How is ThoughtSpot different to Power BI? We break down the key differences.
Updated: Jan 26, 2023
Everything is a report in Power BI
In Power BI, everything is a report. This is the interface. To be able to answer a question, you need to create a report. Even if there is only one visualisation on that report.
The report creator requires a Pro licence as a minimum, which is cheap but there are a lot of limitations on the data that you can connect to and the size of the data you can display. To share that report with a user they need to publish the report to the Power BI service and the viewer will also need a Pro licence.
In ThoughtSpot everything is built around answering questions. So everything is a search. When a user asks their questions, they will see a visualisation. The user has the ability to save that visualisation to their own Liveboard if it is something that would be useful again, or they want to share it with a colleague. But in a lot of cases an answer is all they need, so they can just move on to their next question without having the overhead of creating a report or dashboard.
In ThoughtSpot, licencing is not on an individual user basis so anyone can have access without needing to purchase additional licences.
It’s as simple as thinking of your question and writing it in the search bar.
Self-Service
With Power BI, when they talk about self-service, they are referring to the consumption of information from an existing report or a dashboard. The “self-service” element refers to the ability to apply filters and to drill down. The core assumption of Power BI “self-service” is that someone has already wanted to ask the question and has created the report.
If your answer is not already on a dashboard, then you either need to have the skills to create a dashboard, and that means being able to create a visualisation from the panel below, or rely on an analyst to do it for you.
In ThoughtSpot self service is all about asking your own questions with search. It’s as simple as thinking of your question and writing it in the search bar.
So as a business user, what makes more sense is to find the answer to the question “What are my daily sales over the last 30 days for each item type?” by simply typing this into the search bar:
With the click of the button, you will see is a chart showing you the sales for each product as a line chart, just like this:
ThoughtSpot intuitively selects the best chart based on the search you have entered. The cognitive load on the user is light and they can focus on the insights rather than wrangling a highly complex tool.
Personalisation in Power BI
Personalisation allows a user to modify an existing chart. Personalisation is not available by default, but can be enabled by the report creator. Personalisation allows a user to modify a limited number of settings on a visualisation on a report. Personalisation is limited to:
Add or remove a legend
Changing the visualisation type
Swapping out a metric or dimension
Compare two or more measures
Change aggregations
There are some additional requirements:
The user needs to use Personalise Box, which has a lot of options (see image below) and can be a bit overwhelming for some users
User explorations don’t automatically persist, so the user needs to remember to save their personal views as bookmarks
When you export to PowerPoint or PDF the Personalisation is ignored
Personalisation requires a Pro or Premium licence
Personalisation is not supported for Publish To Web
For fear of sounding a bit like a broken record, in ThoughtSpot anyone can ask their own questions with search, so you don’t need to “personalise” a visualisation, you can easily run your own search at any time. If you do find a visualisation that is close to what you need, you do have the ability to use the Explore feature on any visualisation (see image below).
Explore allows you to add, replace, filter or even compare with the click of a button on the selected visualisation. Explore provides a clean and intuitive UI that guides a business user through the customisation options, making it easier for them to make their modifications. This feature is available to all users and you can save your changes as a new visualisation if you want.
But what about Power BI Q&A?
Q&A is Power BI’s “search capability”. Power BI has has attempted to sit search on the top of the existing Power BI, without truly redeveloping the platform to deliver a true search experience for users. As such, Q&A has a number of limitations:
Limited to performing search on the report it has been applied to, which means it only searches on that individual report and not across all of your data.
Unless the report designer has given you edit permission you will be able to use Q&A, but you won’t be able to save any of the visualisations created with Q&A.
Only supports the following configurations of data sources (essentially Microsoft only products):
Import mode, which then limits the size of data significantly
Live connect mode with on-premises SQL Server Analysis Services, Azure Analysis Services, or Power BI datasets
DirectQuery with Azure Synapse Analytics, Azure SQL, or SQL Server 2019 SnowFlake, Redshift and GBQ are not supported.
The Q&A index stores a maximum of 5 million unique values. If you exceed this number, the index doesn't hold all the potential values, which might decrease Q&A accuracy.
The Q&A index is built based on the first 1,000 model entities (tables and fields). If your data model exceeds this number, the index doesn't hold all the potential model entities, which might decrease Q&A accuracy.
There are serious privacy and security issues as RLS is not applied to search suggestions. This means that a user may have values suggested that they should not have access to.
ThoughtSpot is built for scale, so you can search across billions of rows of data in seconds.
Did I mention that in ThoughtSpot everything starts from a simple search? So you are not limited to applying search to an individual report or dashboard. You can search across all of your data, even if it has a complex data model. ThoughtSpot allows you to direct query to:
Amazon Redshift
Azure Synapse
Databricks
Denodo
Dremio
Google BigQuery
Oracle
Presto
SAP HANA
Snowflake
Starburst
Teradata
Trino
ThoughtSpot is built for scale, so you can search across billions of rows of data in seconds. ThoughtSpot’s security model is pervasive, which means that RLS is applied to everything, even the search suggestions. A user will only ever see data they have permission to see.
So are PowerBI and ThoughtSpot the same?
The answer is no. Power BI and ThoughtSpot both provide BI and reporting solutions, but they have very different approaches.
Power BI has been built for analysts to create reports and dashboards for business users to consume. They have added some ability to interact with a dashboard, but if you want to answer a new question, you will need the ability to create a report or be reliant on an analyst. Power BI is licenced on a per user basis and has some fairly significant restrictions in terms of the volume of data it can handle.
ThoughtSpot has been built from the ground up to support massive volumes of data, it connects to the best of breed data warehouses and it is built with the business user at the centre. ThoughtSpot uses the power of search to make data accessible and easy for business users to be able to ask their own questions about data. The UI is simple and easy for users to understand, switch, and modify to suit their needs. Importantly, ThoughtSpot does not use a per-user licence model, which means that everyone can use ThoughtSpot, without rising licence costs.
ThoughtSpot leverages the power of search, to make self-service an actual reality.
