I collect data daily via URL download to csv (standard tabular format). I use Excel Power Query to transform and summarise this data (groupby row count by column header name) so that it is crunched from rows in excess of 100,000 down to hundreds (Table 1). I then create a second Excel Power Query which adds the contents of Table 1 to an historic version of the same data (Table 2). Finally, I use another Power Query, this time in Power BI, to draw the contents of Table 2 to Power BI where I visualise and share the summarised, historic data with stakeholders. Clearly this is a very manual exercise, but I have yet to find an alternative daily update methodology which simplifies the entire process (in particular, the historic update with daily data). Ideally, I need a process which operates automatically, daily, directly from source to Power BI to stakeholder (ie without needing to do the transformation in Excel)
Any advice or alternative courses of action much appreciated.
Thanks.
PS Recently I have connected the external URL above to MS Dataverse (so that every new line item in the source data is automatically collected real time). MS Dataverse can be connected directly to Power BI (but not to Excel as far as I can see).
Solved! Go to Solution.
Power BI has no memory. You need to do that aggregation and merging in the upstream system.
Hi @seancody ,
Could you tell me if your problem has been solved?
If it is, kindly Accept it as the solution. More people will benefit from it.
Or you are still confused about it, please provide me with more details about your problem.
Best Regards,
Stephen Tao
Hi @seancody ,
It sounds like you need to configure scheduled refresh, please kindly refer to Configure scheduled refresh - Power BI | Microsoft Docs.
In addition, you can clean and transform data directly in Power Query of Power BI. There is no limitation on the rows of data you can import into Power BI currently.
Best Regards,
Stephen Tao
If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.
Power BI has no memory. You need to do that aggregation and merging in the upstream system.