It’s no secret that big data is a hot new industry and the need for data engineers is growing just as fast. Companies today generate an overwhelming amount of data, and because of that, many organizations need help accessing, transforming, analyzing and leveraging that data as quickly as possible.
And in order to succeed in today’s competitive environment, you need a robust infrastructure to store and access that valuable data - and data engineers make that possible.
Not only that but data professionals need to communicate this information to non-data audiences, so they can leverage the insights to make critical business decisions - and that’s where Tableau comes in. Tableau has become an in-demand tech skill since the analytics platform and visualization tool is powerful, easy to use, and growing in popularity.
But hiring for the right data engineer can be daunting. Luckily, we’ve searched for some of the best data engineer questions (and answers to look for), with a focus on vetting for Tableau knowledge. Lets get started.
1. What is Data Engineering?
What you’re looking for:
Someone who understands that data engineering mainly refers to data infrastructure or architecture - they will probably reference the field of ‘big data' since data engineering is a popular term within that industry. The ideal candidate will know that the data that's generated by lots of different sources like mobile, social media, www(internet) is raw data, and it needs to be scrubbed, profiled, molded and aggregated for business needs.
If they use the terms ‘dark data’ or ‘dusty data’ (data that’s unstructured and untapped found in unprocessed and unanalyzed data archives or log files), then the candidate is on the right track.
Ultimately, you’re looking for someone who understands that the practice of designing, architecting and implementing the data process system which helps make the converted data useful business or competitive intelligence, is called data engineering.
2. Tell me what you know about Tableau and its products
What you’re looking for:
A candidate who knows that Tableau is an interactive data visualization product, that’s focused on providing business intelligence. It is used to connect to the data, visualize it and create interactive dashboards for further analysis - so the ideal candidate should be able to connect the dots between data engineering to Tableau.
They should be able to speak to each of Tableau’s products:
• Tableau Desktop: self-service data visualization and analytics tool that’s used to translate data images into optimized queries, and that’s easily accessible for businesses
• Tableau Server: enterprise level software used to publish desktop-based dashboards distributed through web-based Tableau server
• Tableau Online: a hosted version that makes BI more efficient - dashboards can be shared using desktop
• Tableau Reader: a free desktop application which enables you to view visualizations primarily built on Tableau Desktop. Users can filter and drill down into data but no editing features in this version
• Tableau Public: As the name suggests, this is the free version of Tableau software which can be used to make visualizations once the workbooks are saved on the Tableau server which can then be viewed by anyone.
3. From a data engineer perspective, why is Tableau useful?
A candidate’s answer to this question shows you how they understand the big picture role of data, and the role of analytics to make sense of that data.
Their answer should demonstrate that they understand how using a tool like Tableau can help make their responsibility to design, build and manage a company’s analytics databases easier. The tool allows them to churn large data sets and draw actionable business insights from it, while facilitating an uncomplicated communication method for stakeholders who may not be as data-minded as themselves.
4. What are the different kinds of data types in Tableau?
What you’re looking for:
The candidate should know that Tableau supports the following data-types:
• Boolean (true or false)
• Date Values
• Date and Time
• Geographical Values
• Number (Decimal)
• Number (Whole)
5. What does the day to day look like for a data engineer?
You’re looking for a candidate who knows that ultimately, the role of a data engineer is to transform data into a useful format for analysis, which can include things like:
• Develop, construct, test and maintain architectures (like databases and large-scale processing systems)
• Ensure architecture will support business needs
• Discover data acquisition opportunities
• Develop dataset processes for data modeling, mining and extraction
• Recommend ways to improve data reliability, efficiency and quality
6. What are measures and dimensions in Tableau?
A good candidate can provide simple definitions:
• Measures are the numeric quantities or measurable metrics of data analyzed by the dimension table
• Dimensions are descriptive attribute values used for multiple dimensions of an attribute which defines different characteristics
7. Can you explain what is the difference between .twb and .twbx extension?
The answer you’re looking for is:
• .twb is an xml document which contains all the selections and layout you made in your Tableau workbook, that doesn’t contain any data
• .twbx is the ‘zipped’ archive containing a .twb and any external files such as extracts and background images
8. How many maximum tables can you join in Tableau?
You can join a maximum of 32 tables in Tableau.
9. What are the different connections you can make with your dataset?
What you’re looking for:
A candidate who knows that you can either connect live to your data set or extract data onto Tableau, and define the difference.
• Live: Connecting live to a data set leverages its computational processing and storage. New queries will go to the database and will be reflected as new or updated within the data.
• Extract: An extract will make a static snapshot of the data to be used by Tableau’s data engine. The snapshot of the data can be refreshed on a recurring schedule as a whole or incrementally append data. One way to set up these schedules is via the Tableau server.
You’d use extract over live connection to have the ability to use the extract anywhere and build your own visualization, without having to connect to the database
10. What are shelves and sets?
The answer you’re looking for is:
• Shelves are the named areas to the left and to the top of the view. Views are built using fields by placing them onto shelves.
• On the other hand, sets are the custom fields used to define a data subset primarily based on conditions. A set can be based upon a specific computed condition.
11. What is context filter and what’s the difference between context filter to other filters?
The candidate should know that by default, all filters that you set in Tableau are computed independently. So, they should know that each filter accesses all data rows without regard to other filters, except a context filter. A context filter basically acts like an independent filter and any other filters that are set are defined as a dependent - since they will only process the data that passes through the context filter first.
They should mention why a context filter would be used (improve performance on a large data source for example).
Bonus points if they mention the other filters (quick and normal/traditional) and demonstrate their understanding with an example of how the filters work together.
12. What are the disadvantages of context filters?
Ultimately, the candidate will know both the positive and negative of a feature, so the limitations they should discuss are:
• Slow performance - often times, the context filter is not frequently changed by the user. If the filter is changed, then the database must be recomputed and the temporary table must be rewritten, which can slow down performance
• Reload time - When you set a dimension to context, Tableau creates a temporary table that will require a reload each time the view is initiated.
13. What is aggregation and disaggregation of data in Tableau?
What you’re looking for here is a candidate who talks about how both are ways used to develop a scatterplot in order to measure and compare data values.
The difference between aggregation and disaggregation is that the former is the calculated form of a specific set of values which return a single numeral value and are not user-defined, while the latter refers to viewing every data source row and analyzing data both dependently as well as independently.
14. What is the difference between data blending and data joining in Tableau?
Since the core function of a data engineer is to move data around and store it efficiently, a good candidate will know that data blending is a method for combining data that supplements a table of data from one data source with columns of data from another data source. Data joining is a method of combining related data in common fields from the same data source.
They should be able to discuss situations where data blending is more useful than data joining (for example, combining data from different databases that are not supported by cross-database joins or the data is at different detail levels).
15. What is the Tableau Server?
The answer you’re looking for:
The Tableau Server is a browser-based tool anyone can use for insight. This is for desktop and mobile. You can publish dashboards with Tableau Desktop and teams can see them throughout the business and organization. A very easy tool to use and run.
That concludes the top tableau interview questions you should use to either hire your next candidate or prepare for your next interview. Below I've outlined a higher level understanding of why businesses are looking to hire for this role in particular. After reading, it should give you a better understanding of how to position your value during the interview process and also how to be a better key player inside the business. It should help you understand why Tableau as a tool is becoming more than just that, it's becoming a position to play. At the core, you will be a multidisciplinary cross-functional player. One additional benefit I like to discuss from a leadership decision-making point of view (more on that below) is the ability to help make strategic acquisition decisions. As an employee (great to explain in the interview) you could allude to the fact that data insight can help predict and display future revenue potentials, overall market depth or traction, and point out pitfalls in qualitative thinking.
With an interview, especially when thinking about Tableau, it would be recommended that you understanding competitive products as well as the reason why the Company might be looking to hire for this position. The below information will be helpful for determining the ladder. But in general, the real only competitor to Tableau is QlikView. And the main reasons why Tableau is better is because of its data integration solution, its ability to work within multidimensional data, its support for Powerpoint, ability to have a finer viewpoint on the visuals and overall its scalability. QlikView is a healthy alternative but by industry standards, those are the main reasons why an organization would choose Tableau over QlikView.
Understanding Business Benefits of Data Storytelling
In every aspect of being associated in a business, an individual will comes across data in some form or the other. It is vital for businesses; data is what builds and makes trade possible. Information is intangible but helps make sense out of things in the real world. Data about a particular issue can say a lot of things. Hence, no wonder, companies are investing heavily in data accumulation and analysis to study the market and make informed decisions. Data can only present the facts; you need to have a narrative for data to be able to speak. Narrating or storytelling with data helps understand data more than just staring at a piece of information.
Data science, or the art of studying data from various origins, aids businesses in shaping their strategies for the present and future market to supplement their growth and profitability. This activity is more so important in today’s context of the digital era. Because of the technology, it is possible to monitor every aspect of today’s life and generate some form of data. Be it going to market, buying stuff online, video call with friends, every human action is recorded with the help of various products and services that surround us (Fitbit, Amazon, Skype). Companies analyze this information which helps them figure out how and why things are as they are, how the future could change. Data science means inferring sense from large amounts of information (a.k.a. big data).
Data, in its raw form, won’t be comprehensible to anyone. Big data needs gathering and analysis by the experts who deal with large amounts of data, segregating the useful aspects and making sense out of it. But, how does data science actually help the cause?
Use of data science in technology
This is the guiding light for management and officials to make better decisions. Data science deals with cold hard evidence, there aren’t any assumptions involved in the process. Hence, it acts as a perfect beacon for the top-level decision makers. Management and officials could benefit from a helping hand in having a consensus on important issues. With the help of data science, management and officials can make informed decisions about the various aspects of running the business.
Leading actions based on trend and helps define goals
Data scientists are paid to analyze the big data and infer sense from it by spotting the trends and patterns. These trends and patterns point out to specific controlling factors that resulted in the corresponding data generation, meaning trends indicate continuity in data and in line, help understand the reason for it. It helps organizations pinpoint the cause of rise or fall in business, and hence, strategize accordingly and take actions. Trends also show where the business is heading, so, aid authorities in defining goals based on the trajectory and conditions.
Adoption of best methods and focus on issues that matter
Any analysis performed says a lot of things. But, the vital aspect of any analysis is to get acquainted with the scenario. It helps to understand what is going on, how it is affected by others, does it affect others. Data science reveals the bigger picture of the data. It shows every aspect of business changes, meaning concerned personnel are able to focus on the issues that hold higher priority over others and respond accordingly.
A critical aspect of any business is being aware of market opportunities and taking advantage of them. Data science, through a data scientist, helps businesses identify these market opportunities. The data scientist works on the current system, evaluating its processes, tools, and assumptions; looking to improve the value derived from the data further.
Reasoning decisions with quantified, data-driven evidence
In a time when the concept of data science was not even conceived, companies sometimes had to take a risk in making decisions due to lack of appropriate insight. The emergence of data science has helped the business community by providing the much-needed insight into any matter that needs addressing. Companies can now take steps forward with confidence knowing that their decisions are backed by proven theories.
Evaluation of decisions
For an effective strategy deployment, businesses should be able to plan, implement and evaluate the strategies. In addition to the planning, data science can also help in evaluating the business initiatives implemented after acting upon a decision. A data scientist can benefit the evaluation by identifying and measuring the key parameters that are related to the changes and quantify success.
Identification and refining of target
The consumer market continuously evolves. Hence, products have to adapt to the scene. Some aspects of market evolution are the change in customer preferences or customer themselves. These market changes are reflected in the big data. Data scientists identify these changes and inform companies about the transformation. Thus, companies can tailor products and services concerning the demand and preferences to suit the customers.
Apt talent recruitment
Apart from the analysis of the big market data, data science is also capable of suggesting appropriate candidates for a particular job. With the massive amount of information available from various sources, companies are better able to examine the candidates and determine their worth with regards to their capability and needs.
This digital era forced companies to collect and look at their data in a different perspective to be competitive in the market. This change in form and medium of data accumulation led to businesses employing people that were capable of understanding these complex set of information and derive sense out of it. Data science is still the new kid around the block, hence, one won’t find many individuals associated with it. That said, data scientist, according to Harvard Business Review, is the sexiest job of the 21st century. And the best part is, there is a growing demand for data scientists, making it a lucrative career option.
Now, let’s look at some of the other job titles are associated with the data field.
Apart from data scientists, there are three more job titles, namely Data Analyst, Business Intelligence (BI) Developer, and Data Engineer.
Data Analyst role
In an organization, data analysts deal with querying and processing data, create reports by summarising and visualizing data. They need to have a strong understanding of the tools and methods associated with analyzing and presenting data. However, data analysts do not deal with big data and aren’t expected to provide insight into developing algorithms for problems.
Required skills: basic understanding of statistics, data visualization, exploratory data analysis
Tools: Microsoft Excel, SPSS, SAS, SAS Miner, SQL, Tableau, Microsoft Access
Business Intelligence (BI) Developer
While a data analyst compiles reports, BI developers work to understand the stakeholders’ reporting needs and provide corresponding solutions. They are concerned with understanding and creating reporting solutions for businesses based on the requirements. They are responsible for designing, developing and providing support for data centers, dashboards, and analytical reports.
Required skills: ETL, developing reports, OLAP, cubes, web intelligence, business objects design
Tools: Tableau, dashboard tools, SQL, SSIS and SPSS Modeler.
Storytelling with data
Data is only helpful if you can understand what it means. It is necessary that you are able to understand data to deduce findings. What I mean is, data is only good at telling what happened. To know what actually caused it, that requires some other expertise. Concerned data personnel rely on storytelling with data to make others understand the cause and effect. Storytelling with data means presenting information through stories. Why stories? Because that’s what we grew up with.
Stories add that sense of chronology to events. They create smooth transitions from one event to the other. Any piece of information that’s in the form of a story is much more comprehensible than its standard form.
In data analysis context, data tells you the ‘what’ and the story tells you the ‘why.’ While data speak about what happened, stories convey how did that happen. A story helps make you understand the whole scene behind the data.
There are three aspects of data storytelling (your core use case for Tableau responsibility): data, narrative and visuals.
Data comes from various sources about different aspects. Every act generates data. Businesses heavily rely on data for creating and understanding the market. The market acts as the source and also as the end result of data collection. One thing to remember is that data is not capable of doing anything by itself. For it to be effective and contribute, data has to take help of analysis, to provide meaning. Data only presents the facts; it does not explain the reason for it.
A narrative is the use of language in a format that enables easy understanding for the audience. It makes more sense to humans, being associated with the socializing aspect. Socialization is a critical aspect of our existence, and narratives or story form its base. Everything we convey has some story to it.
Concerning data, narrative helps data convey its existence in a medium that’s comprehensible to humans. It helps make sense of things the way we understand. A narrative provides a voice to data. Data alone can only stare at you, but when combined with a narrative speaks volumes about why things are the way they are.
Visuals are the graphical medium of presenting information. They are the most effective form of presenting data as they are capable of presenting things while being short and precise. Visuals are capable of showing but not explaining things.
What makes for a compelling data story?
For any data story to be active at what it says, it needs to incorporate the three elements – data, visual and narrative. Data forms the base of the story. It could have various characters, defined by trends and patterns, which help determine the course of the story. Visuals help form conclusions to a scene. They depict a particular scene in a single space. Visuals present data graphically for easy understanding. Narrative combines the elements together. Narratives give voice to data to speak for itself. It helps present data in a way that connects with the audience.
Every customer action today results in some form of data generation. The market is large, so it results in the accumulation of massive amounts of data in relation to customer behavior and interaction. Big data is crucial to keeping the digital market ticking, however, they themselves cannot bring out any change in the market. It is the analysis of big data that catalyzes everything. Data science supervises the big data analysis, helping businesses make better decisions, identify trends and set goals while also identifying opportunities and evaluating decisions.
Data science and/or Tableau as a tool does do the heavy lifting, figuring out what’s effective and what’s a wastage in commercial terms, but, at the end of the day, data stories connect with businesses and users the most.