Creating dashboard without a layout

There should always be a sketch of your dashboard discussed and agreed with your stakeholders before you start actual development. This will save you a ton of time and effort. People normally develop a dashboard then get it reviewed which cause numerous redoes and loss of time and effort.

Too many No. of pages

Avoid adding too many pages on desktop file. We can’t restrict the number of pages your Power BI Desktop file should contain but as a best practice, limit it max to 7 to 10 main pages and around 5 pages for drill down.

Large Number of visuals on One page

Never put everything on one page (beginners do this most of the time). It will make the dashboard ineffective, full of clutter and will only distract the audients. A fair number of visuals on one page is four to six. If you need more, add another tab.

Dark backgrounds and background images

People coming from Excel background tend to do this mistake more often. Even you are creating dashboard in Excel or in Power BI, never ever use dark backgrounds or distracting background images. We are mentioning this based on our years of experience. If you do this, the whole purpose behind creating the data visualization will suffer as it puts heavy cognitive load on your audience.

Don’t use One Power BI file for multiple type of dashboards

Never use one desktop file for multiple type of dashboards such as Sales, Purchase, Production, Financial performance and so on. Always use separate files and create the dashboard in Power BI Service.

Excessive use of Bookmarks

Bookmarks are great in data storytelling, but their excessive use particularly on one page makes it difficult to accommodate future changes and additions so use them with care.

Lack of page navigation / homepage

Your dashboard should have proper page navigation icons and a home page. We recommend using infographics to make it more effective.

Frequently changing the Dashboards

Once the dashboard is final and agreed with your target audience, avoid the temptation to make changes frequently. As a best practice, it should be fixed for certain period and should be reviewed on specific intervals for example on quarterly basis.

Where should dashboards be developed?

You should not develop a dashboard on your local machine if it is intended to be shared with others and a schedule refresh is required. You are not using your system 24/7 so always avoid this. People normally do this mistake and later face issues.

Your IT department, is very helpful in this matter. Ask them to give you access to a virtual machine (VM) via remote desktop. This will give you the numerous benefits.

• The VM is much faster and high in specs than your regular work machine and is available 24/7, so it’s the best choice for development and sharing on the web.
• IT has a frequent data backup policy, so your files will be secure
• The windows login password is required when you setup auto refresh in Power BI service. In most of the organizations it needs to be updated every month for local users and if you do this, all the data gateway connections in Power BI Service must be updated with new password. In case of VM, this can be avoided.
• The data gateway in also installed on the same VM
• If one developer leaves the organization, the successor need not to face issues in working with existing dashboards

Conclusion

To become Business Intelligence expert, you need a combination of technical as well as conceptual skills. Working in Power BI desktop requires a considerable level of such skills. You should always adopt guided learning approach and avoid the common mistakes people do. Remember, the best way to learn, is to learn from other’s experience.