Similar to talking to ChatGPT, Rasgo's AI understands some questions better than others. Here's a guide to craft the best possible question prompts for your data question.
Rasgo's AI is using your question to study its notes, pick the right table or tables, and then craft an accurate SQL query. That's a lot of steps! The more detail you can share in your prompt, the better AI will do at navigating that task. Details that can be helpful to include:
- How should AI aggregate the data, if at all (i.e. Total, Average, Max, etc...)
- Do you have an idea of what table or tables AI should look at? Even a guess can be helpful (i.e. "Use our Master Sales table")
- Want your query to apply filters to the data? Include both the column and the values in that column you want to filter, if possible (i.e. "... for customer Acme Corp", where
customeris the column and Acme Corp is the value in the
customercolumn that you want to filter to)
Just like a 5 year old, you don't want to give AI the wrong ideas... It's not very good at discerning the "negative" instruction (i.e. "Don't use column ABC"). Instead, focus on telling AI what to do whenever possible (i.e. "Use a column starting with DEF").
If you're not sure where to start, this is a good template that very reliably translates to SQL:
Sum <Column A> and <Column B> by <Column C> for the current time period using <Date Column D>
Generative AI can generate some impressive SQL but it never hurts to start with the basics.
- Rasgo AI can only process feedback on the SQL it generates today. This means that if you ask a follow up question about the Data Visualization, Headline, or natural language interpretation sections, Rasgo's AI will not know what to do.
- Rasgo AI, like all generative AI, is token limited. If you ask a question that generates a particularly large query, Rasgo won't be able to intelligently interpret the output data, and you will only get back a Visualization and a SQL Query Summary.
- Rasgo AI has to search through your data warehouse metadata to find the right table to use. That means if there is no human documentation of the tables and columns in those tables, all the AI has to go off of is the column names and table name when deciding whether or not to pick that table.