Business Analytics 101: Data Visualization and Storytelling Dashboard Design KPI Selection Layout Drilldown Filters Tooltips

What This Is

Dashboard design is a crucial aspect of business analytics, as it enables stakeholders to make informed decisions by visualizing key performance indicators (KPIs) and trends. A well-designed dashboard should be intuitive, easy to navigate, and provide actionable insights. For instance, a retail company might create a dashboard to monitor sales, customer engagement, and inventory levels, allowing them to identify areas for improvement and optimize their operations.

Key Formulas & Metrics

• Mean Absolute Error (MAE) = (1/n) Σ|yᵢ − ŷᵢ|: The average absolute difference between actual and predicted values, where yᵢ is the actual value, ŷᵢ is the predicted value, and n is the number of observations.
• R² = 1 − (SS_res / SS_tot): The proportion of variance explained by the regression model, where SS_res is the sum of squared residuals and SS_tot is the total sum of squares.
• Coefficient of Variation (CV) = σ / μ: A measure of relative variability, where σ is the standard deviation and μ is the mean.
• F-statistic = (MS_model / MS_res): A test statistic used to determine the significance of the regression model, where MS_model is the mean square of the model and MS_res is the mean square of the residuals.
• Adjusted R² = 1 − ((n-1)/(n-k-1)) * (1-R²): An adjusted measure of the proportion of variance explained by the regression model, where n is the number of observations, k is the number of predictors, and R² is the unadjusted R².
• MAPE = (1/n) Σ|yᵢ − ŷᵢ| / yᵢ: The mean absolute percentage error, where yᵢ is the actual value and ŷᵢ is the predicted value.
• Spearman's Rank Correlation Coefficient (ρ) = 1 - (6 * Σ(dᵢ^2)) / (n^3 - n): A measure of the strength and direction of the relationship between two variables, where dᵢ is the difference between the ranks of the two variables and n is the number of observations.

Step-by-Step Procedure

1. Define the purpose and audience: Identify the stakeholders who will be using the dashboard and determine the key performance indicators (KPIs) that need to be displayed.
2. Choose the layout: Select a layout that is intuitive and easy to navigate, such as a grid or a matrix.
3. Select the KPIs: Choose the KPIs that are most relevant to the business problem and display them in a clear and concise manner.
4. Add filters and drill-down capabilities: Allow users to filter the data and drill down to more detailed information to gain a deeper understanding of the trends and patterns.
5. Use tooltips and hover text: Provide additional information and context to help users understand the data and make informed decisions.
6. Test and iterate: Test the dashboard with stakeholders and iterate on the design and layout to ensure that it meets their needs and is effective in communicating insights.

Common Mistakes

• Mistake: Confusing correlation with causation.
• Correction: Correlation does not imply causation, and it's essential to identify the underlying mechanisms and relationships between variables.
• Mistake: Misinterpreting p-values.
• Correction: P-values are not the probability that the null hypothesis is true, but rather the probability of observing the data (or more extreme) if the null hypothesis is true.
• Mistake: Using the wrong error metric for a business problem.
• Correction: Choose an error metric that is relevant to the business problem, such as mean absolute error (MAE) for forecasting or mean absolute percentage error (MAPE) for inventory management.

Software / Tool Tips

• Tableau: Use the "Show Me" feature to automatically create a dashboard based on the data, and then customize it to meet the needs of the stakeholders.
• Power BI: Use the "Dashboard" feature to create a dashboard that is interactive and easy to navigate, and then add filters and drill-down capabilities to provide more detailed information.
• Python with pandas and scikit-learn: Use the "matplotlib" library to create visualizations and the "seaborn" library to create heatmaps and other types of visualizations.

Quick Practice Problem

A retail company wants to forecast sales for the next quarter. The historical sales data is as follows: 100, 110, 105, 120, 115. What is the moving average forecast for period 4?

Answer: 115.5 Explanation: The moving average forecast is calculated by taking the average of the last few periods. In this case, the moving average forecast for period 4 is the average of the last 3 periods: (110 + 105 + 120) / 3 = 115.

Last-Minute Cram Sheet

• Dashboard design is crucial for business analytics: A well-designed dashboard should be intuitive, easy to navigate, and provide actionable insights.
• MAE is a measure of absolute error: The mean absolute error (MAE) is a measure of the average absolute difference between actual and predicted values.
• R² is a measure of variance explained: The R² statistic measures the proportion of variance explained by the regression model.
• CV is a measure of relative variability: The coefficient of variation (CV) is a measure of relative variability, where σ is the standard deviation and μ is the mean.
• F-statistic is a test statistic for regression: The F-statistic is a test statistic used to determine the significance of the regression model.
• Adjusted R² is an adjusted measure of variance explained: The adjusted R² statistic is an adjusted measure of the proportion of variance explained by the regression model.
• MAPE is a measure of absolute percentage error: The mean absolute percentage error (MAPE) is a measure of the average absolute percentage difference between actual and predicted values.
• Spearman's rank correlation coefficient is a measure of relationship strength: Spearman's rank correlation coefficient (ρ) is a measure of the strength and direction of the relationship between two variables.
• Correlation does not imply causation: Correlation does not imply causation, and it's essential to identify the underlying mechanisms and relationships between variables.
• P-values are not the probability that the null hypothesis is true: P-values are not the probability that the null hypothesis is true, but rather the probability of observing the data (or more extreme) if the null hypothesis is true.