Combined Sets to Analyze data

On last week project we had to analyze a fake brand data. I needed to see some insights about which products were benefiting the most by paying social media influencers to endorse their products.

The data I was using had a lot of fields but the only ones I needed where these .

Creating Parameters

I Created these 2 parameters to be able to allow the Dashboard user to create their own view. The Top N Influencer Pay Parameter will allow users to adjust how many results they would like to see in their set by Sum of “Influencer Pay”. The other set will allow the same but for the amount of Sales.

Creating Sets

Then we need to create sets to see which products meet my criteria. In this case I needed to see the Top N Products Names by Influencer pay Sum.

To do this Right click on Product Names and click on Create> Set. Go to the Top tab and choose the parameter you already created for top N influencer pay. Then select the measure and aggregation method, in this case Influencer Pay and Sum.

Now create another set from Product Names but this time select your other parameter for top n sum of sales. Also change the measure and aggregation method to Sum of sales. Now that we have both sets created we will be able to combine them.

Combining the Sets

Our goal is to see which Product names are included in both criteria selecting the Top N we want. To do this we need to create a Combined Set. Ctrl +click both sets and right click to create a combined Set. Once the Combined set window comes up we have to give it a name, and select the “Join” type. Combined Sets work as a join so we can manipulate what the outcome will be like. In this case we need only “Shared members in both sets”, to exclude all the product name that don’t meet both our criteria.

Creating the chart

To create the chart, we need to drag our combined set onto the rows and also the Product Names. Then we can drag the Profit onto the columns. This will allow us to see which Product Names that have the highest sales and highest influencer pay make most profit.

Another way to analyze this data would have been to choose Bottom N by Influencer pay in our first set to see which products where benefiting most from influencers that get paid less.