Case ADE: Data-driven product development leads to satisfied customers and improved competitiveness

UnSeen helps in developing the Agile Data Engine DataOps platform, which automates the creating and utilization of data warehouses (ADE). By collecting and refining data, features are developed to enhance ADE’s competitiveness, enabling end-users to achieve more, faster, and at lower costs.

ADE is utilized by several large Finnish and international companies, such as DNA, Fortum, Ponsse, S-ryhmä, ISS and Skandia. Since December 2023, ADE Insights Oy has operated as its own subsidiary as part of Solita group.

Challenge: Increase Momentum in Global Market Conquest

You are either in or you fall behind

This is how ADE Chief Product Officer Harri Kallio describes the role of AI on IT industry at 2023. This was the reason why ADE development team wanted to find out what would be the possibilities with leveraging data and AI. Important target was also to give boost ADE to gain momentum at international markets.

Continuous product development is crucial for this business. We knew that AI could be used to implement new features that would benefit our customers. The challenge was to find the right way to develop and utilize them as part of ADE,’ says Kallio

The Solution: Data science highlights innovation opportunities

To explore suitable development directions, ADE established an artificial intelligence team internally, with UnSeen’s data science expert Jaakko Routamaa as its first member.

We already had a good relationship with UnSeen, and to manage risk levels, we initially wanted to acquire expertise on a consulting basis. The process has been smooth, and very little guidance has been needed, says Kallio.

We already had a good relationship with UnSeen, and to manage risk levels, we initially wanted to acquire expertise on a consulting basis. The process has been smooth, and very little guidance has been needed, says Kallio.

Routamaa began exploring ADE’s data and started, in Kallio’s words, an expedition.

Initially, an exploratory data analysis was conducted, reflecting on development ideas concerning the utilization of ADE’s collected data. Key metrics and visualizations were created to facilitate decision-making, and eventually, some ideas were rejected while potential ones were advanced. After iteration, the most promising idea for a new feature was selected. In its implementation, a machine learning model trained on the company’s own data was chosen, and the feature development was then taken into production.

Although the project started off like an exploration, it was crucial to stay grounded in practicality from the beginning

We didn’t want to do just experimentations for couple of years. Instead, we wanted to immediately incorporate exploitable features into ADE. When vision is merged to concrete actions brilliant things can happen. As we have seen in this case, says Kallio.

Results: The route is open for more satisfied customers and new opportunities

Six months after the start of the project, ADE’s first new feature was ready for release. At the same time, new features utilizing machine learning models are on development already.

Now we have a process that allows us to bring more features desired by ADE’s customers. This way, they get more, faster, and at lower costs. For us, it means an increase in competitiveness as we keep our current customers more satisfied and attract new ones, says Kallio.

Kallio highly values how, with UnSeen’s help, the utilization of data in developing ADE has taken leaps forward.

Starting almost from scratch, we have reached a point where we are developing features using AI every day. So, things have been set in motion, and you can see how development progresses every week. We are still in the early stages, but this initial spark has been very important in order to get started with concrete steps and not just stay dreaming.

ADE in numbers