Nordisk Movie Studio Drives Data with a Small Team and a Big Vision
Modernizing their data warehouse presented a host of problems for a Denmark media company.
- By Brian J. Dooley
- March 28, 2019
As data warehousing and analytics continue to move to new architectures, many firms are struggling to replace or extend legacy warehousing systems. This demands a re-evaluation of plans in place and future considerations. It can be particularly difficult for smaller IT departments and small businesses.
Such was the case with Nordisk Film, a Copenhagen-based film studio that has been making movies continuously for longer than any other studio still in business. It is part of Denmark's Egmont group and highly visible throughout the region through its globally viewed films and TV, operation of the leading cinema chain in Denmark and Norway, and distribution of Sony PlayStation consoles and products.
Need for Change
Nordisk had been using a data warehouse built in 2007 that was set up specifically to serve the company’s Navision ERP system (now Microsoft NAV). Since that time, the business has expanded to include a much wider variety of data sources related to expanding business. The company also needed to reduce IT staff involvement and introduce innovative new analytics including machine learning.
According to Nordisk systems finance manager Mikkel Hansen, "an old data warehouse was making it impossible for us to make the necessary changes in development and to keep up with business needs. Development and integration of new data sources had stopped due to excessive development time. The old solution was running on an old system that was running out of support from the vendor. It broke down repeatedly."
To solve these problems, Nordisk Film decided that it needed to move to a new platform for handling its data. The solution chosen was TimeXtender, which was in use and recommended by a sister division of Egmont. Crucially, TimeXtender’s intuitive GUI permits non-IT personnel to create new applications for the data warehouse. This made it possible to bring data warehouse development in-house with very little training or specific IT expertise. The system is also compatible with the Microsoft-based environment, including Azure and SQL servers for data warehousing.
Adding a New Kind of Platform
The specific implementation was TimeXtender's Azure-certified Discovery Hub. Discovery Hub is designed to centralize data access and management to provide a single access point that streamlines connections and improves security. It simplifies and automates operation of the data infrastructure, providing a connection point for a wide variety of analytics tools including AI and predictive analytics. For decision makers using Power BI, Tableau, or Qlik, data conserved in Discovery Hub is easily accessed and connection to models can be automated. This permits better and timelier business decisions by providing instant access to data without data quality issues.
"Since TimeXtender was implemented, we have greatly expanded the information we are gathering to include ERP navigation, sale systems, business-specific systems such as cinema platforms and gift cards, and a range of external data from user forums, public sources, user feedback, and other data streams," says Hansen.
TimeXtender Discovery Hub addresses data access challenges by presenting a centralized data source with meaningful and trustworthy data rather than hard-to-understand raw tables. Data is combined from multiple sources with quality issues reduced or eliminated. Secure enterprisewide reporting is enabled through governed purpose-specific models that provide easy and casual data access.
The same data model is used across Power BI, Tableau, and Qlik to provide consistent reporting of any changes and the ability to meet compliance needs. This is particularly important within the European GDPR environment where regulation affects data storage in all organizations. To meet these requirements, organizations need sophisticated self-service BI platforms that include metadata, automation, governance, and security.
Immediate Benefits and Future Inspiration
Implementation of TimeXtender provided a range of benefits immediately, such as the ability to add new data warehousing development without coding as well as improved development speed -- completing tasks within days rather than weeks or months as in the previous system. Nordisk has also been able to eliminate dependency on external consultants with special skills and amplify the skills of its small (two- or three-person) team as it serves five to six highly diversified business areas.
Hansen has plans for future enhancement of the system including use of stored procedures or third-party tools such as Rundeck to enable execution of data warehouse packages on demand and move away from traditional scheduled runs. Nordisk is also implementing Qlik Sense as its BI platform.
"TimeXtender is used for extending ML and AI applications for analysis," says Hansen. "Some examples are user RFM (recency, frequency, monetary) models to target marketing and sentiment analysis of forum posts. We are also hoping to use these techniques to get information for analysis to feed our BI visualization platform."
"We consider the TimeXtender implementation a great success," says Hansen. TimeXtender will make it possible for Nordisk's analytics and data warehousing solutions to run more efficiently and move proactively into a new generation of analysis without enormous and costly change to existing vendors or systems.
As more companies need to implement advanced analytics in an increasingly complex data environment, vendors are now beginning to advance self-service into the frontiers of coding and AI. This demands some rethinking of existing systems, but it can provide clear benefits even to small teams without advanced data skills.
Brian J. Dooley is an author, analyst, and journalist with more than 30 years' experience in analyzing and writing about trends in IT. He has written six books, numerous user manuals, hundreds of reports, and more than 1,000 magazine features. You can contact the author at [email protected].