Different datasets are explored at different levels of detail depending on the questions to be asked. Answering these questions requires working with different levels of data granularity. Over the next few minutes, you'll explore the concept of data granularity and discover how it can help inform your data analysis. Over at Adventure Works, the company needs sales data to help make strategic decisions about what products to stock. It must identify the highest and least performing products using annual and daily sales data. You can help the company generate these insights by using data granularity to analyze its sales records. Let's begin by recapping what is meant by the term data granularity. As you might recall, data granularity refers to the level of detail or depth captured in a certain dataset or data field. Granular data provides deeper and more precise insights. This delivers more nuanced and valuable findings. Remember, data granularity isn't about always having the highest level of detail. It's about having the appropriate level of detail. Before you begin your analysis, ask yourself, do you require high granularity or low granularity? The decision should depend on the specific requirements and objectives of the analysis. It's about striking the right balance between detail, manageability, precision, and simplicity. High granularity data is the dataset that records very detailed information about each transaction. This level of granularity provides a comprehensive overview of each transaction including specific attributes and metrics associated with the transaction. Let's look at an example from Adventure Works database. For instance, in Adventure Works data analysis, product related data can be captured as product ID, category, subcategory, name, price, size, and weight. Some benefits of high granularity include in depth exploration of trends, patterns, and relationships within datasets to identify specific behaviors and anomalies. The flexibility to aggregate and summarize data at various levels of detail, and the ability to facilitate accurate decision making by drilling down into specific data points. Next, let's look at low granularity. In low granularity data, information is captured and analyzed at a high level summary or an aggregated level. The data is not broken down into individual records, instead, data is summarized over broader categories or periods. Here's an example from the Adventure Works database. For example, Adventure Works can explore its sales quarter by business quarter or month. The benefits of low granularity include a simplified view that's easy to understand and allows for analysis without an overwhelming level of detail. Improved performance and reduced data volume, which leads to faster query execution and a quick identification of trends and patterns for informed decision making. Let's take a closer look at data granularity and its role in data analysis.In the context of data analysis, high granularity data is often more desirable. It offers a finer level of detail, so it provides greater precision and potential for deeper insights. For instance, tracking sales hourly, high granularity, instead of monthly, low granularity, could reveal patterns like peak shopping hours during the day. However, working with high granularity data comes with its challenges. The more granular your data, the larger your datasets will be, potentially slowing down data processing and analysis. On the other hand, low granularity data, while offering less detail, can provide a broader view of your data. It's also easier to manage because of the smaller datasets. In Adventure Works, the monthly sales data, low granularity, could help identify broader trends, such as seasonal sales fluctuations of certain product lines. For example, bicycle repair equipment sells more during the spring and summer months. This is because customers are more on their bicycles. You can ensure the relationships are accurate and produce consistent aggregations by matching the granularity levels. It also helps with correct filtering and supports drill down analysis. Data Granularity also has a significant impact on building relationships between tables in Power BI. For example, to determine the highest and lowest selling products in the adventure works inventory, you must produce reports of total sales and budget over time using the sales and budget data. The sales data is in the sales table and has daily level granularity. On the other hand, the budget data is stored in the budget table and is monthly. To establish the relationship between tables and produce accurate results, you need to format the date table in both tables and then build a relationship based on a commonly formatted date column. Understanding and manipulating data granularity is a powerful skill that all data analysts must master. The degree of granularity can impact the insights drawn and the ease with which data can be analyzed. With a firm understanding of data granularity, you can now approach your data analysis tasks with a refined perspective. It's time to discover the story that the right level of detail in your data can tell.
en