HR Consultation Insights - Power BI Challenge #3 Entry from Luke

Here’s @Luke entry for Power BI Challenge 3. @Luke, would you like to share how you built this dashboard and what your inspiration is in building it?

Thanks for your submission Luke and great to know that you are developing your skill set well with Power BI.

I definitely think that this report has most of the information that you would need to understand the data well.

The only thing I think I would improve here is the look and feel of the report. Some of the visuals I think could be represented in a slightly more creative way.

But I’m definitely confident this will come in time when you have the opportunity to utilize all of the great resources that we have available with enterprise DNA.

Also don’t forget to use the color theme generator to generate a consistent theme across all of your reports in visuals.

Also some report design inspiration here

https://info.enterprisedna.co/report-design-inspiration/

Talk us through your thoughts process here with your development?

Sam

Thanks for the feedback @sam.mckay ,

I modelled the data into 4 dimension tables (employees, consultations, interviewers and date) and one fact (interviews).

The dims were simple however there was additional modelling for the fact table. It was modelled so that there was just one interview per row

Enjoyed the modelling part of the project and adhering to best practice as much as possible here.

I didn’t use many measures at all just some simple counts and filters.

The design I have tried to keep clean and uncluttered whilst still showing all of the necessary insights and including all of the relevant data and breakdowns by location and job role etc.

The top section of the report shows data across both interviews and offers comparisons and the lower section is filterable by interview number and allows you to drill into the data.

I would have liked something more advanced for the sentiment chart or a more exciting way to display it but i was falling short with my DAX and decided to settle on simply creating a score that is count of happy - count of unhappy.

Cheers!