In my years working in data quality I have seen many skills and attributes of what a good data quality analyst must have under their belt.
As a result I believe a data quality analyst must…
Be proactive: Many times the media catches a snip-it of some disastrous event which was the result of bad data quality. Or the masses that are your company’s customers have congregated on Facebook to strike down the corporate beast you work for, because of bad billing or bad customer service.
One must ask…why is there bad billing or bad customer service? One fundamental reason is…
More Read
In my years working in data quality I have seen many skills and attributes of what a good data quality analyst must have under their belt.
As a result I believe a data quality analyst must…
Be proactive: Many times the media catches a snip-it of some disastrous event which was the result of bad data quality. Or the masses that are your company’s customers have congregated on Facebook to strike down the corporate beast you work for, because of bad billing or bad customer service.
One must ask…why is there bad billing or bad customer service? One fundamental reason is the fact that the data is not quality data. Ultimately, if you bill someone more then they should be billed, you will have a bad reputation and be considered to have poor customer service. It’s all about perception. So what to you do to prevent such things. Be proactive, don’t sit back and wait for someone to say “I think the order provisioning system has bad data?” or “We billed the Smith family $3000.00 when we should have billed them $30.00!” You don’t want those statements to come to you. You want to be proactive and look for the bad data, you want to trend the data so when it finally breaks the trend you know you have something worth looking into. You want to look at the ETL that project ABC is bringing into your data warehouse. You want to review the data dictionary and ensure that all the “t’s” are crossed and “i’s” dotted. You want to establish quality checks up front, before the data is loaded. You want to be proactive.
Be relentless: When a data issue emerges tackle it. Be ruthless, be relentless, when the developers say they don’t know the business and the business says they don’t know the script or code. Bring the two together and resolve the issue. They’ll be a time when someone passes the buck or tries to brush you off. Don’t take no for an answer. Escalate if you want to get answers. Be relentless.
Be technically savvy: Knowing basic SQL in a data warehouse environment is worth your weight in gold. At the least you must know a little SQL so that you can look at the data in different patterns and omit specific values to perform a better analysis of the data to ensure it’s quality. You do not want to rely entirely on a predefined report that someone created before your arrival. Data is always changing and you must be able to adapt and change with it. You must have some basic SQL skills and be technically savvy. Eventually your expertise will increase.
Be personable: You might ask why if I worked with data do I have to be personable? Let’s face it. You are a data quality analyst, if you are questioning the data coming into the data warehouse from the orders department, do you think they want to correct the issue for you. No they want you to fix the problem. So be personable! Scratch a few backs and someone will scratch yours. It’s all about being a team, no matter how large of an organization you are working in, you are all part of the same team, with the same goal and that is to succeed.
Know the business: A data quality analyst who isn’t business oriented would be someone who really reads reports and says, “Ah! Bad data again! We’d better fix it.” An analytical data quality expert would be intimate with the data. They can look at the trends, outliers and more and have an understanding of what the data is saying about the business. So when there’s a quality threshold spike on a data object, you immediately know it’s related to what the business is doing. This will save you investigation time and money.
