4 Habits of a Bad Data Analyst

First impressions are hard to change and if you’ve just started at a company as a data analyst you don’t want to start off with a bad one. I’ve worked with my fair share of bad data analysts over the years and want to share their bad habits and how to avoid them.

1. Neglects Data

The worst bad habit I’ve seen is data analysts that don’t do any exploratory data analysis ( EDA ) on new tables and jump immediately to writing SQL. When I was a consultant I had a new colleague join two tables together without first looking at the data to realize there were duplicates on the join key. This caused a cartesian join that ran for hours and halted the daily ETL jobs until the DBA killed the query.

Good habit — Develop a standard list of things to check in your EDA for tables you’ve never worked with. For example, query the first 100 rows to get a sense of the field values. Check for duplicates on join keys you plan to use. For time series, graph the values to look for outliers and missing days. This will help you understand the data and write the proper SQL to get the correct results.

2. No Validation

A bad analyst will deliver numbers without checking against another source. This source is typically a dashboard or report that has correct numbers and can be used to sanity check the calculated numbers. An analyst once showed me conversion rates that were 20% higher than the company’s dashboard conversion rates. The analyst didn’t even realize the discrepancy until I pointed out the number looked too high compared to conversion rates I’ve seen in the past.

Good habit — Know your company KPIs and always verify your calculated numbers against official dashboards or reports. Bonus points for remembering the ranges off the top of your head. For example, if you know the conversion rate is normally 1% and you just calculated a rate of 2%, this should raise red flags to check if either your SQL or data source is correct. If there are no sources to verify against, ask another analyst or your manager to look over your numbers before you present them to the stakeholder as a sanity check.

3. Lacks Independence

A bad analyst will ask questions without attempting to find the answer on their own. This can be due to a lack of experience for a junior analyst but if this is a senior analyst then it’s a sign of a bad analyst. Most technical questions can be answered with a Google search. A data question can often be answered by looking at the table and field values. A colleague once asked if the order total in the company’s sales table excluded returns. If my colleague had even looked at the table it would’ve been obvious there were multiple fields to chose from with one that excluded returns and one that didn’t.

Good habit — Avoid your first instinct to ask a question right away and try to find the answer on your own. If you still can’t figure it out ask another analyst or your manager and explain what you looked at already. This will show you tried to look for an answer first before asking for help.

4. Speed Over Accuracy

Bad analysts rush through analysis quickly and often make careless mistakes that could’ve been caught if they took the time to check their work. I once worked on a request with the assistance of a junior analyst and while the work was done quickly the analysis was riddled with errors. I ended up using more time to verify the numbers than if I had done the entire request myself. I had to tell my manager I couldn’t work with this analyst anymore because I was spending too much time checking their work and not doing my own tasks. This left my manager with a bad impression of the analyst and could’ve been prevented if the analyst had taken the time to check their work.

Good habit — Don’t rush through tasks and stop to verify your numbers. It’s worse to deliver the wrong numbers to your stakeholder and leave a bad impression than it is to take a little more time to deliver the right numbers.

Final Thoughts

Ensuring stakeholders and senior management have a good impression of you will help with performance reviews and promotion decisions. Now that you’re aware of bad data analyst habits, you can avoid them and follow the good ones.