What is Predictive Analytics?
To attempt understanding what predictive analytics could do to the retail industry – it is vital for us to understand what predictive analytics is and the delta that it offers when compared to traditional Business Intelligence solutions that the whole retail industry was obsessed about when it came into the scene.
Predictive analytics, in essence, is a process of developing data mining techniques that use analytical models to discover hidden patterns and apply them to predict future trends and behaviours. Predictive Analytics takes history into account to forecast and learn to forsee emerging patterns that can be used as a basis to take decisions for the future.
The process of prediction involves the following steps:
- Problem Identification
- Determining the Outcome and Predictors
- Explore data and segregate data
- Test the models
- Apply models to an identified population to predict behavior and evaluate
In most situations, analysts face the callenge of defining the business problem. Successful predictive business problems have quantitive goals.
For example:
- Which of the top 30% of the members are likely to renew the loyalty card, priced at $4 per month?
- What will the contribution of my high margin customers to overall sales if prices were reduced by 13%?
Without an objective business problem, predictive project will not deliver useful results because of the lack of a metric to measure success.
An analytics project entails information about the data that will be used.
The problem definition could include additional specifications such as one or more of the following:
- Where do I get the data?
- Who owns the data?
- How do I get access to the data?
- How much datat is available and how much data is required?
- Does the data pertain to my business problem?
- Do I have enough data?
- Is the data clean, can it be cleansed?
- Where can I process this datat?
If you want to learn more, check out our whitepaper below.