What Research?

Business Analytics: Provide actionable insights to optimize strategies and programs

Identity and define issue:

• What happened?
• How or why did it happen?
• What is happening now?
• What is likely to happen next?

Research:

• Identify data sources
• Gather data (Organizing and group similar material, capture and evaluate source information, make notes with the assumption you are not the only person reading the data gathered.)
• Explore data (Identify relevant variables. What elements are the same or different? Identify patterns, note when patterns deviate, determine gaps or process differences. Is it trending up or down?)
• Qualify data (Ensure data integrity, clean data, reconcile data, model scenarios, and validate assumptions)

Report /Communicate

• State historical performance, benchmarking (comparing to industry standards, best practices, or performance metrics) and summarize results.
• Summarize conclusions based on insights and recommended actionable strategies.  

Research Methods

Qualitative research is used to understand why customers behave and develop hypotheses about that behavior. To understand reasons, opinions, and motivations. Personal interviews and focus groups (a group of 8-12 carefully selected people held in a neutral location). Data collection from participant observation and analyzed by themes.

 Quantitative research is a very structured form that attempts to answer how much. Numbers can be projected of the sample represents to uncover patterns. Telephone, online and mail surveys, polls, or questionnaires are examples of this type of research. Data collection through measuring things, numerical comparison and statistical inferences.

Analyzing Data

Descriptive: What has happened?

Predicative: What could happen?

Prescriptive: What should happen?

Descriptive statistics. It is the summary of collected data points. These are the models that will help you understand what happened and why. Past performance review, identify changes, note similarities and patterns, successes and failures (i.e. counting population and reporting census breakdown) Does not make a conclusion or hypotheses, only describes data.

• Measures tendency (often the average) valid measurement are mean, median and mode
• Measures spread (standard deviation) a spread within a data set

 Inferential statistics. It is a techniques that allows use of data samples to make generalizations about data. Sampling strategy selects random or systematic sample data instead of all data to qualify findings.

 Predictive analytics. It is to predict customer behavior and trends based on historical data to evaluate probability of future behavior. Predictive models exploit patterns found in historical data to identify risks and opportunities to determine probability. (i.e. Credit Scoring, high score more reliable, low score more risk)

 Prescriptive analytics. It attempts to anticipate various outcomes based on what will happen, when it will happen and why it will happen. Prescriptive analytics answers the question of what to do by providing information on optimal decisions based on the predicted future scenarios. Typically used to optimize a process to ensure delivery of services efficiently to create optimal customer experience. (i.e. hospitals use to predict the likelihood of readmission of a patient if they do or do not follow recommendations)