The amount of data in the world is exploding because almost every electronic device in every industry around us produces data. Thus, one of the big issues that most companies face is how to understand and deal with all this data. And that is where data professionals that work in the business intelligence (BI) space come in. Nic Smith from Microsoft BI Solutions Marketing, said this, “BI is about providing the right data at the right time to the right people so that they can take the right decisions”.
So, how do you find a good BI professional? Well, it all starts with what you ask them during the interview process. You need to get a sense if they really understand the tools and methodologies of dealing with data.
Interview questions for business intelligence developers
What does ETL mean?
- ETL (extract, transform, load) refers to a process that extracts data from data sources, transforms the data for storing it in proper format or structure for querying and analysis purpose, and loads it into the final target (i.e. database or data warehouse).
What is SSIS? SSAS? SSRS?
- These are components of Microsoft’s SQL Server software. Integration Services (SSIS) is used as a platform for data integration and workflow. Analysis Services (SSAS) is used for special databases which provide fast analysis of large amounts of data. Reporting Services (SSRS) is used to quickly generate reports.
What’s the difference between the Inmon and Kimball data warehousing models?
- These are the dominant models that are used in data warehousing. At a very high level, Bill Inmon’s approach follows the idea of having a normalized enterprise data warehouse, which sits at the center of the Corporate Information Factory (CIF) (http://www.inmoncif.com/home/). Ralph Kimball’s approach uses a dimensional approach, where transactional data is broken into facts (usually numeric) and dimensions (provides context for facts) (http://www.kimballgroup.com/). In order to determine the level of depth of the candidate, I like to see if they can explain some of the following: advantages and disadvantages of each model, the place of data marts and operational data stores, database normalization, differences between OLTP and OLAP databases, star and snowflake schemas, conformed dimensions, and slowly changing dimensions.
What is Big Data?
- I ask this question because it is very open ended and that will help you see what the interviewee grasps about this subject. One huge plus I look for is if they can give examples of how they have encountered big data and what they were able to do. But, the basic idea is that big data refers to large and complex data sets that are difficult to process using traditional relational database systems. Some of the characteristics of big data are: volume (size), variety, velocity (speed of data generation), variability (inconsistent), veracity (quality), and complexity (http://en.wikipedia.org/wiki/Big_data).
Draw a picture of how the SQL Server stack works and how the following pieces interact (Databases, Data Marts, Data Warehouse, SSAS, SSIS, SSRS)
- Personally, I always like having an interviewee draw or diagram out something. This way you can see how they will communicate while trying to help you visualize something. On my team, we are always using whiteboards for designing and explanation, so I view it as an important skill.
As you can see, although I provided some basic answers, these are NOT meant to be comprehensive. Additionally, I feel like the interviewer needs to have a good measure of knowledge about the questions they are asking. That way they can know whether the interviewee knows what they are talking about. Also, I generally ask general database questions that I feel like every database professional should be able to answer (https://sqlbadboy.wordpress.com/2013/07/31/interview-questions-part-1/) and maybe even some that are specific to DBA’s (https://sqlbadboy.wordpress.com/2013/11/15/interview-questions-part-2/).