Power BI Week

Building a Complete Dashboard from Scratch in Power BI

Hit the like button and subscribe to the channel

Tell me where you are from in the chat

Post on LinkedIn and Instagram Stories and tag me: @leokarpa

Some important notes:

- We will have three sessions: Monday, Tuesday and Thursday
- I am live with you in the chat during the sessions
- The database and materials are available in the link in the description
- Do not try to reproduce at the same time with me, watch first and practice later
- All recordings will be available until Sunday
- On Thursday I will provide my solution (PBIX file) and send the link for the Certificate for those
who watch Live
- I will also open the doors to my complete training program on Thursday with a very special
offer

Pretend you paid a large amount to participate in this event and be committed during this week!

Answer me in the chat:

1) never used Power BI
2) beginner
3) intermediate
4) expert

----------------------------------------------------------------------------------------
The basics

- How can I download Power BI?

https://powerbi.microsoft.com/en-us/desktop/

- How much is Power BI?

FREE

- Excel vs Power BI
 - Excel is a report-based tool. Simple to use but too many copy and paste techniques.
- Power BI is a model-based tool that you can create reports on top of a data model. Once
defined the model, you can build differenr reports with the same data.
- During the development of our dashboard I will compare both tools again.

-----------------------------------------------------
1) Extract

Where does data come from?

Generaly from an OLTP database. But the way we connect to this data can vary across organizations,
mainly due to security reasons.

How do you get the data you work with today?

1) Connecting directly in the transactional database (OLTP)

2) Connecting to a Data Warehouse (DW)
3) Through an Excel file connected to the database
4) Extracting reports from the ERP system
5) Filling out spreadsheets manually

-----------------------------------------------------
2) Transform

- What is the correct structure of a table?

- Always columnar
- Each column must represent a unique type of information

- When you deal with real data, it is usually messy and dirty
- So it is extremely important to TRANSFORM it before loading into the model

- Power Query is our best friend for this. It is the tool to prepare, transform and put the data into the
right shape
- Even better: you do this only once, because it automates all your ETL process

-----------------------------------------------------
3) Modeling

- We always have to think about two types of table:

 1) Dimension tables:
- Describe business entities, the things in our model.
- Lookup tables, the What, When, Where, Who.
- They store the perspectives, the attributes of the entities, their descriptions.
- Must contain a key column that acts as a unique identifier.
- They are used to filter and group the values you want to analyze
- You use them usualy in the axis of the visuals and on slicers.
- Usually short but wide (not so many rows and lots of columns).
- Ex: products, customers, suppliers, accounts, etc

2) Fact tables:
- Data table, the transactions, events that happen over time.
- Usually there is a date associated to an event.
- Contains dimension key columns that relate to dimension tables, and numeric columns.
- Can be very tall (and grow over time) but usualy thin (lots of rows and not so many
columns).
- Ex: sales orders, stock balances, financial transactions, temperatures, etc.

-----------------------------------------------------
4) Calculation

- DAX: Data Analysis eXpressions

- We can create columns, measures or tables with DAX

1) Columns: consume memory and are static calculations. Use it for descriptive information on
the Axis and Filters.
2) Measures: are dynamic and consume only CPU, because they are calculated on the fly. Use
always for Math (numbers that you place in the Values of the visuals);
3) Tables: not used often. It is better to create them in Power Query, with exception to the Date
table.

Where is the Power of DAX? 1, 2 or 3?

- Filter context:
- The combination of all filters that are applied on the report
- They propagate through the relationships and provide a subset of the original table, a filtered
table

- CALCULATE function: it is the only that can modify the evulation context
- Syntax: CALCULATE( <expression>, Filter 1, Filter 2, ... )
-----------------------------------------------------
5) Visuals

- What is Data Storytelling?

-> The power of building a narrative around a set of data to help convey the meaning of that
data in a powerful and compelling fashion
https://tdwi.org/portals/what-is-data-storytelling-definition.aspx
-> In other words, to guide your audience smoothly through the analysis.
-> A dashboard is like a joke. If you have to explain that's because it was not good.

- Recomendations:
1) Identify your audience
- Ask how a dashboard will be used and design for next step actions
- What information does the reader need to be successful?
- How much detail does the reader need?
2) Draw a sketch
- Good design should tell a story with data that does not become overwhelming with
way too much information, clutter or noise.
- Limit content to fit entirely on one screen.
- Keep your dashboard simple with only a 3 to 5 key values, charts, or tables. Avoid
putting too much information on a dashboard.
- If detail tables are needed, place them on the bottom of the dashboard
3) Choose the right visuals
4) Choose the right theme and keep consistency
- https://color.adobe.com/explore
5) Create a background (WOW effect)
6) Use icons and images to add context
- https://logomakr.com
7) Align all visuals properly
8) Use references to get inspiration
9) Use visualization features from Power BI (tooltip, drill-through, bookmarks...)