Business Intelligence (BI) is definitely a hot topic in the business technology space. I read new posts daily about different BI solutions, ‘Big Data’ and trends in the BI space.
I wanted to write a post about my experience of BI from my perspective as an ERP/CRM consultant and analyst. Every business system implementation I have been a part of, business intelligence is a focus at some level and within recent years more business leaders are seeing the value in quality reporting and data discovery solutions. A lot of businesses put huge focus on collecting transactional data, but how can you use that data for positive gain?
Cut Through The Smoke & Mirrors
I feel BI solutions are one of those areas of business software that are very hard to understand what your money is buying. End-To-End BI tools are often responsible for data extraction and transformation and then provide reporting and visualisation processes. You can then look at self-service BI solutions that just provide one particular component of the BI process. E.g. Visualisation.
One piece of advice I can give is that modern BI tools demo and present very well. Popular vendors will showcase dazzling charts, drill down capabilities, drag and drop analytics all of which look fantastic at face value. However, taking these features and deploying them to your own ‘real-world’ data can be a much more challenging task.
Know Your Data
You want better reporting and strong data insight, but do you know where this information resides. Business will often have multiple disparate systems containing silos of data from different departments or processes. Often it is the combination of all this data that will allow proper insight into the business operations. It is critical to understand where the data lives and how it can be accessed before any BI implementation occurs.
If a data warehouse is already in place this may also need to be reviewed to ensure all data considerations have been made. Is the data live? What is the extraction schedule? Perhaps the data warehouse exists only for a specific analytical purpose and not all data silos and considered?
Ultimately almost all business system transactional data will need to go through an ETL (Extract, Transform, Load) process before it can be easily analysed. In simple terms, business software systems have data structures that work well for capturing transactional data but that data is rarely in a state that we easily report on.
Find An Expert
I am a big proponent of engaging the right people to give quality advice and insight. You may think this means hunting down a data analytics firm to guide to BI glory. While this is definitely one option also consider the points above, do the people you are leaning on understand your data? Can they take the ‘ugly’ and make it ‘pretty’ in data terms?
Any business leader could easily spend countless hours reading and reviewing BI solutions all of which will look impressive, but the end result is only part of the process. There is a more complex data journey that must be undertaken to truly uncover your businesses data secrets.