I’m sure you’ve seen the following types of claims from BI vendors:
“Get one solution for all your analytics needs”
“Get more from your data”
“Our tools put actionable information at users’ fingertips”
While these claims are not alternative facts, they do need some clarification.
The term Business Intelligence covers a wide range of toolsets including reporting, querying, dashboards, predictive analytics, data visualisation and more. What does this have to do with architecture? Having worked at one of the world’s leading architecture practices (www.hassell.com.au), I’ve had the advantage of seeing first-hand the importance of blueprints to the success of a project. I used to love walking through the office looking at the beautifully designed models and even more so seeing the models come to life in the streets of Melbourne as well as further afield. It’s impossible to believe that the same success would have been achieved had the builders “just got on with it” and just started building.
In the same way that the architectural blueprint is key for a successful building project, it is also key to a successful business system project, including business intelligence.
What is Enterprise Architecture?
According to The Open Group, a global consortium dealing with IT standards, enterprise architecture has two aspects:
- Formal description of a system, or a detailed plan of the system at component level to guide its implementation
- The structure of components, their inter-relationships, and the principles and guidelines governing their design and evolution over time
Enterprise architecture has several elements which need to be brought together:
Business architecture encompasses the strategy and vision of the organisation, governance, and key business processes.
It’s not data that adds value to the business but rather the information and subsequent knowledge that can be gained from the data. The information architecture designs the framework for translating data into information that will be used by the business users, including the processes to ensure continued data quality.
It is in the data architecture element that we define the facts and dimensions that are needed to provide the information to business. This covers the logical data model design – data warehousing for example.
The applications architecture will cover the functionality required by the data (data warehouse, data mart etc) as well as the BI environment. Available technologies should be identified and reviewed in line with your organisation’s IT strategy.
At this stage you should be looking at the software and hardware capabilities required to support the deployment of the solution.
How does this relate to Business Intelligence?
There is a tendency to start BI projects at the applications architecture stage – we’ve got a brilliant new BI tool that can do X, Y and Z; let’s get it. This is akin to the builder starting to build without a blueprint. Business software of any format, whether it be BI, ERP, WMS, SCM, CRM et al, must have a robust infrastructure underpinning them to be successfully implemented. Business Intelligence depends on accurate data being inputted to be able deliver accurate insight. Taking the time to review enterprise architecture and address any inefficiencies will not only deliver value itself, it can leverage further value from business systems.
What should businesses be doing?
- Ensure you have a clear vision, including measurable objectives
- Review your business processes and how they impact on your strategic objectives
- Define the information that is required by each of the process owners across your business
- Decide how your business is going to deal with data – categorisation, integration and storage
- Review the requirements from each of the previous stages against the various technologies; select the application(s) that best meet your requirements criteria
- Design the infrastructure that will support the selected solution
Business intelligence isn’t just about choosing the latest application that will provide a funky visualisation or run a predictive model, it’s about providing value to the business through improved decision making capabilities. For this to be achieved the underlying architecture needs to be in place.
To Learn More About Enterprise Architecture & BI, Download our White Paper HERE