Hi all! I’m a marketing intern in Devices who has been set a project to try create a model which can help calculate sales uplift. I am just learning how to use QuickSight now and am wondering if anyone can help me work out if QuickSight is really the right tool to build this model! I’m looking to use QuickSight over Excel because the amount of sales data will exceed the Excel row cap.
My big question is can QuickSight do ‘what-if analysis’? At the moment I can only really see that it visualises data, but there isn’t space to ‘alter’ inputs on the dashboard, and then see how that changes the numbers. Eg. I would want to ‘increase XX fixture in store’ and then see how much predicted sales would change by. Not sure if this is possible, but could I connect the QuickSight to a Quip spreadsheet and then change the numbers that way?
^In that case, is it possible to combine datasets as well? There is an existing marketing sales dataset, but I’d also need data like number of fixtures in store, fixture → sales uplift etc. so is there a way I can combine those two together in QuickSight or do I need to make a copy of the original data somehow and make my own new dataset?
Welcome to the QuickSight Community!! Interesting post with multiple doubts!
Let me try to help out with a few answers.
Parameters is one way you can consider allowing user input in dashboard. Just note currently you cannot store your input anywhere for future reference but you can alter inputs and use them in your calculations.
You can use these parameters in calculations and perform analysis, creating relevant calculated fields using various functions available.
However there is no way to connect a Quip to QuickSight for now. If you want to input a set of values or tables, you can ingest a .csv file and use it as a dataset.
You can combine datasets by joining them through a foreign key or a related field. This can be done within QuickSight Dataset configuration.
You can also use unrelated datasets, but cannot join them. Instead you can use them within an analysis based on your use case.
Hope this clears few of your doubts. This is a great workshop that you can look into for answers to your how-tos considering you are starting to explore.
Hi @Brett , thank you for your message!
My questions were answered by @prantika_sinha - thank you so much Prantika!
Super helpful, will be sure to ask any more questions I have in future on here
Happy for the topic to be closed, have a great rest of the week all
