It may be worth exploring other tools if that's your end goal (R, Excel, the Sand Dance custom visual).Īs other posters have mentioned, data types are important. A scatter chart with 100,000 data points might not perform as well as a traditional graphing tool. A column chart with 10,000 columns is going to be pretty unusable. In my opinion, it's good that Power BI puts the onus on you to add that context as a column to your data.īear in mind, with some of my de-aggregating suggestions, that if you're working with large datasets, Power BI isn't going to perform all that great. Even if each row represents a single customer or a single month, it makes more sense to provide the customer ID or month name in your chart, than for Power BI to use a meaningless row number. For that reason, charts need to provide context (such as what a column represents). Power BI is designed as an interactive data analysis tool. Should Power BI have a concept of "row" when displaying data, so you can de-aggregate data to the lowest level without having to manually add your own column? I personally don't think so. This could be as simple as going to "Edit Query" and adding an Index column (though I'd recommend adding something that makes sense to your problem domain). In any of these cases, if you don't have a field to break your chart down to the level you want, you'll need to add one. Instead, you want to put the field you want to break your chart down by into the "Details" or "Legend" section. If you know that you have only 1 value per period (say), then using an aggregation such as average or sum shouldn't matter because an average/sum of a single value is that value (as long as you're breaking down your chart by period).įor a scatter chart, the same principals apply but in this case, there is no axis. If, as in the case of the original poster, you want to break down your actual & goal by period, then period needs to be on the axis. There are a few ways around this. If you want to break your column chart down to have a single column for every sales record, then you need sales record ID (or another attribute that's unique for every record) on your axis. Even if you know there's only one data point per month, Power BI doesn't know that will always be true. If you have "Month" on the axis and then you put "Sales" as the value, what do you expect Power BI to do when there is more than one sale in a given month? It needs to know if it should sum, average, count, etc. If you think about it, this does make sense. You can change that aggregation (sum, average, count, min, max, etc) but you cannot select "Do Not Summarize". The exception (I know of) is the table visual.įor that reason, if you put a column into the "Values" section of a chart, Power BI will select an aggregation for you. Most visuals in Power BI expect an aggregated value in the "Values" section of the chart.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |