This webinar discussed the purpose of data analytics and how it can be a light in the darkness for your organization to make better decisions for the future. The webinar covered the purpose of data analysis and its definition, the fundamental steps to take to perform data analysis to problem solve, and closed with next steps that attendees can take to further develop data analysis and business intelligence within their organizations.
During this webinar, attendees learned about the following:
- How data analytics functions to help your organization improve.
- The process for using data analytics to solve problems.
- Next steps to take to build data analysis within your organization.
2. Introduction
Samuel BowerCraft, MSIS, CISA
• Senior Manager in the Internal Audit and Management
Consulting Group.
• Security consulting related to financial data, information
systems, and assets.
• Experience with strategic oversight and planning;
management; operations and installation of technical
infrastructure; software; and systems.
• M.S., Information Systems.
• Certified Information Systems Auditor (CISA).
3. Introduction
Janice Snyder, CPA
• Partner.
• Leads the firm’s audit segment.
• Serves for-profit and foreign owned business in food
manufacturing, distribution, and services.
• Serves healthcare, human service organizations, and other
non-profit groups in PA and MD.
8. An Exercise
• What measureable objectives do you have in your organization?
9. Data Analysis - Defined
Process of inspecting, cleansing, transforming, and modeling data.
With the goal of:
- Discovering useful information,
- Suggesting conclusions, and
- Supporting decision making.
https://en.wikipedia.org/wiki/Data_analysis
10. Data Analysis - Methods
Qualitative
• Interviews
• Focus groups
• Discussion
• Introspection
• Evaluation
Quantitative
• Units
• Prices
• Proportions/percentage
• Rates of change
• Ratios
• Scoring/ranking
11. Data Analysis:
Methods and Considerations
Qualitative
• Subjective
• Exploratory
• Observational
• Flexible/dynamic
• Contextual
• Continuous view
• Semi-structured
Quantitative
• Objective
• Generalizing
• Testing
• Fixed/controlled
• Variables are known
• Pre/post measurement
• Structured
The Open University: 6 Methods of Data Collection and Analysis
14. Strategy
- Vision
- Mission
- Objectives: What are your key metrics?
- Strategies:
- Activities:
Context – Vast Data, Specific Action
15. Strategy
- Vision
- Mission
- Objectives: What are your key metrics?
- Strategies: What are the themes you have to achieve your key
metrics?
- Activities:
Context – Vast Data, Specific Action
16. Strategy
- Vision
- Mission
- Objectives: What are your key metrics?
- Strategies: What are the themes you have to achieve your key
metrics?
- Activities: What activities support your strategies?
Context – Vast Data, Specific Action
17. Data Galore
• More and more info.
• Varied sources.
• Varied formats.
• Uncertain relationships.
• Uncertain usefulness.
Limited Action
• Time limits options.
• Resources limit options.
• Choice limits future options.
• Not all possibilities can be
investigated.
Context – Vast Data, Specific Action
18. The Scientific Method
“I will never need this
when I grow up.”
- Every kid ever.
https://www.sciencebuddies.org/science-fair-projects/science-fair/steps-of-the-scientific-method
19. Data Analysis - Defined
Process of inspecting, cleansing, transforming, and modeling data.
With the goal of:
- Discovering useful information,
- Suggesting conclusions, and
- Supporting decision making. (which leads to action)
https://en.wikipedia.org/wiki/Data_analysis
20. The Scientific Method
1. Question: Choose something you are curious about.
2. Hypothesis: Make and educated guess at your question’s answer.
• Set your expectation regarding the outcome.
• Be aware of your reasoning for your expectation.
3. Experiment: Put your hypothesis to the test.
• You already did this; your business is in motion.
21. The Scientific Method
4. Data: Record the outcome of the experiment and your
observations.
• Financial data.
• Customer satisfaction data.
• Market share.
• Etc.
5. Analyze: Review the results.
• What does the data indicate given your experiment’s activities?
• If you change your activities, how do you think the data will change?
22. The Scientific Method
6. Report: Show the results of your experiment and discuss whether
your hypothesis is correct.
• Did sales increase when you increased marketing? What did the data indicate
over time?
• Did customer satisfaction go up or down after you implemented the rewards
program?
• How did you measure it?
• Does the data show causation or correlation?
24. Data Analysis Process - Goals
A. Achieve the goal of the analysis.
B. Answer the questions that were asked.
C. Create new informational insights to support decision making and
new behaviors.
D. Record relevant information to support conclusions and results.
27. Framework
1. Identify and articulate your goal.
2. Document the purpose of the analysis.
3. Identify the necessary data.
28. Framework
1. Identify and articulate your goal.
2. Document the purpose of the analysis.
3. Identify the necessary data.
4. Evaluate the data available, structure and content.
29. Framework
1. Identify and articulate your goal.
2. Document the purpose of the analysis.
3. Identify the necessary data.
4. Evaluate the data available, structure and content.
5. Document the assumptions made and any potential impact or
influence on your analysis.
30. Framework
1. Identify and articulate your goal.
2. Document the purpose of the analysis.
3. Identify the necessary data.
4. Evaluate the data available, structure and content.
5. Document the assumptions made and any potential impact or
influence on your analysis.
6. Design a model for the analysis / plan of attack.
31. Framework
7. Articulate your expectation with
regard to the analysis for later
comparison to the findings/outcome.
32. Framework
7. Articulate your expectation with
regard to the analysis for later
comparison to the findings/outcome.
8. Prepare the data and evaluate the model/approach.
33. Framework
7. Articulate your expectation with
regard to the analysis for later
comparison to the findings/outcome.
8. Prepare the data and evaluate the model/approach.
9. Perform analysis.
34. Framework
7. Articulate your expectation with
regard to the analysis for later
comparison to the findings/outcome.
8. Prepare the data and evaluate the model/approach.
9. Perform analysis.
10. Report on findings:
• Compare to expectations
• Evaluate differences noted, issues identified, and limitations on
your findings.
• “We found no evidence to suggest that we could not rely on the
results from our analysis.”
35. Framework Benefits
• Structure
• Building from a pre-existing foundation
• Identify vulnerabilities
• Analyze or evaluate the risk associated with that vulnerability.
• Determine appropriate ways to eliminate or control the
vulnerability.
• Efficiency: Cost Savings (time and dollars)
• Effectiveness
• Support
36. Framework Drawbacks
• While structure is good, understanding is better.
• Limitations:
• The framework versus your environment.
• “No battle plan survives contact with the enemy.”
- Helmuth von Moltke the Elder
• Clarity of Responsibility: you and the framework
38. Starting Your Journey
1. Select an area where you want to see improvement.
2. Set your expectations; test your ideas.
3. Get the resources you need: ideas, data, people.
4. Perform the process: goal, data, analysis, conclusions.
5. Consider what you have learned: how can this new perspective
inform your future actions.
6. Act: based on what you have learned.
Ask attendees to submit some of their measureable objectives online.
Mention:
Data driven decisions.
Better information leads to better understanding … leads to better behaviors / action
Mention: Understanding the methods lets us evaluate and decide which method to use at a given time.
Breakfast: qualitative is fine.
Business acquisition, large capital purchase: “feelings” may not be the best indicators.
Modeling, Action, Outcomes, Evaluation (analysis), new Model
Planning, action, planning based on outcomes, new (or same) action.
Model, outcome, comparison to objectives, evaluation (analysis), remodel, repeat.
Mention:
Data driven decisions.
Better information leads to better understanding … leads to better behaviors / action
Question: refer back to your measurable objectives… you are curious about: what makes this better, OR how do I keep this “good”?
Expectation: you think you know why this works a certain way, but you need to validate this.
Beware of reverse validation.
Ideas are different than beliefs.
If you believe, then all your evidence will support your belief.
Hypothesis (does not equal) conclusion.
Experiment: take action.
4. Data: information about what happened and to what extent.
5. Analysis: evaluation of the information.
6. Report: conclusions, support, answers.
Caution: Rationalization is a form of self-deception.
Seven Habits: Begin with the end in mind.
Do you have enough information to analyze?
If not, how can you obtain it? Is it worth the effort.
Cost/benefit analysis. Value proposition.
You will make assumptions.
Communicating them is easier if you write them down.
Defending them is easier if you write them down.
Accepting them is easier if you write them down.
Excel has tons of space, add a tab.
Draw a picture. This helps ensure you aren’t missing large chunks of information or key relationships.
Make a guess about how you think it will turn out.
Test that guess.
This process helps us to better evaluate how well our judgement is working… and helps us to improve it.
Backward rationalization is a trap and slows our improvement.
Talk through the plan.
Project management is a balance of EFFECTIVENESS and efficiency.
We want high quality, fast, but they don’t always go together.
It is not enough to do your best; you must know what to do, and then do your best.