If you want to have different types of Cost (Goods, R&D...) and the total on the same plot then you will likely need to create a summary table in your data model that contains at least 3 columns Year, Cost and Cost Type, where the cost type field contains possible values of Total, Goods, R&D, and Services). If you want to be able to parse to other levels of detail it could contain more columns, like maybe Quarter or Month or Department. Once you have the summary table, then the plot in QlikSense has two dimensions (Cost Type and Year) and 1 measure Cost.
To get the Cost to show as Billions, you will need to divide the measure by 1 Billion and then adjust the number format or label
Chris - The method to create the summary table will depend on your data model. If you data model has a transaction table that looks something like this right now will each cost category as its own field then to create the summary table you would use:
Transaction# Date FY Goods R&D Services zzz 1/1/2016 FY15 $10 $5 0 zzz2 1/1/2015 FY14 0 $100 0 zzz3 1/1/2014 FY13 0 0 $50 zzz4 1/1/2013 FY12 $25 0 $10
sum(Goods) as Cost,
'Goods' as CostType
Group by FY;
You would repeat this for each type (Goods, R&D, Services,etc...). This would link back to the rest of your data model by FY (or you could link by date if you prefer).
If your data model looks more like this with all the costs in one field.
Transaction# Date FY Costs Type zzz 1/1/2016 FY15 $10 Goods zzz 1/1/2016 FY15 $5 R&D zzz2 1/1/2015 FY14 $100 R&D zzz3 1/1/2014 FY3 $50 Services
Type as CostType,
sum(Costs) as Cost
Group by FY, CostType;
You would not need to repeat and it would also link back to your data model by FY.
For more on the group by and other load statements I would suggest this nice write up:
Hope this helps.
First, Create 4 measure with respect to FY12 ...FY15 with the help of set analysis. Drag 4 bar chart and use dimension which has value as ,'Goods','R&D','Services' .
Note. By above statement, you have to make new chart for FY16 or I can say it will not a dynamic visualization. But for now it will solve your purpose.
Please mark the appropriate replies as CORRECT / HELPFUL so our team and other members know that your question(s) has been answered to your satisfaction.