tl;dr - Definite is a better way to do analytics. It's like Figma for data with a healthy dose of magic (AI) sprinkled in. We’re opening up the product to more teams (sign up here).
I’ve been writing SQL for 17 years. It was the first programming language I learned. It felt good being one of the few people that can “pull data” on my team.
That feeling quickly wears off when you’re drowning in a flood of ad hoc data requests. Knowing SQL is a superpower, there's never enough people with the power to do all the data work your team needs.
There’s been a lot of action in the modern data stack. Tools like Fivetran have made writing custom ETL scripts a thing of the past. dbt has changed the way transformations are done. Snowflake went head to head with Google and Amazon and won.
But it still takes an inordinate amount of time to go from raw data in a source system to a final result people can use to do their jobs. Most people pull up a dashboard, can’t find the answer they’re looking for and Slack a “data person” to get it for them.
I see three core problems:
It takes way too long to set up analytics.
Once you set them up, there’s always “one more thing” needed from people that don’t know SQL.
Most analytics tools are bad at ad hoc analysis and exploration.
You deserve a better way to leverage data and it’s definitely Definite:
Definite automates the most tedious parts of analytics with AI.
Anyone in your company can setup analytics workflows using natural language so your data engineers and data scientists can finally have time for deep work.
An ‘infinite canvas’ design space for your team to collaborate on data problems, perform ad hoc analysis and produce presentation-ready results.
It's like Figma was built for analytics and deeply understood your data warehouse.
Connect to your warehouse and get the analytics work you need automatically generated for you with AI. How can we do this? We scan your database schema, query history and dbt models to gain a deep understanding or your data without you lifting a finger. We’re training our AI on commonly used datasets like Stripe, Salesforce, Hubspot and Shopify so we understand what the data means and can output meaningful results. We’re leveraging the countless hours of work you’ve poured into your data stack to make analytics a breeze.
You also shouldn’t need to visit a dashboard or set up Zapier to get live data from your data warehouse. Definite has Slack alerts built-in that you can toggle on with zero overhead.
#2 ChatGPT for data
Have you tried using ChatGPT to write SQL for you? It’s not bad. But you need to give it so much context that you’re probably better off just writing the SQL yourself. Don’t know SQL? You’re out of luck.
At Definite, we’ve become prompt whisperers. We know just what to tell to the AI to get it to do what you want. We’re using your teams query history and database schema to inform the AI of precisely what it needs to get work done. The data nerd in me is insanely excited to see where we can take this.
#3 A drafting table for data
Most software gives me claustrophobia. Notebooks, SQL IDE’s and BI tools are no exception. I always felt cramped when doing analytics work.
When I first used Figma, the first thing I thought was, “how the hell are they doing this in the browser?” The second thing was, “I wish this was connected to Postgres.”
Definites’ Canvas is like a drafting table for your data. See your schema, the results of that gnarly CTE, your final query and the visualization all in one surface. We use DuckDB to keep a temporary in-memory database of your current Canvas which lets you work across databases (e.g. joining Postgres data to Snowflake). You can also query parquet files on S3, CSVs or Google Sheets directly without importing them to your data warehouse.
If your team wants to be data-driven, but you don’t have the time it takes to set up the analytics you need, let’s talk. I think you’ll love what we’re up to here. The team (Steven, Michael, Alex, Stefan, Kai, Jichao, Jess and Kevin) is incredible and shipping ridiculously fast.
Sign up below and find us on Twitter here.