ThoughtSpot, DBT partner to join BI and data modeling
The analytics vendor and open source tool have already developed integrations that combine self-service BI and semantic modeling, and now plan to fine-tune their functionality.
ThoughtSpot and DBT Labs have partnered to better enable engineers to develop self-service analytics capabilities.
Self-service analytics tools have grown in popularity in recent years as a way to ease the burdens on centralized data teams. These easy-to-use tools make analytics accessible to more than just data experts, enabling organizations to be more agile and foster growth.
ThoughtSpot, founded in 2012 and based in Mountain View, Calif., is an analytics vendor whose platform enables customers to search and analyze data using natural language processing (NLP), similar to how Google enables users to search by typing into a search bar.
DBT, which stands for data build tool, is an open source tool founded in 2016 that helps engineers transform data. Some of its data and analytics partners include Alation, Sigma Computing and Veezoo.
Working together
Earlier in 2022, the two vendors developed one integration that connects ThoughtSpot with the DBT semantic layer and another that brings DBT models into ThoughtSpot's work environment.
Already, shared customers including Chick-fil-A and Nasdaq use the integrations to generate value, according to the vendors.
Given the complementary capabilities of ThoughtSpot and DBT -- one providing the BI layer and the other data preparation -- the partnership stands to benefit both companies' users, according to Doug Henschen, an analyst at Constellation Research.
I see [the partnership] as a good fit for ThoughtSpot because it can attract the large and fast-growing base of customers using the semantic modeling power of DBT.
Doug HenschenAnalyst, Constellation Research
"I see [the partnership] as a good fit for ThoughtSpot because it can attract the large and fast-growing base of customers using the semantic modeling power of DBT," Henschen said.
Semantic modeling is a method of data modeling that reveals how data points are related to one another, in effect creating a knowledge graph that removes much of the data preparation work that can slow data analysis.
"These modeling capabilities have long been in demand in BI and analytics circles, with venerable examples including [SAP] BusinessObjects' Universes and MicroStrategy's centralized management capabilities," Henschen continued. "DBT brings the same benefits of data consistency and flexible reusability to 'modern data stack' platforms."
Similarly, ThoughtSpot co-founder and CTO Amit Prakash highlighted DBT's semantic modeling capabilities as a reason for establishing the partnership.
"DBT is becoming one of the de facto tools people use to transform their data in a modern cloud setup," he said. "We use it for defining the semantics of that data. And where ThoughtSpot customers benefit hugely from working with DBT is their time to value is shrunk dramatically because they don't have to build worksheets and do all the modeling."
While DBT's semantic modeling capabilities stand to benefit ThoughtSpot users, DBT users will also benefit from the partnership, according to Tristan Handy, founder and CEO of DBT Labs.
He noted that DBT has been clear about what it does and does not want to do; it does not want to move data or provide business intelligence. For its users to move their data and analyze it, DBT needs partners that specialize in data migration and analytics.
"We have to stay focused on what we're good at, which is helping users code their knowledge about their business into these data sets that can then be queried," Handy said. "So the thing we're always looking for as a company that does one thing well is partners that do the other things well."
Other platforms specializing in self-service analytics include Qlik, Tableau and Power BI from Microsoft -- among others -- but ThoughtSpot is the only BI platform developed around the concept of natural language search.
However, ThoughtSpot's NLP capabilities no longer exceed competitors, Henschen noted. In recent years, the vendor focused on making its capabilities cloud native and on building a more full-featured analytics stack.
"That detour gave bigger incumbents time to catch up on natural language query capabilities, though I wouldn't say they've all caught up," he said.
That said, DBT's users stand to benefit from integrations with ThoughtSpot's toolkit just as ThoughtSpot's users will benefit from DBT's capabilities, according to Henschen.
"This gives any DBT customer a shortcut to analyzing their DBT-modeled data in ThoughtSpot," he said.
A sample ThoughtSpot dashboard displays an insurance company's data.
Future plans
While ThoughtSpot and DBT developed two integrations before formalizing their partnership, the partnership solidifies their plan to continue developing capabilities together.
Prakash noted that ThoughtSpot and DBT wanted to get their initial integrations to market before formalizing the partnership. The integrations are complex, and throughout their development, the vendors worked with customers to use the integrations in production and get feedback on what works and what doesn't.
"Now, it's ready for prime time," Prakash said.
However, there is still work to be done, according to Handy. While ready for use by joint ThoughtSpot and DBT customers, the integrations are only in their first iterations, and functionality will be added over the next 12 months.
For example, the integrations don't yet support joins between different tables, Handy noted.
"It's incredibly obvious, and we must support that, but in the spirit of getting things out into the world and iterating on them -- which is how we built DBT -- we had to make choices about what to ship with and what not to ship with," he said. "You'll see our roadmap over the next 12 months fill in functionality."
Prakash added that additional plans for collaboration between ThoughtSpot and DBT center around streamlining data workflows.
