Mistakes in Power BI dashboard design

comentários · 68 Visualizações

Power BI certification course online helps you learn its concepts and prepare for the PL-300 Exam. Become a certified developer with our Microsoft Power BI training.

Not showing variances

Since we're on the subject of variances, I believe that error fifth is most crucial. The most serious flaw of dashboards we see is not displaying variances. The idea of skipping variances isn't acceptable.

Did you remember the sales dashboard we discussed in the beginning of the blog? Have a second look.

There's no comparison with the plan in this dashboard. There's no strategy in this chart, so there's not a comparison to the previous years. While it's possible to see trends here, or observe certain seasonal patterns, you'll don't know if the actual results are positive or not. The most important thing to remember is that there is no method of determining what your company is performing.

Let's examine the next opportunity.

There's actually a form of comparison in this case that is one that includes a estimate, bottom line cost and the actual cost. However, they are shown in separate charts, which makes it difficult to compare these different categories of data. The most significant problem is that they're displayed in pie charts which makes it difficult to comprehend the dimensions of the various categories. It's not a good concept to design your dashboards this way.

This dashboard displays the comparison between sales and sales from the year before or the target sales that the user is able to alter. It's a great idea. But, it is not very effective when it comes to visualisation. When people look at actuals versus budget or actuals with the prior year, they simply place them in the form of a side-by-side chart. That means you'll always see lots of yellow and blue colors. But what you're really looking for is the difference in between them.

Different methods of displaying different ways of showing

People look to various methods to think about this. Consider this as an example:

One method is to utilize those traffic signals which are shown to the left. The result is green and red and possibly yellow. It's pretty but it isn't without its problems. You'll always get a bunch of green, red or yellow bubbles. The question is, which is the most vital in the present? It's possible to apply a simple rule that says that if you have a number that is greater than your budget, it's green . And it's red if it's below the budget. In this scenario the category could be labeled red if only half a tenth or a couple of dollars or euros lower than the budget.

This means that you aren't able to see the magnitude of the variance or its significance. It is possible to change to a display in which the dimension of the bar is colored to show the importance of the variance.

Avoid using gauges and odometers since they're utterly overused. They take up a lot of space for a single analysis.

It is possible to employ the half-circle chart to the right, but that occupies a significant amount of space. The issue with this visual is that it doesn't show gaps. In this case we're looking at the gap, and we can see that the actual sales fell below the budget. This gap above the budget line should be the red.

The graph shows that our goal was to sell 200k and should direct your focus to this. Visualisation should help you notice this.

Let's take a look at an native Power BI that is used by people who use frequently.

The issue is that the whole image is either red or green regardless of how much variance. In this case, it's an amount for sales and aggregated showing the year-to date information. The issue is that, when it's an example of a sales-versus-target chart, there's a target hidden within, but isn't evident.

Confusion of page layouts

The confusing layouts of pages are the top six. Take a look at this dashboard which is frequent in the business world.

There's no way to know what to look for. There are too many variables that mean that you cannot effectively communicate. There are some great decisions to make - Interactive elements should be at the top, and KPIs should be placed on the left, and everything else is an unorganized mess. There is no logic in the arrangement and order of charts. The designer wanted to combine all the charts on one page.

I'd like you to get a suggestion for a better dashboard.

The most important rule to follow when creating dashboards is that the people will read in order from the left side to right from top to bottom. This means you should always place the most important visualizations and the most important KPIs to on the left side and on the top. In the example below the KPIs are located at the top left, while details are displayed beneath or on the left.

Particularly, if you're creating for the management level You don't have to incorporate the entire set of charts. Sometimes you may only require brief written notes and that's often absent in reports and dashboards..

Mistake: Scaling your charts

This dashboard has brought me back to the previous error on the dashboard. If you carefully look at this dashboard, you'll notice that we have multiple charts of various sizes: one chart to show EBITDA One chart for EBIT one chart to track free cash flow and a chart of net earnings. What's interesting is that each the charts have been scaled. It means that the highest point of the Y-axis across all charts is identical. This is referred to as smaller multiples and is an example of the idea of scaling that is is almost entirely absent from the majority of Power BI dashboards.

I'm unsure if I've seen any dashboards within Power BI that use proper scaling and multiple charts. In certain situations, it is necessary to sync the axes to increase the quality of your dashboards to a higher level. That's why we selected the absence of scaling as the last item in the list of mistakes.

Scaling is required whenever you create different visuals on a page or dashboard. Power BI always scales them to the highest amount. 

Learn more about the mistakes of Power BI in the complete Power BI syllabus mentioned along with a course that you can join immediately. 

 

 

comentários