With Power BI Desktop’s July update now out for a few weeks, I wanted to write a quick review of the new features, focusing on my Top 3 for this month’s update.
1. Visual Headers
Having the ability to completely remove the hovering visual headers from each visual, in addition to moving it inside the visual greatly enhances the report aesthetics. This is a great way to have items such as Images, Report Titles, Last Refresh Dates, etc., appear in the report without distracting the user when they move the cursor over them as they try to interact with another visual.
Moving the header to be in-line with the title also helps keep the report tidy while still providing all the features available (which can also be individually controlled).
Please note that to reflect this on existing reports, you will need to enable it in the Options dialog under the Report Settings section.
2. Wallpaper Formatting
This is another great feature that enhances the report aesthetics and enables some new design combinations when authoring a report. Instead of being stuck with a light-grey frame around our reports, we now have full control over that area, giving us the ability to change the color, transparency, and image.
This is great for reports that are embedded, published to web, or that have a page size that doesn’t take up the majority of the screen.
With the new Theming update, you can also set the wallpaper across all your report pages.
3. Composite Models
While still in Preview mode, I think the capabilities that are being enabled with this feature really open up the door for some new data modeling options, data storage considerations, and an overall improved querying performance in Power BI.
Having the ability to combine DirectQuery and Import queries removes the limitation of having to choose between the two as one of the first things when authoring your report. I would still like to see some improvements here to support, for instance, the ability to connect to SSAS Tabular and combine that with an Imported dataset.
The many-to-many relationships will make it easier to join datasets where neither side has a unique list of values to join on. While there are scenarios where this may make sense to be used, if you have a dimensional modeling mindset like me, you may be scratching your head a bit with this one. It will be interesting to see how this capability evolves over time.
Lastly, the storage mode is a game-changer when working with really large datasets. I would highly encourage you to watch this session from the Business Applications Summit by Christian Wade, where he shows how to configure a data model in Power BI to support querying 1 trillion rows.
There are many other features in this month’s release, if you want to check them all out, make sure to visit the Power BI blog here.
