The SaaS Revolution Show

Episode: Nikola Mrkšić, PolyAI: Enterprise AI Sales, Pricing, & Millions in ROI
Date: October 2, 2025
Host: Alex Theuma
Guest: Nikola Mrkšić, Co-founder & CEO, PolyAI

Episode Overview

In this episode, Alex Theuma sits down with Nikola Mrkšić, CEO and co-founder of PolyAI, ahead of his appearance at SaaStock Europe. They discuss PolyAI’s journey from a research-focused startup to a leading enterprise AI company, the evolution of AI technology, sales and pricing strategies in the enterprise space, and the real-world impact of voice AI agents. Nikola shares candid insights about founder sales, the realities of pricing AI solutions, and how AI is changing the nature of customer service across industries.

Key Discussion Points & Insights

1. Nikola Mrkšić’s Background and Founding Story (02:24–06:41)

• Growing up in Serbia: Nikola describes his classical Eastern European upbringing, shaped by the region’s political instability. Despite a passion for science fiction writing, he was steered toward STEM and competitive mathematics.
• Early entrepreneurial spirit: Shared anecdotes about selling Star Trek-inspired stories in local cafes.
  • Quote: “I had this idea of just going around like cafes and trying to like sell them to people… I came up with this like packaging where basically you could like buy three or five at different prices… I made like 2 or €300, and the average Serbian salary that year… was 2, €300 a month. So I felt like there’s a way to hack it.” (03:03)
• Journey to Cambridge: Pursued computer science and math, leading to involvement with VocalIQ, a deep learning startup spun out of Cambridge which was acquired by Apple—setting Nikola on a path toward building enterprise AI solutions.

2. Narrowing the AI Vision to Enterprise Customer Service (06:41–08:38)

• Recognized the gap between sci-fi expectations and real-world voice assistants like Siri. Refocused on more manageable, measurable B2B problems—specifically, customer service.
• Customer service as a ripe area for AI: “About 2% of the working population in the UK and the US works in contact centers… despite all the automation… it hasn’t really shifted all that much.” (05:38)
• AI has tremendous potential in this space but industry-wide transformation is still in progress.

3. The Impact of AI on Contact Centers and The Human Factor (06:41–09:11)

• Nikola predicts a significant reduction in the required human labor in contact centers—potentially a 5x reduction within a few years.
• AI brings not just efficiency, but a new type of “air traffic controller” knowledge workforce to oversee and guide these powerful tools.
  • Quote: “If you have AI doing 80% of the work… it will start to resemble air traffic control, where it’s going to pay to have levels and levels of supervision and just, continuous improvement.” (07:56)

4. PolyAI’s Growth and the Mainstreaming of Enterprise AI (09:11–10:49)

• PolyAI spent initial years as a “fake company”—more like a research lab than a business.
• The explosion of generative AI (ChatGPT, LLMs) accelerated both capabilities and enterprise interest: demand shifted from early adopters to board-level mandates.
  • Quote: “The interest of enterprises and their belief that this can be done with technology has changed profoundly…” (10:04)

5. Challenges of Rapid AI Development (10:49–13:11)

• The pace of model improvement forces constant product adaptation—sometimes requiring rewrites or complex transition plans for enterprise clients.
• Smaller companies remain more agile; as scale increases, supporting all clients becomes more complex.

6. Founder Sales and the Enterprise Go-to-Market Path (13:11–15:51)

• Nikola addresses the European stereotype of the technically weak founder; in reality, B2B SaaS growth requires technical founders to learn and embrace sales themselves.
• Caution against the myth that hiring a “killer head of sales” replaces the need for the founder in early-stage sales—founder-market fit is key.
  • Quote: “Founder sales all the way. You have to lean in… as you’re selling, you’re tweaking the product… the best way is to have a few… tightly integrated owning the whole thing.” (14:13)
  • Quote: “The killer head of sales who can really sell any software easily… there aren’t that many of them. They’re working at very large companies making seven figure… paychecks every year.” (15:23)

7. Enterprise AI Pricing: Outcome-Based vs. Consumption (15:51–18:54)

• PolyAI uses a mix of consumption and outcome-based pricing, tailored to customer preference.
• Outcome-based pricing is attractive but complex to execute; not a silver bullet, particularly at early stages.
  • Quote: “That’s all very nice and theoretical... If I knew everything about your business, I could price it fairly… There are companies that think that way… but a lot of others look at it as analogous to the picks and shovels…” (16:14)
• Early-stage vendors must recognize that customers invest significant time and resources simply by engaging, not just by spending money.

8. Delivering Real ROI With Voice AI Agents (18:54–23:00)

• PolyAI’s strongest product-market fit began in hospitality (restaurants, casinos in Vegas) by directly boosting customer bookings and revenue.
  • Quote: “For casinos, we increase annual revenue by 5 to 10 million; for large restaurant groups, tens of millions of revenue uplifts.” (19:21)
• In enterprise, factors like compliance, risk management, and revenue generation often outweigh cost-cutting in buying decisions.
• AI improves customer experience—as shown by significant improvements to NPS for banking clients.
• AI is not eliminating jobs as many fear; rather, companies struggle to hire enough agents, and AI fills the gap, enabling growth.

9. Nikola’s Daily AI Tool Stack and Productivity (23:00–26:15)

• ChatGPT is now Nikola’s “operating system” for most daily tasks: drafting memos, ideation, even pricing documentation.
  • Quote: “ChatGPT has become like my operating system for absolutely everything… I don’t even use Google anymore.” (23:21)
• Other recommended tools: Lovable (for code prototyping and discussions), Cloud Code, Granola (preferred over Gong for note-taking), Cursor (internal coding).
• Admits that while ChatGPT accelerates business writing, he avoids using it for personal LinkedIn posts due to content “degradation”:
  • Quote: “There’s a definite kind of degradation in the level of content you see on LinkedIn and stuff… It’s obvious when the tone of the person is not coming through.” (25:04)
• Discussion about AI-written content and its effect on authenticity and algorithm performance.

10. What’s Next for PolyAI? (28:31–29:48)

• The North Star metric: Number of clients where PolyAI replaces the work of 1,000+ people.
• Plans to double that number repeatedly over coming 12 months, focusing growth on large enterprises in the US, UK, and continental Europe.
  • Quote: “The only real measure is the impact…how sticky they are and really just how good our product is.” (29:22)

Notable Quotes and Memorable Moments

• “To build a company you need to delight customers, get more usage, get paid for it, get data, improve the product, delight them even more. So the cycle there…was very obvious—customer service.” — Nikola Mrkšić (05:24)
• "Founder sales all the way…as you’re selling, you’re tweaking the product…You have to have a very quick iteration loop." — Nikola Mrkšić (14:13)
• "A dollar earned is a lot better than a dollar saved." — Nikola Mrkšić (19:21)
• "If you have AI doing 80% of the work… it will start to resemble air traffic control." — Nikola Mrkšić (07:56)

Key Timestamps

• 02:24 — Nikola’s background and entrepreneurial start
• 06:41 — Why PolyAI focused on enterprise customer service
• 08:38 — AI’s impact on knowledge work and contact centers
• 09:11 — PolyAI’s “fake company” research phase and inflection point
• 13:11 — Moving from technical founder to enterprise salesperson
• 15:51 — PolyAI’s real-world approach to pricing enterprise AI
• 19:21 — How PolyAI delivers measurable ROI for clients
• 23:21 — AI tools and ChatGPT in Nikola’s daily workflow

Takeaways for SaaS Founders and Operators

• AI is reshaping enterprise customer service, but transformation is complex and nuanced—not just about replacing people.
• Technical founders must embrace sales responsibilities in early-stage SaaS.
• Pricing innovation (outcome-based models) is attractive but operationally challenging.
• Real ROI from AI comes from revenue generation and customer experience uplift, not just cost savings.
• AI tools are increasingly central to daily workflows—even for top executives—but authenticity remains crucial for thought leadership and brand voice.

For more SaaS founder insights and upcoming events, check out SaaStock.com