One of the ways to improve the responses you get from InsightBase is to ask the right way, this is covered in the Crafting Perfect Queries in InsightBase article. It's a good way to get you started.

However, there are more things you can do to improve the responses even further. The concepts we'll cover in this article could be a little bit technical, and if you need some help then you can always reach out to us, or to a technical person from your organization.

It is possible that the question you're asking is not very clear, and if it's not formulated properly then the system could have problems detecting the right table to start with. This is why specifying the table name could help a lot, especially when the question you are asking is a little bit ambiguous.

For instance, you might have a "users" table for the users who created an account on our website, as well as an "accounts" table that holds user account information, including whether the user is active or not. In this case, some of the questions like "How many active customers do we have" could be ambiguous in the sense that the system doesn't know what a "customer" is - is it a "user" or an "account"? Yes, this is clear for you, but not for the system. In this case, you could specify the table name the system should use, like "How many active customers do we have (use accounts table)?".

If the system returns more than you want, you can always specify the fields you want to get back. For instance, if you want to see a list of the users that spent the most on your app, but you only want to see the user's email and country, just say "Show me the users that spent the most (return only the email and country)".

Check the Data Source Schema article for more details about providing descriptions. This single step could have a huge impact on the responses you get.