Business Intelligence Notes PDF

Preview

Summary

Chapter 1 notes - Power BI...

Description

Week 1 Notes - Prework ● Business Intelligence: (KNOWN) broad categories of applications and technologies for gathering, storing, and analyzing to help organizations make better business decisions and manage business performance ○ Example: how do this year's sales compare to last years sales ● Advanced Analytics: (UNKNOWN) sophisticated techniques and tools to discover deeper insights, make predictions, or generate recommendations (might not see this on our own) ● All known values are known as KPIs ● KPI: a metric that demonstrates how effectively a business is achieving a business objective (using dashboards) ● Analytics Maturity Curve ○ Starts with raw data and increases productivity ● Components ○ Data sourcing - where does data come from ○ Data staging - data transformation, combining data from multiple sources ○ Data analysis/ discovery ○ Information delivery - where dashboards come into play ● Types of Dashboards ○ Strategic - quick overview on the health of the business, reports on high level measures such as KPIs, little information, used by executives, static snapshots ○ Analytical - used by analysts, drill down data, complex data, sophisticated display, statistical snapshots ○ Operational - used to monitor operations, alerts for critical events, detailed data as well as simple display mechanisms, real time data ● Design is important in dashboards - need to draw attention to the viewer ○ Need to reduce the noise ● ARTICLE: What is the Difference between Business Intelligence and Analytics? ○ Both are connected in their approach - focused on resolving business issues and providing insights on past and present data and defining future decisions ○ BI is more focused on the present moment of data, making the decision based on current insights ○ Business Intelligence: deals with what happened in the past and how it happened leading up to the present moment. It identifies big trends and patterns without digging too much into the why’s or predicting the future ○ Business Analytics: deals with the why’s of what happened in the past. It breaks down contributing factors and causality. It also uses these why’s to make predictions of what will happen in the future ○ Seen in football - want to review most recent game ■ BI - would identify statistics and plays that led your team to winning. Identifies the trend that your right side of the field was instrumental in retaining possession through excellent passing

■

●

●

●

BA - would be concerned with why you had possession of ball for longer than your opponent and why your right side of the field did so well at passing ■ Asking good BI questions will lead to better BA ○ Causation vs. Correlation ■ Correlation - when one thing happens, the other tends to happen at the same time ■ Causation - one thing leads directly or indirectly to the other happening ■ Applies to business by understanding factors that are causing your business to fail rather than just the factors that are associated with your business success or failure. ● Need to know what happened and how (BI) before you can say why it happened (BA) ○ Use case scenarios ■ To figure out what products would be most likely to succeed (BA) you need to figure out, ● What products have been most successful in the past (BI) ● The seasonal trends that had influenced success for past launches (BI) ● Why customers bought the past successful products (BA) ○ Important in HR roles, Procurement (is it possible to outperform my supply delivery process?), and sales (how to decrease the sales cycle length?) The importance of KPIs - airplane scenario ○ Pilots need to focus on many factors, not just one - this also applies to businesses ○ It is important for organizations to: identify key measures, organize them, and to design an evaluation system The importance of measuring performance - faculty evaluation ○ Research, teaching, citizenship/service ○ Need to look at quality and quantity Power BI ○ Allows to create, collaborate, and share ○ Start with power BI desktop - can generate reports, and access data sources to make visuals ○ Power BI services to share reports with others

Introduction: ● Dashboard: is a visual display of the most important information needed to achieve one or more objectives; consolidated and arranged on a single screen so the information can be monitored at a glance. ○ objectives : KPI given by a balanced scorecard approach ○ Timely ○ Customizable ○ Situational awareness

●

● ●

●

○ Clarity of goal ○ Response process - who and how are they responding Design principle: provide the greatest degree of understanding with the least amount of effort ○ Dashboard must be truthful and functional, but there are reasons you should go the extra mile to make dashboards that are elegant as well ○ Remove ink - make as simple as possible Data must have context - make sure units are clear and place them into context (what direction are they going, how does it relate to last year, etc.) Few’s ideas on dashboard design ○ Lack of context for data ○ Don't use too much detail or precision (don't need decimals) ○ Express measures directly ○ Use right display media ○ Don't use too much variety - use only a few types of charts ○ Watch for poor scaling ○ Poor arrangement ○ Poor highlighting ○ Cluttered display ○ Misuse of color - start with greyscales ○ Unattractive display - will use readers attention, want to be seen as professional Functionality of Power BI ○ Start with data, then build visuals, put them on report page, pin to dashboard know hierarchy

The Big Book of Dashboards: Chapter 1 (Data Visualization) ● Preattentive attributes: things our brains process in milliseconds before we pay attention to everything else ● Color, size and hue ● Types of Data: ○ Categorical: represents things and are mutually exclusive labels without any numerical values (examples: gender, city, etc.) ○ Ordinal: similar to categorical except has clear order (examples: educational experience, salary bands, etc) ○ Quantitative: is the numbers, can be measured and aggregated (exam scores, profit, sales, etc.) ■ Discrete or continuous ● Discrete - predefined, exact points ● Continuous - the inbetween ● Color should be used to draw attention to something important ○ sequential , diverting, categorical, highlight, alert ● Common chart types: ○ Bar chart - uses length for measure, best used to spot biggest or smallest items ○ Time series - show change over time

○ ○ ○ ○

Scatterplot - compares two different measures, useful to show relationship between variables Dot plot - allows you to compare values across two dimensions Chloropleth - uses differences in shading or coloring within predefined areas to indicate the values or categories Symbol map - shows values in specific places

Chapter 33: The Allure of Red and Green ● Colorblind friendly visualization: provide option to swap the colros ○ Use light and dark colors to distinguish between red and green ○ Leverage blue in the green colors Chapter 34: The Allure of Pies and Donuts 13 Common Mistakes in Dashboard Design 1. Exceeding boundaries of a single screen 2. Supplying inadequate context for the data 3. Displaying excessive detail or precision - ex. Decimal places 4. Expressing measures indirectly 5. Choosing inappropriate display media 6. Introducing meaningless variety 7. Using poorly designed display media 8. Encoding quantitative data inaccurately 9. Arranging information poorly 10. Highlighting important information ineffectively or not at all 11. Cluttering the display with visual effects 12. Misusing or overusing color 13. Designing an unattractive visual display Learning Data Visualization: ● Figure ground - used in business logos ○ Bottom vs. top ● Proximity - groupings ○ How we recognize patterns ● Similarity - allows to categorize ● Parallelism - when you have things in parallel, we assume that they are together ● Common fate - can see in moving diagrams ● Closure and continuity ○ Closure - see things that aren’t there ● How to show differentiation ○ Position ○ Size ○ Color ○ Contrast

●

○ Shape The Importance of Scale ○ Chart type ■ Can change numbers on axis (not beginning of 0) ○ External Reference ○ Avoid Bias...