Build a data-driven app with ThoughtSpot Embedded
A hands-on guide, built around one real embedding project.
The scenario🔗
Imagine you’re a full-stack developer at SpotStay, a short-term rental company. Your team runs an internal portal that hosts use every day to check bookings, occupancy, and maintenance status across their listings. Right now, if a host wants to know why occupancy dropped for a listing, they file a ticket and wait for the data team to pull a report.
Your mandate: give hosts a self-serve view inside the portal they already use.
-
Embed a Liveboard so hosts see booking volume, occupancy rate, and average daily rate at a glance.
-
Embed Spotter so hosts can ask follow-up questions in plain language, such as "why did occupancy drop for our Austin listings this week?". Spotter answers without you building a custom query interface for every possible question.
-
Style and lock it down so it feels like part of the portal, not a third-party iframe bolted on the side.
|
Note
|
Liveboard is an AI-augmented dashboard within ThoughtSpot. |
This guide covers the build end to end, the same path you would take for your own app. It matches the SpotStay example on GitHub if you want to work through it in code. == Before you start
-
Confirm access to a ThoughtSpot instance (trial or org instance) with Developer or Admin privilege.
-
Add your app domain to the CSP and CORS allowlist (Develop > Customizations > Security Settings).
-
Confirm you have content to embed: a connected data warehouse, a data model, and a Liveboard.
-
Confirm Spotter is enabled on your instance. This is on by default for most Cloud instances. If you don’t see it, ask your ThoughtSpot administrator to enable it.
|
Note
|
Trial users: sample data is pre-connected. No data setup is required. |