One of the main reasons for creating a dataset, particularly an import mode dataset, is to provide a performant data source for reports and dashboards. Although Power BI supports traditional reporting workloads, such as email subscriptions and view-only usage, Power BI empowers users to explore and interact with reports and datasets. The responsiveness of visuals for this self-service workload is largely driven by fundamental data model design decisions, such as the granularity of fact and dimension tables. 

Additional performance factors outside the scope of this chapter include the hardware resources allocated to the dataset, such as with Power BI Premium capacities (v-cores, RAM), the efficiency of the DAX Measures created for the dataset, the design of the Power BI reports that query the dataset, and the volume and timing of queries generated by users. Beyond the DAX measures described in Chapter 4, Developing DAX Measures and Security Roles these other factors are outside the control of the dataset designer and will be addressed in other chapters, such as Chapter 13, Scaling up with Power BI Premium and SSAS.