We launched our Regression Chart recently. It hit Microsoft’s AppSource on last Friday. The visual had been cooking in our kettles for some time, so it is great to see it finally reach the users.
Now, as it is gaining users it is a good time to look in more detail why we are so excited about it.
1 — Dynamic loading of data
Firstly, the visual contains an element that we call dynamic data loading. It means that data is loaded and shown to the user in multiple chunks. So instead of traditional monolith read-handle-draw sequence, we might do a dozen of them, depending on dataset size.
What is so great about this? It forces the visual to be responsive.
Power BI is an excellent platform. By far the best, in my opinion. And I’m not alone in this view. If you don’t believe, ask Gartner.
But despite being the best, Power BI is not perfect. One of the larger problems that it’s suffering at the moment, is the relative slowness and lack responsiveness of the reports.
Microsoft is acting on this, and they have promised better tooling support for report designers. Tools should arrive in April.
Tooling support is, however, not enough. Profiling tool will allow report designers to identify problem visuals. But what there is to do? You can remove the badly behaving visual, you can reduce data, or you can reduce the number of visuals on the report. All of these will leave user and report readers high and dry.
Think about it.
You are either missing data or missing wanted visualizations.
We think the answer is dynamic data loading, and we say that every visual should enable this. It is the only way to allow large datasets and multiple visuals while ensuring responsiveness.
2 — Small multiples
Regression chart includes small multiples or faceting. This is such a great way to slice, dice, and report data that we wrote own blog post about it.
3 — Fast render mode
The visual also includes something we call fast rendering mode. It only works in modern browsers, say Chrome, and Power BI Desktop, but it allows real fast rendering of 10k+ dataset sized.
It uses completely rewritten data visualization, and it does not break under heavy load, as the regularly used visual tech does. You have to see it to believe it.
Everything in this world has pros and cons, as does the fast rendering mode. The result visualization does not look as crisp on some displays than the traditional tech, but if you need to visualize larger dataset, this is usually an acceptable trade-off.
As with any new technique, fast rendering has had few glitches. These relate mostly to differences in browsers. Gladly, our users report these errors, and we’ve been able to iron them out.
4 — Step towards statistical analysis
Most of the Power BI visuals are aimed toward business intelligence and management reporting. Craydec Regression Chart calculates the linear regression for the dataset. This is a step towards statistical analysis of data. This is something we would really like to see more in Power BI community.
Power BI is such a great data platform, and there is is no need to focus it just on business intelligence.
5 — First step of the movement
Regression Chart is our first published visual. We are on a mission to bring the very best Power BI visuals to power users. Make sure that you’ll check this one out, as there are many more to come.
But we can’t make the best visual alone. We need your input. So if you have any comments, you can mail them to support(at-the)craydec.com.
Like what you’re reading?
Want to read more? Subscribe here.
Originally published at www.craydec.com on February 27, 2019.