Metabase is an open source business intelligence server.
* a way to ask simple questions
* a way to save them for later
* group them into dashboards
* and share questions + dashboards with others
It lets you ask questions based on your own data. Answers are displayed in formats that make sense, whether that's a bar graph or a detailed table.
You can mark your questions to be saved for later, so you can come back to them. Or you can group your questions into visual dashboards. It's easy to share your questions and dashboards with the rest of your team.
### A Short Overview of Databases
Before we talk about metabase, it's useful to give a short overview of the terminology we will use.
A database is a collection of tables.
Before we jump into working with Metabase, it's useful to go over a few key database terms we will use.
A database is a **collection of tables**.
#### Tables
Tables have one or more columns and one or more rows.
A row is made up of cells where each cell has a value for the columns in the
An example of a table:
Tables contain one or more **columns** and one or more **rows**.
A row is made up of cells and each cell has a value that corresponds to the column it falls under.
Here's an example of a table:
| Name| Age |
| Name| Age |
| ---- | --- |
| John | 25 |
| Jenny | 31 |
Here the columns would be `Name` and `Age`. The first row would contain two cells, one with `John` and one with `25`
#### Columns
Each field has a type that describes what kind of data a field stores.
All the cells in a column contain the same type of information. For example, in the sample table above, the `Name` column contains names in each cell, while the `Age` column lists ages.
Each field has a type that describes what kind of data is stored in the field.
Examples of types:
* String Types (TEXT, CHAR, VCHAR etc) - These store things like names, addresses, and anything else that is free text.
* Numerical Types (Integer, Float, DoubleFloat, Decimal, etc) - These store numbers. Integers are whole numbers while Floats and Decimals are ways to store numbers with decimals in them. These store things like age, bank account balances, costs, latitudes, longitudes, etc.
* Time Types (Timestamp etc) - These are a special format of a number to store dates and times (or both). Sometimes databases store an integer timestamp which is either seconds or milliseconds since `00:00:00 Coordinated Universal Time (UTC), Thursday, 1 January 1970`. This is a convention that allows for compact storage of timestamps.
* IDs - Also called primary keys
***String Types** (TEXT, CHAR, VCHAR, etc) - These fields store things like names, addresses, or anything else that is free text.
***Numerical Types** (Integer, Float, DoubleFloat, Decimal, etc) - These fields store numbers. Integers are whole numbers; Floats and Decimals are ways to store numbers with decimals in them. Numerical types store things like age, bank account balances, costs, latitudes, and longitudes, to name a few.
***Time Types** (Timestamp, etc) - These fields are a special format of a number used to store dates and times (or both). Sometimes databases store an integer timestamp which is either seconds or milliseconds, such as `00:00:00 Coordinated Universal Time (UTC), Thursday, 1 January 1970`. This convention allows for compact storage of timestamps.
***IDs** (Also called **primary keys**) - This field uniquely identifies each row. For example, imagine a car reservation app where you can book a car in advance. The ID of the reservation could be the reservation number (no two reservations would share the same reservation number, making the ID unique).
**Customer**
| ID | Name| Age |
| ---- | --- | --- |
| 11| John | 25 |
| 12| Jenny | 31 |
In the above table, the `ID` field is an ID (primary key). The `Name` field is a string type and the `Age` field is a numerical type (specifically an Integer).
#### Relationships
Tables can contain references to other tables. For example, imagine a car reservation application where one can book a car in advance.
Our application would have two tables, one for reservations, let's call it `Reservations` and one for customers, called `Customers`.
To connect reservations to customers, it is typical in databases to use a field called a `Foreign Key` where the contents of the field are the same value as the ID value of the connected row in the other table.
For example, in our hypothetical application, we could connect each reservation to the customer that made the reservation by having the `customer` column of the reservation contain the same value as the `id` column of the customer who made the reservation.
Tables can contain references to other tables.
For example, in our hypothetical car booking app, we would have two tables, one for reservations (let's call it `Reservation`) and one for customers, (we'll call this one `Customer`).
To connect the reservation data to the corresponding customer data, you can use a `Foreign Key`. A **`Foreign Key`** is a field used in databases when the contents of its field are the same value as the ID value of a connected row in the other table.
For example, in our hypothetical car booking app, we could connect each reservation to the customer that made the reservation by having the `Customer` column of the reservation contain the same value as the `ID` column of the customer who made the reservation.
##### Reservations
**Reservation**
| ID| Customer | Time |
| --- | ---- | --- |
| 1 | 11 | 12/20/2015 |
| 2 | 12 | 1/2/2016 |
| ID| Customer | Time
| ---- | --- | --- |
| 1 | 11 | 12/20/2015 |
| 2 | 12 | 1/2/2016
##### Customers
**Customer**
| ID | Name| Age |
| --- | ---- | ---|
| 11| John | 25 |
| 12| Jenny | 31 |
| ID | Name| Age |
| ---- | --- | ---|
| 11| John | 25 |
| 12| Jenny | 31 |
If your database contains these kinds of relationships (and many do), you can use linked objects in this way. So if our hypothetical application database were examined with Metabase, we could ask questions like:
-- Example of a join --
If we wanted to analyze our hypothetical app's database with Metabase, we could ask a question, like:
-- Example of a join --
### Questions
The core concept in Metabase are Questions and their corresponding Answers.
Metabase's two core concepts are questions and their corresponding answers. Everything else is based around this functionality.
Questions are made up of a number of parts:
* source data - database + table. Typically you will be working with a single table in your database as the starting point for you questions.
* optional filter - this narrows down the source data to an interesting subset, like "active users" or "bookings after June 15th, 2015"
* And a view -
* either bare rows or
* aggregation clause -, count, etc
* optional group by field or fields
* Advanced Options
* Limit to 1, 10,1000 or more entries
***Source Data** - Source data comes from a table in your database. Typically, users use a single table from their database as the starting point for their questions.
***Optional Filters** - Filters narrow down the source data to an interesting subset, like "active users" or "bookings after June 15th, 2015"
***View** - There are a few ways to view the answer to your question
**Bare rows* are an output of each individual record that matches the criteria of your question.
**Aggregation Clause* allows you to ask Metabase to aggregate a question's answer in a particular way, such as taking the sum, average, or total count of values.
**Group* data by a particular field of your table.
***Advanced Options**
* Limit your results to 1, 10,1000 or more entries
* Sort by a column: either by clicking on the column header or by selecting the column in the advanced section
And can be visualized in a number of ways
Answers to questions can be visualized in a number of ways:
*scalar
*table
*charts
*maps
*Scalar
*Table
*Charts
*Maps
Once answered they can be saved, favorited or downloaded
Once a question is answered, you can save, favorite, or download the answer.
### Dashboards
* made up of multiple cards in a given position
* used to share groups of questions that should be
A **dashboard** displays groups of questions and their corresponding answers. Each question is on its own card that you can move or resize as you see fit.
Have a few key performance indicators that you want to be able to easily check? Make a dashboard featuring questions that show your key metrics and share it with your team.
You can make as many dashboards as you want. Keep them private or share them with others on your team.
## Asking questions
## Asking Questions
### Raw Data:
Just a table with the rows in the answer.
Useful to see the data you're working with, or for exploring small tables.
Useful when combined with Filtering to see segments of interesting Users, venues, etc. These segments can be saved and passed around.
Raw Data is a table with the answer listed in rows. It's useful when you want to see the actual data you're working with (rather than an aggregate sum) or when you're exploring a small table with a limited number of records.
### Different kinds of metrics
What is a metric? It's a number that is derived from your the source table (filtered by any filters you've added described later).
Apply filters to see segments of interesting users, venues, or groups, etc. You can save these segments and share them with your team.
### Different Kinds of Metrics
What's a *metric*? It's a number that is derived from your source table and takes into consideration any filters or elements you asked Metabase to apply to your question.
#### Basic Metrics
* Count: Total of number of rows in the answer
* Sum: Sum of all values in a column
* Average: Average of all values in a column
* Advanced Metrics
* Number of Distinct Values: Number of unique values of a column among all the rows in the answer
* Cummulative Sum: Additive sum of all the columns, eg. total revenue over time
* Standard Deviation: Number which expresses how much the values of a column vary among all rows in the answer
### Breaking out metrics
Metrics are single numbers.
Often you'll want to know more detailed information.
For example, the sum of all invoiced amounts is a metric.
It's natural to want to look at this metric across either time or another dimension, such as whether the invoice is paid or not.
You can do this by adding a "Grouping" to your question.
You can breakout by any date or time dimension you have, as well as any category field.
If you break out by a single dimension you get a table where the leftmost column is the dimension and the rightmost column is the value of the metric for that point.
Two dimension breakouts are equivalent to a pivot table in excel and are one of the workhorses of the business intelligence world.
If you break out by additional dimensions, you will add columns to the left of the dimension.
***Count:** Total of number of rows in the answer. Each row corresponds to a separate record.
--Exmaple of a 2 and 3 d breakout--
***Sum:** Sum of all the values in a column
***Average:** Average of all the values in a column
### Filtering your data
#### Advanced Metrics
***Number of Distinct Values:** Number of a column's unique values among all the rows in the answer (Ex. number of different types of items sold in store—*not to be confused with the total number of items sold*)
You can filter by any fields in the table you're working with or any connected tables through foreign keys in that table.
***Cumulative Sum:** Additive sum of all the columns (Ex. total revenue over time)
***Standard Deviation:** Number which expresses how much the values of a column vary among all rows in the answer
### Breaking Out Metrics
Metrics are a single number. Often you'll want to know more detailed information.
For example, the sum of all invoiced amounts is a metric. It's natural to want to look at this metric across time or another dimension, such as whether the invoices are paid or not.
You can do this by adding a **Group** element to your question. You can break out your answer by any date or time dimension in your table, as well as any category field.
If you apply a *single dimension* to your question, you get a table where the leftmost column is the dimension and the rightmost column is the value of the metric for that dimension's value.
*Two dimension* breakouts are equivalent to a pivot table in Excel and are one of the workhorses of the business intelligence world.
If you break out by additional dimensions, you will add columns to the left of the dimension.
### Filtering your Data
You can filter your data by any fields in the table you're working with or any connected tables through foreign keys.
Different fields can have different filters based on them.
All fields can be filtered by:
There are four universal filters that can be applied to any field. These filters are:
* is a value, eg. "Status is 'closed'"
* is not a value eg. "Status is not 'closed'"
* is null, i.e. it isn't set in the record
* is not null
**is a value,* eg. "Status is 'closed'"
**is not a value* eg. "Status is not 'closed'"
**is null*, i.e. it isn't set in the record
**is not null*
Some fields will have a limited number of possible values. Metabase will pick up on this and limit the choices in the filter selection to only valid values.
Some field (eg. price) will have too many.
Some fields have a limited number of possible filters. Metabase will pick up on this and limit the choices in the filter selection to only valid values.
Some fields (eg. price) will have many possible filters.
Fields that are comparable, like numbers or dates, can also be filtered using the following operators:
* Less than a value you enter
* Greater than a value you enter
* Between two values you enter
**Less than* a value you enter
**Greater than* a value you enter
**Between* two values you enter
If filtering by dates, a datepicker will appear to allow you to input dates easily.
If filtering by dates, a datepicker should appear to allow you to input them easily.
## Digging into Individual Records
Click on a record's ID to see more information about a given person, venue, etc. You can see all fields related to that ID and all connected tables that are hidden for readability in the standard dispaly.
## Digging into individual records
* you can click on ids to see more info about a given person, venue, etc
* you can see all the fields that are hidden for readability
* you can see all connected tables
## Asking more Advanced Questions in SQL
If you ever need to ask questions that can't be expressed using our interactive GUI (guided user interface) query builder, you can use **SQL**.
## Asking more advanced questions in SQL
*What's SQL?*
* If you ever need to ask questions that can't be expressed using our interactive GUI query builder, you can use SQL.
* SQL is short for Structured Query Language and is a widely used standard for getting data from databases. To learn more about it read: [SQL Tutorial](http://www.w3schools.com/sql/default.asp)
* Even if you don't understand SQL or how to use it, it's worthwhile understanding how to use SQL inside Metabase as sometimes other people will share SQL based questions that are useful to you.
* You can switch a card from GUI mode to SQL mode by clicking on the cursor icon in the upper right
* Once there you can write SQL directly.
* To see how it work try running `select count(*), date from purchases group by date`. Don't worry if you don't understand this just yet. Click 'run query' and note the table that comes back is the same as if you had graphed "count" broken out by "date"
* Note also that you can treate the result the same way you would treat a result of the GUI, and yu can save it, download the results, or add it to a dashboard.
* SQL (pronounced "sequel") is short for Structured Query Language and is a widely used standard for getting data from databases. To learn more about it read: [SQL Tutorial](http://www.w3schools.com/sql/default.asp)
Even if you don't understand SQL or how to use it, it's worthwhile to understand how to use it inside Metabase because sometimes other people will share SQL based questions that are useful to you.
###Using SQL
* You can switch a card from GUI mode to SQL mode by clicking on the cursor icon in the upper right hand corner.
* You can write SQL directly into the text box that appears.
* To try it out, type the command `select count(*), date from purchases group by date`. Don't worry if you don't understand this just yet. Click 'run query' and note the table that comes back is the same as if you had graphed "count" broken out by "date".
Questions asked using SQL can be saved, downloaded, or added to a dashboard just like questions asked using the GUI.
## Visualizing results
While tables are useful to lookup information and to pull out hard numbers, it is usually easier to see trends and make sense of data using graphs.
Metabase offers a number of charting options.
### line graphs
### Area graphs
### Bar + Pie graphs
## Maps
### US State maps
While tables are useful for looking up information or finding specific numbers, it's usually easier to see trends and make sense of data overall using graphs.
Metabase offers a number of charting options:
### Line Graphs
### Area Graphs
### Bar & Pie Graphs
### Maps
Metabase offers three different map visuals:
### United States Map
* default GeoJSON
* requires a column that is a state in the result
* expect states as abbreviations or full names
### Country maps
* Creating a map of the United States from your database requires there to be a column with states
* State names will be abbrievated or appear in full
### Country Maps
* default GeoJSON
* requires a column that is a country in the result
* expect countrys as full names or abbreviations
### Pin maps
* Countries will appear as full names or abbreviations
### Pin Maps
* requires your admin to set up a google maps api key
* requires both a latitude and a longitude marked column in the result
* heat maps vs pin maps
##Saving Questions & Dashboards
---
##How to Save a Question
##What is a dashboard?
A **dashboard** is a collection of saved questions that you want to refer back to regularly. Questions on a dashboard are meant to be consumed together. For example, you could have a dashboard devoted to metrics about the number of reservations. It could show the number of reservations by day of week, time of day, geographical region, and more.
## Saving questions + dashboards
### what are dashboards?
You can have as many dashboards as you want. **Public dashboards** (public only to those within your organization!) can show canonical KPIs that your whole team regularly checks in on. Use your **personal dashboards** for projects or for keeping track of your own areas of interest.
* a collection of saved questions that are meant to be referred back to regularly
* meant to be consumed together
* public dashboards contain canonical KPIs, etc
* personal dashboards can be used for projects and deleted or for long standing areas of interest
### arranging dashboards
to make any changes to a dashboard, click on the edit button
this will allow you to resize, reorder or remove cards
##How to Arrange Dashboards
To make changes to a dashboard, clicking **Edit** will allow you to resize, reorder, or remove individual question cards.
* resizing cards
clicking on the borders or the handle on the lower right corner will allow you to drag the cursor to resize the card
* reordering cards
to reorder cards, click+drag the card to its new place. Note that other cards will rearrange as you do this.
* removing cards vs deleting cards
when you click on the "x" icon t
* To resize a card, click on its borders. You can also use the handle in the lower right corner, which will let you drag your cursor to resize the card.
* To reorder your cards, click and drag the card to where you want it to be. Other cards will rearrange as you move the card to its new place to make room for it.
* To remove a question card, click the **X** icon on the card.
### deleting dashboards
when you delete a dashboard, you aren't deleting the cards
be careful deleting public dashboards
##How to Delete a Dashboard
Deleting a dashboard does not delete the individual question cards on a dashboard. The question cards will still be saved. Deleting a dashboard deletes the grouping of the questions (but not the questions themselves). Be careful when deleting public dashboards - make sure other users in your organization do not need the dashboard.
### tips on creating dashboards
##Tips on Creating Helpful Dashboards
To make a dashboard that is actually helpful, decide what you want the dashboard to tell you about your data. What questions will give you insight into what you want to know?
* figure out which cards you need
* once you have all the cards, try to find a structure that will allow a user to consume related information together
* if you have more than 10 cards, strongly consider
Once you know what questions you want to know the answer to, make a question card for each question and save them. Then save the cards to the dashboard. Once you have the cards you want, structure the cards so that users can easily consume the information. Place the most important cards near the top of the dashboard or make them bigger than the other cards. Doing so, will naturally draw users' attention to the cards.
## Getting help on your data model
If you have more than 10 cards on a dashboard, consider breaking the dashboard into two separate ones. You don't want to overwhelm users with too much information and each dashboard should revolve around one theme. Remember you can make as many dashboards as you want - there's no limit!
##Getting Help on your Data Model
---
* You can get to the data model reference at any time by clicking on the book icon
* You will see a list of databases available to you and the tables in them
* pick the table that is relevant to you
...
...
@@ -224,4 +249,3 @@ be careful deleting public dashboards
