Hello ,I am looking to build a data story using ‘Q’ and want Q to assimilate both data and business inputs around that data so that it generates a summary of what caused the variances on particular metrics and how these metrics contributed to the main metric. I have created multiple charts explaining the relationship between the metrics and provided text inputs for business context, but the output from Q is not able to combine data and business inputs and provide a meaningful summary. How do I make Q understand from multiple lines of text and correlate it to the data and charts provided separately? I have tried providing inputs through prompt but Q still doesn’t generate insights through them
Hi @sacagarw ,
Are you using Data Stories or the Executive Summary? I have found that Data Stories is more tunable and will generate better text by changing the visuals that I give it. Also, it does generate different responses based on your build prompt.
That said, there are a couple points to note from the docs (see here); 1) it’s still in preview and not recommended for production usage, yet, and 2) It’s a first draft, so you’ll still have to tweak the text to be perfect for your audience.
Hope that helps.
ws
Many thanks @wstevens01 ! I am using Data Stories and have tweaked my charts learning from the output generated in each iteration and even changed the prompts to force Q to glean the business inputs and relate it with the charts provided but to no use. I have provided the business inputs as a dataset also and built topic on top of it trying to keep the literature similar to the data but it seems that my visuals are not able to provide the data that the prompts are looking for.
Hello @sacagarw, were you able to make any progress on this or get relevant information from the data story builder? I am curious to see if you were able to get it to work, and if so how you were able to accomplish that.
Hello @sacagarw, since we have not heard back from you, I will archive this topic. If you still require assistance with this issue, please post a new question in the community and link this topic to provide relevant information. That will ensure you are at the top of the priority list to receive a response from one of our QuickSight experts. Thank you!