AI enters everyday business in three ways: How companies in Espoo can benefit from the change

AI is no longer something used only by large technology companies or research labs. Over the past year, it has become an everyday tool for many small and medium-sized enterprises, especially in marketing, software development, customer service and analytics.

Espoo is home to many technology companies, growth companies and research organisations that develop new solutions. More and more businesses are asking the same question: how can AI be used in a sensible and controlled way in their own operations? In our collaboration with companies, three key development trends stand out. Recognising these trends can significantly improve the use of AI in business.

Companies collect large amounts of data from online services, production equipment, customer systems and analytics tools. AI promises to use this data efficiently, but in practice the first challenge is often the quality and fragmentation of the data. In many organisations, data is scattered across different systems, classified in different ways or its origin is unclear.

A company may, for example, monitor the quality of customer service in three channels: email, chat and phone. Each system categorises contacts differently, so when this data is combined for AI-based analysis, the results become distorted because the systems use the same terms to mean different things.

Often the most effective approach is to fix the data fundamentals:

• map where your data comes from and how it is collected
• harmonise key terms and classifications
• identify the most important data sources
• start AI solutions with a small but high-quality dataset

When the data flow is clear, using AI solutions becomes much easier.

Marketing, software development and expert work are areas where generative AI tools are already used daily. This often happens without shared guidelines, because AI has entered workplaces through employees.

For example, a marketing team may use AI to draft texts and generate ideas. At the same time, they may feed the tool with information about customer groups, campaign plans and internal documents, even though they do not know how the service processes or stores this data.

Companies benefit from guiding the use of AI through clear rules:

• create simple instructions for using AI
• define three tiers: approved tools, caution zone and prohibited tools
• provide an AI tool that the organisation has approved
• organise a short AI literacy training session

Even a one-hour training session can significantly reduce misuse and security risks.

Small, specialised AI solutions may well become a Finnish strength. Public discussion often focuses on large language models, even though in practice many needs can be solved with a small AI model built for one specific task. This fits well with Espoo’s business landscape, which has strong expertise in software development, electronics, industry and research cooperation.

A company may want to predict maintenance needs for a large fleet of equipment. A general-purpose language model is heavy and expensive for this task, whereas a small model trained specifically for the purpose can make accurate predictions from a limited dataset.

A good AI project often starts with one clear use case:

• choose one problem where AI can help with prediction or classification
• collect a high-quality dataset for this specific task
• test a lightweight model that can run in the company’s own environment
• document the results and insights

The first successful pilot often becomes the foundation for broader use of AI.

Introducing AI does not need to be a large project. For many companies, a good start is three simple steps:

Understand your own data.
Find out where your data comes from and how it is used in decision-making.

Create rules for using AI.
Guide employees to use AI safely and effectively.

Start with a small pilot.
Choose one targeted use case and test an AI solution in practice.

We offer support for using AI

Companies in Espoo have exceptionally good opportunities to benefit from AI. The city has a strong research and innovation ecosystem, and companies can access practical support for testing new solutions.

One option is FAIR EDIH (Finnish AI Region European Digital Innovation Hub). It offers support for SMEs and public organisations in using AI, data analytics, high-performance computing, cybersecurity and XR technologies.

For companies, this may include:

• mapping AI opportunities
• background research and evaluation of solution alternatives
• proof-of-concept trials with research organisations
• opportunities to test solutions in real environments

For many companies, the most important thing is that they do not need to solve everything alone. Once the first pilot begins, the next steps often become clearer.