AI cannot fix poor data or poor management
A survey conducted by Tickingbot in autumn 2025 among 24 Finnish organisations found that many still struggle with the basics in the age of digitalisation and AI. Technology is not the main problem. People and structures are.
Strategy on paper, chaos in practice
Most organisations recognise data as a strategic opportunity, yet implementation is fragmented and dependent on individuals. Excel files live separate lives on specialists’ computers, systems do not communicate, and no one genuinely owns the data. Without shared processes and accountable owners, data collection becomes a side task, leading to poor quality and unreliable analysis.
Five bottlenecks
- Lack of intent: organisations have not clearly defined what information they need and why.
- Unclear responsibilities: ambiguous data ownership causes duplication and inconsistent reporting.
- Quality problems: definitions are missing and manual work consumes time.
- Technical debt: fragmented systems, dependence on Excel and poor interoperability.
- Cultural resistance: recording data is seen as extra work, information is treated as a source of power, and decisions are based on instinct instead of evidence.
AI cannot save you if the foundation is missing
AI is seen as a way to speed up routine work and information retrieval, but security, legal constraints and ethical risks cause concern. Some organisations report tangible efficiency gains from chatbots, translation tools and generative solutions. Elsewhere, experiments remain uncoordinated activities by individual employees.
Change starts with leadership
Change management is a key success factor. Successful organisations assign responsibilities, invest in training and embed new practices in daily work. Management commitment means actively using data in decision-making, allocating resources to development and making achievements visible.
What should you do?
- Start with intent: define what information you need, why you need it and where it comes from.
- Appoint owners: make data management someone’s job, not everyone’s side task.
- Pilot boldly: choose low-risk use cases and measure the results.
- Train in practice: invest in hands-on support models.
- Celebrate progress: make the benefits visible and shareable.
Developing data-driven operations is not primarily a technology issue. It is a challenge of leadership, culture and capability development. Organisations that understand this are already moving forward.
Order the free report and book a presentation
Contact us if your organisation wants to develop data-driven management, assess its data maturity, identify concrete next steps or receive sparring for data and AI development.
