Are you wondering what a data analyst does? Or maybe you are wondering what the difference between a data analyst and a data scientist is? Well either way, you are in the right place! In this article we tell you everything you need to know to understand what the average data analyst does and how data analyst roles compare to other data roles.
This article starts out with a discussion of what a data analyst and how the average data analyst spends their time. After that we compare data analyst roles to similar roles such as data scientist roles. Finally, we provide some examples of projects that a data analyst might work on to give you a better idea of what a day looks like for a data analyst.
What does a data analyst do?
So what does a data analyst do? Job titles and responsibilities will vary from company to company, but in general the data analyst should be the first one of defense that business stakeholders turn to when they have questions that need to be answered with data. That means that data analysts typically work closer with their business stakeholders than other data professionals.
The job of a data analyst starts at the very beginning of the data collection process, where they are generally responsible for developing high level metrics and ensuring that proper tracking is in place to collect the data needed to calculate those metrics. After the data has been collected, the data analyst will clean, aggregate, and analyze the data to answer questions for their business stakeholders. In addition to answering ad hoc questions, they will create long term dashboards to display key metrics and trends over time.
Responsibilities of a data analyst
What are the responsibilities of a data analyst? Here are some common responsibilities of a data analyst.
- Work with business stakeholders to identify opportunities for data-driven optimizations
- Define health metrics for key initiatives
- Ensure that appropriate data tracking is instrumented
- Build dashboards to surface key metrics
- Plan and analyze simple experiments and AB tests
- Analyze data and perform statistical tests
- Create intuitive data visualizations
- Present the results of their work to technical and nontechnical audiences
How do data analysts spend their time?
How much time do data analysts spend on ad hoc requests?
How much time does the average data analyst spend on ad hoc requests? Data analysts tend to spend more time on ad hoc requests than other data professionals. That is not to say that data analysts never work on long term projects, but data analysts should expect to be faced with ad hoc requests from their stakeholders.
Data analysts tend to face more ad hoc requests because they are the go to people that business stakeholders turn to when they have questions that need to be answered with data. This makes data analyst roles great for people who like to work on a variety of problems with a variety of people.
What tools do data analysts use?
What tools do data analysts use? Here are some common examples of tools that are used by data analysts.
- SQL is used to extract, clean, and analyze data
- Tableau is often used to create dashboards that surface key metrics and insights
- R may be used to analyze data and perform statistical tests
- Excel may be used to analyze data
How much time do data analysts spend in meetings?
How much time do data analysts spend in meetings? Data analysts tend to spend the most time in meetings out of all of the data professionals, especially at the individual contributor level. There are two reasons for this. The first is that data analysts tend to work more closely with their business stakeholders than other data professionals. The second is that data analysts tend to spend more time on short term projects and ad hoc requests than other data professionals.
Each time a data analyst is pulled into a new project, they will need to have a meeting or a series of meetings to understand the context of the problem. There may be additional meetings to flesh out the project requirements as the data analyst starts to get some initial results. Once the data analyst is done with the analysis, there will generally be another meeting or two to share the results of the project with stakeholders.
Who do data analysts work closely with?
What kind of stakeholders do data analysts work closely with? Here are some examples of stakeholders that data analysts work closely with.
- Business stakeholders. Data analysts work very closely with business stakeholders and generally meet with non-technical stakeholders more than technical stakeholders. Data analysts are often embedded within a specific business unit that they work with regularly. They are more likely to consistently work with the same group of stakeholders than data scientists, who may work with different groups of business stakeholders on different projects.
- Developers and data engineers. Data analysts may also meet with developers and data engineers to ensure that data is being tracked correctly. This is especially true when new features that do not yet have any tracking are being implemented. These meetings help to ensure that the underlying data is complete and accurate.
- Data scientists. Data analysts may meet with data scientists for a variety of different reasons. For example, a data analyst might meet with a data scientist that they are collaborating with on a project. They may also meet to exchange knowledge about different datasets and metrics. Sometimes a data analyst will meet with a data scientist to be mentored if they wish to move into a data science role.
- Other data analysts. Finally, data analysts will also meet with fellow analysts on their team to give updates on their projects and get feedback on the methodology they are using.
Types of data analyst roles
Are you wondering what types of data analyst roles exist? Here are some examples of different types of data analyst roles.
- Embedded analysts. The first type of analyst we will talk about is the embedded analyst. This is an analyst that is embedded within a small group of business stakeholders and developers that focus on a specific product. These analysts will work primarily with the team they are embedded in and do not spend as much time meeting with other analysts in their organization. This type of analyst tends to be the most heavily involved in business decisions since they work the most closely with their business stakeholders. That means that embedded analyst positions are great for analysts who have a lot of input to give on business decisions.
- Centralized analysts. Centralized analysts, unlike embedded analysts, tend to work with a variety of different stakeholders. They may or may not have a specific business unit they are dedicated to, but if they do then it is usually a larger business unit that has multiple analysts dedicated to it. For example, there may be a centralized team of analysts that are all dedicated to advertising. These analysts will work exchangeably on a variety of projects and products spanning the advertising business. Centralized analyst roles are great for analysts who want to work on a variety of different products.
- Analyst-scientist hybrid. The next type of analyst we will talk about is the analyst-scientist hybrid. An analyst-scientist hybrid may operate within an embedded or centralized framework. What distinguishes the analyst-scientist hybrid from other types of analysts is the methodologies they work with. This type of analyst will have some more advanced knowledge of statistics and be comfortable using simple machine learning models like linear and logistic regression models. They may work on a mixture of predictive and descriptive work.
How do data analyst roles compare to other data roles
Difference between data analyst and data scientist roles
Data analyst roles are generally more similar to data scientist roles than other data roles. Data analysts and data scientists often work on projects with similar objectives, but they generally use different methods to achieve these objectives. For example, a data analyst and a data scientist might both work on a project with the goal of creating segments of users to help understand the customer base. The data scientist would likely use unsupervised clustering algorithms to complete the project, whereas the data analyst may use more simple data aggregations.
Data scientists tend to work on projects that use more complex methodology and take more time to complete. This means that data scientists tend to work on just one or two projects at a time, whereas data analysts often work on multiple projects at once.
Examples of data analyst projects
What are some examples of projects that a data analyst might work on? Here are some examples of projects that data analysts might work on..
- Design and analyze an experiment. One common task that data analysts do is design and analyze experiments. These experiments are often used to test new features that are going to be added to the product. Before the experiment is run, the analyst will work with their stakeholders to estimate the sample size that is required for the experiment and determine how observations will be randomized into groups. They will also ensure that any data tracking mechanisms that are required are properly implemented. After the experiment is run and all of the data has been gathered, they will analyze the data and perform statistical tests to determine whether there are differences between the experimental groups.
- Develop health metrics and dashboards to track them. Data analysts will also develop health metrics to monitor how certain products are performing and build dashboards to monitor these results. These dashboards will refresh frequently to keep the results up to date
