DESCRIPTIVE ANALYTICS: require further investigation

Predicting staffing and resourcing needs

The descriptive analytics process is as follows:
PRESCRIPTIVE ANALYTICS builds on descriptive
1. Ask a historical question that needs an and predictive analysis by recommending
answer, such as “How much of product X did we courses of action that will reap the greatest
sell last year?” benefit for the organization. In short,
2. Identify required data to answer the question prescriptive analytics tells you what should be
3. Collect and prepare data done in a given situation. It helps executives,
4. Analyze data managers, and employees make the best
5. Present results decisions based on available data.

A good example of prescriptive analytics is the

Examples of descriptive analytics include:
field of GPS-based map and direction
applications. These applications provide route
- Summarizing historical events such as sales,
options to a destination based on traffic volume,
inventory, or operations data
road conditions, and maximum speed. It can
- Understanding engagement data such as likes
then prescribe the best route based on user-
and dislikes or volume of page views over time
defined objectives, such as shortest distance or
- Reporting general trends like revenue growth
quickest time.
or employee injuries
- Collating survey results

DIAGNOSTIC ANALYTICS enhances the

PREDICTIVE ANALYTICS
descriptive analytics process by digging in
deeper and attempting to discover the cause(s).
The predictive analytics process is as follows:

1. Identify anomalies (inconsistencies) in data

1. Ask a forward-thinking question, such as “Can
sets
we predict how much product X we will sell next
2. Collect data related to the anomalies
year?”
3. Use statistical techniques to uncover
2. Collect and prepare data
relationships and trends that could explain the
3. Develop predictive analytics models
4. anomalies
4. Apply models to the prepared data
5. Present possible causes
5. Review models and present results

An example of diagnostic analytics is using

Examples of predictive analytics include:
subscription cancellations, correlated with
customer comments and ratings, to determine
Forecasting customer behavior, purchasing
the most common reasons why users cancel
patterns, and identifying sales trends
subscriptions.
Predicting customer preferences and
recommending products to customers based on Another example would be determining
past purchases and search history whether there is a correlation between the
Predicting the likelihood that a given customer demographics of consumers and their
will purchase another product or leave the store purchasing patterns at specific times of year.
Identifying possible security breaches that
 to draw inferences and conclusions. Often,
analysis relies on statistical techniques and
software tools such as spreadsheets and
visualization applications.

6. Presenting the results - This is usually the last

step for data analysts. It is the process of
communicating the results to decision-makers.
This can be done in the form of a report,
graphical representations, or a combination of
both. Sometimes the data analyst is also asked
to recommend actions based on results.

1. Asking the Question - The analytics process

always starts with a question to be answered.
Some questions are simple, such as “which
bicycle color is most popular with our younger
buyers?” Others are very complex, such as “why
are certain types of cancer cells exhibiting
resistance to radiation treatments?”

2. Getting the data - This step involves the

process of locating and obtaining data that is
relevant to the question, and then determining
if there is enough data to complete the analysis.

3. Investigating the data - Data comes in many

forms and from many different sources. This
step involves determining if the data is
complete and contains the relevant information
for the analysis.

4. Preparing the data - This step can involve

many tasks to transform the data into a format
appropriate for the tools that will be used to
analyze and present the data. This process is
sometimes referred to as “cleaning” the data,
because there may be blank records or obvious
errors in the data set.

5. Analyzing the data - Analysis is the process of

identifying patterns, correlations, and
relationships contained within a data set or sets