the primary builder to being the manager of a team that utilizes standardized AI-native workflows. Building a community or a "funnel before you need it" through content creation is also vital for establishing the authority needed to close mid-market deals.
Frequently Asked Questions
What is an agentic AI system compared to a chatbot?
An agentic system is an event-driven, deterministic workflow where the AI acts as an executor of tasks rather than just a conversational partner. Unlike a simple chatbot that answers questions, an agentic system follows a multi-step logic to achieve a business outcome, such as processing a legal claim or conducting an investment underwriting analysis.
Why is the value of pure development trending toward zero?
As large language models become better at coding and non-technical users gain access to sophisticated build tools, the act of writing code is becoming commoditized. The real value is shifting upstream to the strategic understanding of business problems and the ability to architect complex systems that solve those problems reliably.
How do you choose between a lifestyle business and a high-valuation exit strategy?
A lifestyle business is focused on maximizing current income and founder freedom, often leading to founder-dependent services that are hard to sell. A high-valuation exit strategy requires building repeatable frameworks, hiring for a scalable organizational structure, and targeting higher-revenue clients to build enterprise value that can eventually be acquired by private equity or a larger firm.
What are the main traps that prevent AI agencies from scaling past $100k per month?
Common traps include time-and-materials billing (which penalizes efficiency), the founder being involved in every delivery step, and taking on too many one-off, undifferentiated projects. To break through, a firm must switch to value-based pricing, implement a productized methodology, and focus on high-concentration mid-market clients.
Starting from ZeroDevin's advice for those just entering the AI services space.
