Like crude oil, data needs to go through refining steps, such as collection, hygiene and transformation, before it becomes meaningful. At the pinnacle of this data refinement process is analytics, which is the process for generating hindsight, insight and foresight to make the data actionable and useful for business decisions.
Analytics can be broken into four major types, each employing various analytical methods. As a rule, most business decisions require a combination of different types of analytics.
- Descriptive analytics – Finds out what is happening. It aims to summarize a data sample, rather than learn about the population that the sample of data is thought to represent:
- What happened? What is the problem? How many, often and where? How many participants registered for an event? What marketing channel and event type provided the best ROI? What day of the week provides the maximum registrants?
- Traditional Business Intelligence (BI), dashboard, standard reports, ad-hoc reports, alerts and queries fall into this category.
- Diagnostic analytics asks why something happened:
- Why did the event not generate enough registration? How many different segments of participants are we dealing with? Where are they, and what do they look like? What are they interested in? What is their income, age, number of children, occupation, and regional breakdown?
- Data exploration, intuitive visuals, profiling and segmentation fall into this category.
- Predictive analytics asks what will happen next? Predictive analytical models exploitpatterns found in historical and transactional data to identify risks and opportunities:
- What is making this happen and what will be the future impact of this? Who will respond to this campaign and through what channel? What are the potential values of each customer and prospect?
- Predictive analytics models can generate “propensity to participate in an event” scores for each participant and for other types of future consumer behavior. This is very complex analytics where trained statistical analysts work with all kinds of custom variables and employ technics like regression modeling.
- Prescriptive analytics answers what if. It not only anticipates what will happen and when it will happen, but also why it will happen. Prescriptive analytics recommends options on decisions and shows the implications of each potential decision:
- What is the best that can happen? What if marketing dollars on are spent on email campaign vs. online ads? How will event registration be affected if the price of an event is increased by certain percentage for a certain time frame?
- Optimization, simulation and random testing are two techniques used with perceptive analytics.
Data discovery is an important step that happens for each type of analytics, especially for predictive and prescriptive analytics. This process is where a trained statistical analyst can learn from the data and start building correlations and causality – are two attributes somehow related and does one cause the other to happen.
Soon, ACTIVE will be unveiling its ACTIVE Network Activity Cloud™ product. ACTIVE Network Activity Cloud™ is designed to utilize some aspects of all four types of analytics to drive value for our customers. Descriptive and diagnostic analytics will generate hindsight about events and competition. Diagnostics and predictive analytics will generate insight into forecasts and future results. Predictive and prescriptive analytics will generate foresight to drive options and recommendations for specific actions for improving the performance of the events.
Visit http://www.activitycloud.com/contact-us to talk to an ACTIVE Network Activity Cloud™ expert.