Today, large organizations routinely use “business intelligence” (BI) tools to figure out what’s going on in their operations. This gave rise to many unwieldy behemoths in the software world. Now, British startup Fluent has completed a $7.5 million seed round investment led by Hoxton Ventures and Tiferes Ventures, aiming to apply artificial intelligence-based large-scale language models (LLM) to business databases to make it easier for ordinary business people Query them.
Essentially, BI tools connect to business repositories, use SQL to build visualizations and build BI dashboards. There are many big companies involved in this space: Tableau (owned by Salesforce), Power BI (owned by Microsoft), Looker (owned by Google), and QuickSight (owned by Amazon), among others.
And the market is huge. According to a report, the global business intelligence market will be US$27.11 billion in 2022 and is expected to grow from US$29.42 billion in 2023 to US$54.27 billion in 2030. Gartner believes this market could be even larger if AI and LL.M. are more widely used. .
However, data teams spend a lot of time building these dashboards, especially for large organizations. There’s always the challenge of getting people to actually see them—a difficult task when the data team groans at the thought of filling requests that might take days to build.
Instead, Fluent wants to be a “conversation layer,” using a natural language LLM that sits on top of a company’s data warehouse. It converts these questions into SQL and automatically produces the answers faster. So anyone, regardless of technical skills or business background, can ask questions and gain insights about their data in plain English, the company claims.
Of course, this could mean huge improvements in response times. “Advisors went from waiting two weeks to get insights to 30 seconds,” Fluent CEO Robert Van Den Bergh told me in an interview. “That means they ask more questions, use data more in their work, and data Now it’s within their reach.” Fluent’s customers already include Bain & Company.
While he acknowledged that Fluent “primarily uses Azure OpenAI’s GPT4 model,” he emphasized that it is not a new startup with an “OpenAI wrapper.”
This simplistic approach does not work in the work of generating accurate SQL and therefore cannot correctly answer data questions in the context of a BI tool, he claimed: “Through 18 months of work, we have been able to build a method to do this Organizations like Bain & Company can trust and leverage these answers across their organization.”
“Fluent’s platform helps us leverage LLM to query large, complex data sets and deliver insights,” Bain & Company partner Ian Weber said in a support statement. “Fluent enables our advisors to get what they need quickly, efficiently and accurately.” Answers, especially for questions that are too complex or specific for pre-built data dashboards.”
“Is it really true that all business users want answers to their questions? They don’t want to be models. They want to know how this customer is doing compared to this customer? Or how are we doing here? This time,” Vandenberg said. How effective was the marketing campaign?” He said that while other players in the market are targeting data users, Fluent is targeting the business market, not data.
The field of natural language querying has only recently become possible, so it’s not a crowded market.
For example, Metabase is an open source analytics and business intelligence application that makes it easier to build dashboards. The San Francisco-based company has raised $51 million to date.
Einblick is a US company that was recently acquired by Databricks (which is preparing to go public). It seems to be the closest player to Fluent on the market, but Fluent claims their product is geared toward more technical users on data teams.
Thoughtspot is valued at $4 billion and now also has a natural language query system.
