Overview
Finland, Poland and China, SDK Freja is one of the largest transport and logistics companies in the Nordic countries. Their services include freight forwarding within road, air and sea, project cargo, contract logistics, as well as stevedoring, port agency, customs clearing, chartering, and liner and cruise services. In addition, the company offers specialist logistics solutions within healthcare, warehousing, and refrigeration.
With a robust set of services, a growing customer base, and multi-lingual operations across borders, teams struggled with timely and clear communications, complex documentation, and a massive sales pipeline. That’s when they discovered StackAI.
In early 2024, SDK Freja’s Head of AI, BI, and Automation, Alexander Kristensen saw the potential for secure, scalable AI programs deployed across teams with varied technical backgrounds. After a disappointing attempt with Microsoft’s Azure AI Foundry—which, in Kristensen’s words, is “all marketing,” unable to deliver what it promises and requiring massive investment in external consultants—he discovered StackAI and made a free account. He soon saw how easy it was to create a working chatbot with different interfaces, and that the time from idea to an enterprise-ready published project was dramatically reduced.
The rest is history.
From simple chatbots that are still used thousands of times per week, to increasingly specialized AI applications that utilize the advanced forms interface, SDK Freja has continued to build out a suite of AI agents that save time, reduce manual work, and increase operational clarity across borders, leading to:
Instant IT support from an AI-powered chatbot
Dozens of tender contracts processed weekly in minutes—automating a process that used to take days
80,000+ hours of audio translated, improving cross-border collaboration
Millions of dollars in new pipeline from sales research AI agents
General-Use, Widely-Adopted Chatbots
SDK Freja began their AI journey with a set of general-use AI agents, the first of which was simply an internal version of ChatGPT. Kristensen notes that the cost of buying subscriptions from providers themselves was quite high, but because all leading models are supported on StackAI, it made sense to create a version for employees to use in-house. They also created an IT service desk chatbot.
The Problem: Overloaded IT Support Teams
With thousands of employees across multiple countries, IT support teams were overwhelmed by repetitive requests—password resets, wifi issues, and system access. Response times were long, frustrating employees and delaying work.
The Solution: Automated IT Support Agent
StackAI powered an AI help desk agent trained on internal IT guidelines. Employees now get instant, accurate answers via chat, reducing the need for manual intervention. Average wait time for support dropped to just 11 seconds.
Faster Contract Reviews for Logistics Agreements
Kristensen and his team then moved to tackling one of the thorniest parts of SDK Freja’s transport business: the tender process.
The Problem: Complex, Lengthy Tender Agreements
When competing for new customers, often large production companies with large logistic and transport costs, SDK Freja provides detailed information in response to a submitted tender. Tenders can span many variables: the kinds of goods that must be picked up, when and where they should be picked up, special terms, and more. As tender managers continued receiving a high volume of tender agreements, and each agreement spanned 20 to 50 pages, reviewing each contract manually was slow, error-prone, and limited the ability to respond quickly to win new business.
The Solution: Automated Contract Parsing
Freja quickly deployed a solution tailor-made for the need. Affectionately named “Tendermate,” this AI agent utilizes around 10 sub-agents (LLMs) and extracts over 10 key terms per document—covering liability, payment terms, and delivery windows—so managers can receive insights and respond to tenders more efficiently. Managers simply upload a tender package, press submit, and receive around 10 separate outputs containing extracted information that they can use to understand the tender at a high level. An advanced form interface makes this process seamless and legible to users of various technicality levels. The best part? Tender review took days to complete before StackAI—but now just takes minutes.
Removing Barriers Across Languages
The Problem: Inconsistent Communication Between Teams
Operating across multiple countries created daily challenges in translating emails, memos, and internal documents. Misunderstandings introduced errors, and typos and grammatical errors slowed collaboration.
The Solution: Enterprise-Grade Translation and Transcription Tools
SDK Freja soon rolled out a set of translation AI agents for text and audio, delivering nuanced, context-aware translations. These agents remain widely used today, including one that can transcribe audio, translate it into a desired language, and adjust the formality level of the translation—from bullet points, to a project description, to an email. Kristensen observes that this tool has been adopted by many, but is particularly helpful for handicapped employees who speak rather than write, and for those who are unconfident in their writing, and is a large improvement for the entire company: more than 80,000 hours of audio have been translated to date, transforming global communication.
Accelerating Sales Pipeline Across Borders
The Problem: Inconsistent, Time-Intensive Reporting
An important part of SDK Freja’s sales process is the reporting step, wherein salespeople must write comprehensive reports after visiting different customers in Denmark, Norway, Sweden, Finland, Poland, and more. After observing the difficulty of this process, and high variability in the quality of written reports between teams, Kristensen was determined to build a solution.
The Solution: A CRM-Integrated Reporting AI Agent
With StackAI, sales teams simply upload an audio file of their meetings with the customer directly into SDK Freja’s CRM, where an AI agent automatically transcribes the conversation and extracts key data to fill in various fields of the visit report, dramatically reducing the time spent on this step of the sales process.
The Future: Deeper Sales Research
Now, Kristensen says he’d like to build out even more sales resources to help SDK Freja determine which business opportunities to focus on. In development is a sales intelligence AI agent, which takes a large list of prospects from various industries and geographies, researches to find specifics about their production locations, runs searches on LinkedIn, and more. Having the power to bring a list of 1,000 companies down to 200 tailored opportunities, for example, is a huge benefit to the sales team, Kristensen says. The initial version of this agent has already helped to win customers in SDK Freja’s Poland business, which Kristensen calls a definite success story.
Conclusion
As SDK Freja continues to build out use cases for AI agents driving ROI, Kristensen emphasizes that “the time from idea, to an actual proof of concept, to production is quite fast” on StackAI. “AI is an enabler for automating processes and developing new products, and StackAI makes it real in our business,” he says.
StackAI is proud to be a trusted partner of SDK Freja, empowering communication, collaboration, and business development across borders with powerful internal AI solutions. Want to see how StackAI can do the same for your enterprise? Get a demo here.
Make your enterprise smarter with AI.
Deploy custom AI assistants, chatbots, and workflow automations with white-glove support from AI experts.
Customers
Explore More Customer Stories
The Future of Participant Service: How the YMCA Retirement Fund Delivers 24/7 Member Support with AI Agents
The YMCA Retirement Fund wanted to scale participant support and internal knowledge access without increasing staff or operational complexity.
From Weeks of Research to Minutes: How NobleReach Became the AI-First Nonprofit Leading Tech Transfer Innovation
Manual research, competitor analysis, and tech transfer reports took a full week per project, slowing impact and overwhelming a small nonprofit team.
How Varos Saved 800+ Hours With an AI-Powered Categorization Agent
The operations team spent countless hours manually scanning company profiles, analyzing product offerings, and categorizing leads
