Forum Discussion
Conditional formatting for "version control"
Hi Community!
I need a little help with something that in mind seems rather easy but I am failing to actually get done. As the tittle say I want to use some conditional formating on a table ( or a matrix if possible) to do some version control. Let me give some insights on the type of data and information im handling and the potential expected results:
I can retrieve ( manually for now) an excel file with a list of sw versions for multiple devices from my companys customer.
We have as a company an online repository of all the latest sw releases for every device we have. This is a Smartsheet list.
I have this PoweBI report done were I query the excel file( currently uploaded to a sharepoint) and the Smartsheet ( using the smartsheet connector). All these work fine and I can transform and present some neat tables.
I want to have a table with ( among others) a column from the excel "customer" info and another with the smartsheet "latest released" and compare them, if the value in the customer cell is not the same as the lastest release then it should be higlighted in red (for example).
I know I can solve this in powerquery editor by nesting the 2 tables with the "device" values and do the transformations there and create a custom column holding something like "is different" and then if the values are "yes" highlight them.
I just think this is overcomplicating the things and it should be easier to do with just the plain data in the tables and conditional formating.
Can i get some help or ideas for this?
Many thanks!
3 Replies
I think you'll have to flag them in some way (much like you mentioned) for the effect you're after.
Yes, you can do it with conditional formating. But its very similar if you nesting both tables or use a a relationship between the tables, in the end you have to create a measure which compare both values and you can create a rule for it in the conditional formating or you create with the measure the color code for the conditional formatting. So you have different, simlar ways 🙂
AutoJL If it is a small data, merging tables in PQ can do the job but I will avoid it because it is a slow transformation but rather focus on creating a data model that can achieve the goal. One of the advantages of this approach will be that the solution can be scaled tomorrow to meet other requirements.
I agree with andhiii079845 approach. Basically, you need a common dimension between these two tables, connect both to this new dimension table, and then do the comparison for conditional formatting
