Databricks Data Analyst: dashboards and Genie
Creating Dashboards and Visualizations (16%) and Developing, Sharing, and Maintaining AI/BI Genie spaces (12%) are 28% of the Data Analyst Associate exam between them — around 12 of the 45 scored questions and the largest block on the paper.
This is also the most distinctly analyst content, and Genie is the section most candidates have never touched.
Dashboards and visualisations (16%)
Choosing the right visualisation for what you are showing:
| Showing | Use |
|---|---|
| Change over time | Line chart |
| Comparison between categories | Bar chart |
| Part of a whole | Stacked bar, used sparingly |
| Relationship between two measures | Scatter plot |
| A single key number | Counter or big-number tile |
| Precise values, many columns | Table |
The exam favours clarity. A question describing trend over twelve months wants a line chart, not a pie.
Parameters and filters let one dashboard serve many questions — a date range, a region selector — rather than building five near-identical dashboards.
Scheduled refresh keeps a dashboard current, and costs compute each time it runs. Matching refresh frequency to how often the data actually changes is an examinable judgement: refreshing hourly a table that updates nightly wastes money for no benefit.
Sharing and permissions — who can view, who can edit, and the fact that a dashboard does not bypass the underlying data permissions.
AI/BI Genie spaces (12%)
The section with no equivalent anywhere else in the Databricks certification track.
What a Genie space is. A curated area where business users ask questions of data in natural language and get answers, without writing SQL. The analyst’s role is to set it up so the answers are trustworthy.
Developing one:
- Choosing which tables to include. A focused, well-modelled set produces better answers than everything you have.
- Providing context and instructions — what the business terms mean, which column represents revenue, how to handle ambiguity. This is the single biggest determinant of quality.
- Adding example questions and trusted queries so the space has known-good patterns to follow.
Maintaining one:
- Reviewing the questions users actually ask, which is rarely what you anticipated.
- Correcting bad answers by improving instructions rather than telling users to phrase things differently.
- Adding context as the business vocabulary evolves.
Sharing and permissions. A Genie space respects the underlying data permissions — it does not grant access to data the user could not otherwise query.
The mental model the exam wants: a Genie space is a curated, maintained product, not a switch you flip. Questions describing poor answer quality are usually asking about insufficient context or too broad a table selection.
Sample questions
Question 1. A dashboard refreshes hourly. Its underlying table is loaded once each night at 02:00. What should the analyst change?
- A. Increase the warehouse size so hourly refreshes complete faster
- B. Align the refresh schedule with the nightly load
- C. Remove the refresh schedule and refresh manually when asked
- D. Add more visualisations to justify the refresh cost
Show answer
Answer: B
Refreshing more often than the data changes consumes compute for identical results, so aligning the schedule to the nightly load removes waste without affecting freshness. A larger warehouse increases cost, removing the schedule leaves the dashboard stale, and more visualisations is unrelated.
Question 2. Business users report that a Genie space gives inconsistent answers about 'active customers', because the term means different things in different teams. What should the analyst do?
- A. Add more tables to the space so it has additional context
- B. Restrict the space to fewer users
- C. Tell users to write SQL instead
- D. Add instructions to the space defining what ‘active customer’ means
Show answer
Answer: D
Ambiguous business terminology is resolved by adding instructions and context defining the term within the space, which is the analyst’s core maintenance responsibility. Adding more tables increases ambiguity, restricting access does not fix answers, and telling users to write SQL defeats the purpose of a Genie space.
Question 3. An analyst must show how monthly revenue has moved across the last two years. Which visualisation is most appropriate?
- A. A line chart with month on the horizontal axis
- B. A pie chart of revenue by month
- C. A scatter plot of revenue against month
- D. A counter showing total revenue for the period
Show answer
Answer: A
A line chart is the standard way to show a measure changing over a continuous time period, making the trend immediately readable. A pie chart shows composition at one point, a scatter plot shows relationships between measures, and a single counter cannot convey a trend.
What to practise
Build one dashboard with a date parameter driving several visualisations, and set its refresh to match the data’s actual update schedule.
Then build a Genie space over two related tables, ask it five business questions, and improve the instructions until the answers are right. That second exercise is the only way to learn a 12% section that no amount of reading will cover.