What Does a Data Analyst Actually Do?

Key Takeaways

George is fit vit's first data analyst they're an ambitious startup building a unique Fitness and Nutrition app that provides users with workout ideas and offers a custom nutritional plan tailored to individual goals lifestyle and current training regime before George's arrival fitfit operated with little information on the business's performance regarding growth acquisition retention and user activity developers were too busy building the product while the business team didn't have the necessary technical skills to analyze the data collected in the database the company's Founders have a background in finance but they need to get used to working with a relational database that contains millions of rows that Excel cannot handle the founders struggled to answer such essential questions as how often do users come back to the fitfit app what is the average session duration are there functions rarely used by clients for this reason fitfit recruit a data analyst George's solid SQL skills ability to pre-process data in Python and proficiency with visualization tools like Tableau and powerbi made him the perfect candidate once George accepted the offer from fitfit the founders became excited as they hoped this would soon give them visibility on the many unknowns surrounding current performance in his first weeks on the job George had a chance to talk to everyone on the team and gain a better understanding of the business given that fitfit is a relatively small company it didn't take long to learn how they were organized and what business Logics had been incorporated into the product a proper introduction to the stakeholders in a company is essential for every data analyst to tackle business problems using data the analyst needs to understand the challenges of the business moreover George believes that it's always helpful to build a good rapport with other team members as later they'll work together to solve actual business problems and by the way thank you so much for watching our videos this means a lot to us please like the video and consider subscribing and clicking on the notification Bell if you're interested in a data analyst career check out our 365 data science program we provide a structured curriculum to help you learn all the necessary skills all right let's continue with George's story soon after onboarding George received his first ad hoc request he was asked to analyze fit vit's checkout process what happened once a user had clicked on buy inside the app the founders thought this was a suitable first task as any quick wins from this analysis would show the entire team the benefits of datadriven decision- making George agreed that this analysis had the potential to create measurable business value and proceeded to explore the data collected in the database George was glad to see that developers had built tables that collect information about all clicks made by customers during the checkout process at the same time he had difficulty working with the format the developers had used as it was code rather than meaningful names that could be analyzed and built into a funnel therefore George spent significant time mapping and pre-processing the data assigning meaningful names to each type of action once that was done he created a funnel VIs visualization that indicated how many people initiated a checkout how many dropped out during different stages of the process and how many ultimately purchased George was surprised to discover that many customers dropped out at the process's first registration stage the current way the product was designed asked for too much information and 80% of users did not continue to the second stage of adding their credit card information George presented his findings to business state holders and everyone agreed they could reduce the amount of information asked at registration saving these questions for later when clients had already purchased and invested in the product this led to a significant Improvement interactions and gave George his first quick win besides ad hoc data analysis requests George knew that a data analyst's main task was to create reports and dashboards the decision makers and stakeholders could regularly use to satisfy their informational needs in his conversations with stakeholders George understood that everyone on the team did not know how excited users were about the app how often they used it and whether they took advantage of some of its more sophisticated features so George proceeded to share these observations with key decision makers they agreed it would be optimal to create a dashboard enabling employees to track user activity and give everyone on the team a sense of common purpose bu business stakeholders defined the key questions to be answered George continued by reflecting on the types of visualizations and kpis that could be observed to answer these questions over time George then built an ETL process to provide the source data for his dashboard he used internal data sources data collected within the app and external data sources via API the company used an external payment service once the data was collected it was transformed into a format allowing the analysis in SQL George created several views combining several data tables which could be saved and used as a source for the visualizations in the dashboard to automate his report in the future he used stored procedures as these jobs will be requested regularly before creating the actual visualizations George cross referenced data from different sources to test for mistakes and ensure solid data quality then it was time to create create compelling charts in Tableau once ready George discussed the structure of his dashboard with team members who would later use it ensuring it would satisfy the informational needs of intended stakeholders we hoped you like this story if you did make sure you subscribe for our Channel and hit the like button please