Keynote - Unlocking Real-World ROI with Gen AI Chatbots
Juha Sulkakoski — Founder, Upseller
Generative AI chatbots aren’t just theoretical—they’re delivering measurable returns in sales and customer service today. In this keynote, Juha Sulkakoski, Founder of Upseller, dives into how leading Nordic brands are successfully integrating Gen AI chatbots into their operations, enhancing customer experiences, accelerating sales cycles, and driving tangible business outcomes. Through real-world cases, practical examples, and actionable insights, Juha will outline how companies can leverage conversational AI to generate clear, quantifiable ROI.
"One is that they actually trust it more and it seems that they start using it more because they get more answers, it goes more deep. And especially if you have a good data, if you integrate it to customer data, it is bringing you all the way to the end."
Summary
- Transition from traditional chat solutions to Gen AI chatbots. - Gen AI enables human-like dialogue, improving customer service. - Automated systems reduce labor costs and increase efficiency. - Real-world examples show significant sales and service improvements. - Continuous development and integration with customer data are crucial.
Article
The AI revolution in customer service: How generative AI is transforming business interactions
Finnish tech expert demonstrates the real commercial impact of AI chatbots at Helsinki tech summit
The quiet revolution in how businesses interact with their customers took centre stage at the VERGE AI Frontier event in Helsinki yesterday, where industry veteran Juha Sulkakoski presented compelling evidence that generative AI is fundamentally changing customer service economics.
Speaking to a packed audience at his keynote session "Unlocking Real-World ROI with Gen AI Chatbots," Sulkakoski drew on a decade of experience implementing chat solutions to outline how the latest AI technologies are delivering measurable returns for businesses.
"What chatgpt can do or what LLMs can do is create a human-like dialogue with the customer. And this is a huge change," Sulkakoski told attendees at the technology summit focused on creativity, security and collaboration in artificial intelligence.
From basic chat to intelligent conversation
Sulkakoski's presentation traced his company's evolution from implementing basic human-operated chat systems on company websites to developing sophisticated natural language processing (NLP) chatbots designed to handle high-volume customer service inquiries. The most dramatic shift, however, has occurred in just the past six months as they've integrated generative AI technologies into their solutions.
The impact of this transition extends beyond just technological advancement. As Sulkakoski noted, the fundamental way people seek information has transformed: "Getting information is changing. We all were using Google search before one year ago and I would say that mostly like myself, I'm using chatgpt for search."
This shift from query-based search to conversational information-seeking represents a profound change in customer expectations. Yet according to Sulkakoski, most business websites remain stuck in the past, with "ninety-five percent of the solutions are old school" chatbots that rely on predetermined buttons and menus rather than natural conversation.
Real-world business impact
The presentation's most compelling elements were the case studies demonstrating tangible business outcomes. Sulkakoski highlighted two clients – Roki, an international building products company, and IMO Park, a Scandinavian parking services provider – that have implemented generative AI chatbots.
For IMO Park, the solution handles emotionally charged customer service scenarios, such as parking ticket disputes, with a level of consistency and patience that human operators struggle to maintain. "It doesn't get angry; it doesn't hang up on you. Even at night, even if somebody is drunk and nasty, it handles that really nicely," Sulkakoski explained.
The numbers tell a persuasive story of return on investment. Within three months of deployment, chatbot usage among website visitors rose from 5% to 21% – meaning one in five visitors engaged with the AI assistant. With an internal cost of €12 per human-handled call, Sulkakoski calculated savings of approximately €1.8 million. Customer satisfaction metrics showed 91% of users were pleased with the AI-generated responses.
The future of customer engagement
What makes these systems particularly valuable is their ability to handle both high-volume routine inquiries and more specialised questions that would typically require extensive training for human staff. "You can handle the big traffic, you can handle the common questions, but you cannot basically teach the team of chat people or the customer service all that's necessary. But LLMs are capable of that," Sulkakoski said.
Perhaps most promising for businesses considering similar implementations is that these systems continuously improve without additional investment. "The good thing is that these models are getting better without us doing anything. The rule of more is still holding on. The computers are getting better and faster all the time," Sulkakoski noted.
The presentation highlighted how Finnish language capabilities have advanced to the point where specialised models are unnecessary for most applications. "The common chatgpt model like four point zero is very good in Finnish. You don't need a Finnish model for that," though he joked that regional dialects might still benefit from specialised training.
As the VERGE conference continues today in Helsinki, Sulkakoski's presentation stands as a pragmatic counterpoint to more theoretical discussions of AI's potential. By focusing on measurable returns and real-world applications, he demonstrated that the AI revolution in customer service is already delivering value – not just promising it for some undefined future.
Part of VERGE | The AI Frontier: Creativity, Security & Collaboration