Skip to main content

Set Up Omni

Are you setting up your Omni app for your organization? Follow this guide to learn how to connect to your database, query your data, and promote your data to the shared model.

This guide is designed to walk through how to set up your new Omni app as an Admin user. We'll get you and your team started with a strong foundation 💪 so maintaining your environment in the future is a breeze. We recommend taking these steps when getting started with Omni:

  1. Connect Omni to your data
  2. Query your data
  3. Save resuable data to your semantic model

1. Connect Omni to your data

If you’re the first user to log into Omni, things will probably be pretty barren. That’s a-ok, all great artists have to start from a blank canvas! 🧑‍🎨

In order to query your data, first, you need to connect Omni to your database. This should be easy-peasy as long as you have the database permissions to create users and/or roles on your database and are an admin in your Omni app. You'll need credentials from your database to plug into Omni to get your data connected!

2. Build a query

Now that you have made a successful connection to your data, let's get to the fun part of using Omni - querying your data!

For quick reference, here's a simple diagram outlining the different pieces of an Omni workbook that will be referenced throughout this guide:

Workbook anatomy
  1. Menu settings, options and shortcuts
  2. Query tab
  3. Field picker
  4. Views (schema tables)
  5. Toggle to see other topics or all views and fields in the schema
  6. Toggle between query results table and visualization charts

a. Navigate to a new workbook

  1. Navigate to your Omni app's homepage:
  2. Click on the ** + New ** button in the top left corner of your Omni app's homepage.
  3. If you have more than one database connection set up, select the database connection you want to use. (If you don’t see any options here, head back to the set up steps to connect your data to Omni.)
  4. Click the Schema button and choose the database schema you want to start working with.
  5. From here you can expand the different schema tables and find the one you’d like to start with (we’ll show you how to add joins below).

b. Select your fields

Start building your query but selecting any desired fields from the field-picker menu. Beyond selecting existing fields, you have further options to customize your query outlined in the next sections of this tutorial.

c. Add a query filter (optional)

Add filters to your query to downsize the query size or target the specific data you're looking for.

i. Use a field in the query

  1. In the results tab of the workbook, hover over the field column header, click on the drop-down menu that appears and choose Filter.
  2. Set your filter conditions and value, then hit update and your query will now reflect results based on the filter set.

ii. Use a Field Outside of the Query

  1. In the workbook field picker (left-hand panel), hover over the field you want to filter on, select the three-dot menu and choose filter from the drop-down menu.
  2. Set your filter conditions and value, then hit update and your query will now reflect results based on the filter set.

d. Add a quick aggregation (optional)

Quick aggregations allow you to create new aggregation fields quickly, without writing any SQL or developing in the modeling layer. The aggregation options shown for a particular field are dependent on the field type – for example, you’ll see different options if your field is a string vs number. Learn more about the quick aggregations available by field type.

e. Add Excel Calculations (optional)

These are a quick way to add calculated fields – think basic math, rolling sums, string functions, date diffs – directly from the table (your finance teams may fall in love with this feature, don't say we didn't warn you!). For common calculations, Omni offers table calculations directly from a field's column-header menu as well as the ability to create custom calculations.

Note, the field you reference in your table calculation must be included in the query results.

i. Quick add calculations

ii. Add a custom calculation

f. Create a custom field (optional)

Custom fields allow you to reference existing fields within your topic and create a new field. For example, if you have first and last name fields in a user table, but you want a full name field instead, you can create that using a custom field. These fields can then be promoted to your model for other users to use when they navigate to this topic to explore data.

3. Add a join

Use fields from other tables in your chosen database schema by joining additional tables.

4. Create a topic

A topic is a collection of tables, fields, and joins that are pre-modeled and defined to enable quick exploration of that collection of data. Topics are especially useful for organizing your data and creating self-service for users.

5. Build a visualization

We've been mainly working in the Results tab of the workbook but now let's navigate to the Chart tab to select a visualization for our first query. The default visualization shown is based on the data in the query.

6. Promote your changes to the semantic model layer

Right now, the work you have done (e.g. creating calculations, custom fields, adding join(s), creating topic(s)) only exists in the context of this workbook. This is what we call the workbook layer. If you want to make this logic usable for other users building documents (workbooks or dashboards), we recommend promoting these workbook changes to the semantic model layer. There are two options for promoting your changes to the model layer: promote all changes at once or incrementally promote specific changes.Learn more about shared model promotion in our model management documentation.

7. Create a dashboard

Now that you have created your first query and customized your workbook, you can add that query to a dashboard. Your dashboard will be tied to the workbook you were working in. Any changes you make in the workbook the dashboard is based on will be reflected on the dashbaord once saved.