- Collect data from a companies operation systems and/or external sources,
- Clean and validate the data,
- Integrate the data and store it in a central repository organised by subjects,
- Deliver the data in a form accessible and understandable by non-technical users, and
- Analyse and Present the information through reports and analytical tools (spreadsheets, pivot tables, statistical analysis, data visualisation, data mining, etc.).
The Analytical Pattern will make it possible for organizations to:
- Leverage a shared data environment where data is collected, cleansed and integrated once for multiple uses.
- Use common, proven architectures, technologies and processes, and
- Operate in a simplified and shared utility computing environment.
When to Use
The Analytical Pattern should be used any time analytics functions of data warehouses are implemented. The Analytical Pattern can be used for a module of an application, a single application, or a suite of applications sharing a common data warehouse and/or operational data store as well as dashboards.
The Analytical Pattern should be used for any of these functional requirements related to structured, semi-structured and/or unstructured data:
- Analyse the past, present, and/or predict future trends. Typical functions include financial analysis, forecasting, usage analysis, performance metrics, lead finding, member segmentation, predictive analysis etc.
- Discover previously unknown patterns in large amounts of structured or unstructured data.
- Implement complex business rules (e.g. license shares, member distributions, optimising investment portfolio, detecting fraudulent actions, etc.).
- Report or analyse data from multiple sources (internal and external). Data cleansing, integration and standardisation is usually required.
- Historical reporting. This may involve producing reports from multiple, different systems that were in use at different times and with different data structures.
- Business Intelligence dash-boarding which integrates the reporting environment into one user-friendly interface.
- Improve reporting, reduce costs, or improve performance of operational systems:
- Simplify user access (making data easier to understand and navigate).
- Provide more flexible report selections and formatting options.
- Improve report performance.
- Offload reporting workload from the transactional processing environment.
- Include complex analytics in operational systems (e.g. process control charts and trend alerts; customise the user experience with accumulated knowledge about user behaviour, preferences, etc.; fraud detection; business process management; etc.)
- Data must be analysed along multiple business dimensions.
- Data quality, simple user access, and query performance are critical.
- Rapid and complex analysis of large amounts of data is required
Benefits of an Analytical Pattern
An Analytical Pattern is designed to deliver very flexible, cost-efficient business intelligence systems and robust data warehouses that can support sustained growth and respond quickly to changing business needs. The pattern’s loosely-coupled, standards-based architecture makes it possible to quickly and easily adapt to changing business needs and usage levels.
Using the Analytical Pattern helps improves a companies return on investment (ROI) for analytical applications by:
- Reducing development and implementation time and cost.
- Delivering flexible, adaptable solutions that are easy to change or extend.
- Achieving economies of scale by leveraging common infrastructure and data.
- Reducing total cost of ownership over the life of the application.
- Minimising technical barriers to exchanging or integrating information.
- Developing and sharing key analytical and data management competencies.