- Augmented Analytics
- Big Data Analytics
- Business Analytics
- Data Governance & Data Fabric Architecture
- Data-Driven Strategy & Transformation
- Performance Management & Improvement
- Qui sommes nous
- Insights Room
- Webroom & Events
Effective usage of predictive analytics is only possible in a data-driven organization.
Oct 17, 2017
What is predictive analytics?
There is more than one answer to this question. But maybe a good one for this could be:
A technology that learns from experience (data) to predict the future behavior of individuals in order to drive better decisions. (Predictive Analytics: The Power to Predict Who Will Click, Buy, Lie, or Die, Eric Siegel)
We can say from this that the best way of predicting what may happen in the future is to look at past events. That's where predictive analytics comes in.
The analysis isn’t magic—but it is normally done with a lot of past data, a bit of statistical wizardry, and some important assumptions.
The main components of predictive analytics
- The Data: predictive analytics is not about the size of the data, but about using the data to improve decision support based on the data quality;
- The Statistics: the set of mathematical techniques, ranging from basic to advanced, which is applied to the data to derive predictions;
- The Assumptions: the key factor in any predictive model – the assumption that underlies it.
Types of business analytics
Gartner distinguishes four types of business analytics:
Types of business analytics
Financial industry must take up the challenge
The challenge is to move from descriptive (and/or diagnostic) to predictive analytics and continuing toward prescriptive analytics once the technical and cultural capabilities are in place.
There are several BI & CPM tools in the market that companies are using currently, some example of this tools used can be Board or BPC SAP among others. An example of this challenge is the way they are using these tools. Most cases within the financial sector, where managers and analysts think that they already are using predictive analytics, in fact, they do not. They are still in the descriptive area.
They use this tools to make forecasting and planning. An example of this could be a sales dashboard with monthly sales. Imagine it shows that sales of product A are dropping month after month, and the management concludes that the product is no longer interesting enough and decides to remove the product from their protocol.
This future decision is based in a forecast trend line where underlying assumptions are a linear model, stable data quality, no seasonal influence and no interference with any other new or competitive product.
This shows the descriptive character common in most of the analytics within the financial industry. Since there is no fact-based prediction supporting the management decisions, none of these examples should be considered as being predictive analytics use cases.
However, these tools already offer the chance of analysis on different scenarios to reflect the impact of external variables on the forecast, as well as, evaluate the characteristics of each time series for a suitable model and they also include statistical functions and algorithms.
When will predictive analysis be possible?
Effective usage of predictive analytics is only possible in a data-driven organization. It all comes down to having access to right "past data" and using right skills, techniques and tools to find business-relevant patterns that can be used to solve similar problems in the future.
Predictive analytics in the financial industry
Most of the business analytics at financial institutions are currently still focused on the "rear-view mirror" resulting in descriptive analytics.
In fact, this majority over the last two decades has been synonymous with transaction reporting. Historical data-oriented reporting systems and better visualization tools have been focusing on "rear-view mirror" organizational activities such as sales, transactions, risk management, customer satisfaction measures or various operational issues.
Financial industry still must invest much more in understanding and anticipate the complete context in which each customer consumes their services.
But there’s a lot of potential
The financial industry has a competitive advantage here over most other industries thanks to the data they possess.
Predictive analytics is becoming essential for digital financial services and used effectively, there are several areas of application for predictive analytics where financial institutions could make profitable investments while at the same time improving the attractiveness of their services. There are some examples of its applications like card-linked offers, next best action, pricing, claim handling, risk assessment to mention few.
Through predictive analytics, the company can predict changes in customer needs, target customers with well-timed and appropriate product offerings, and build loyalty by offering contextual information and advice that improves the customer experience.
Now is the time because of:
These are some key areas where predictive analytics creates value for financial decision-making:
- Card Trading analytics: incrementing Financial control;
- Manage trade instruments across the entire trade life-cycle;
- Pricing: decreasing risk losses;
- Based on past customer behavior, a bank can model the in- and outflow of money;
- Risk assessment: incrementing risk control;
- More accurate predictions about future behavior would give the financial institutions the chance to assess risk even better;
- Claim handling: incrementing HR Management & Operational Efficiency.
Based on the value of the claim, nature of injury, place, characteristics and customer’s insurance history.
- From static offers to real-time recommendations;
- Continuous ranking of possible actions;
- The next best action;
- Mobile is a still growing channel for financial services.
Hence company providers of BI & CPM software products are constantly trying to aim at these goals, supplying tools which can automatize processes in order to create solutions in an omnichannel environment.
Tools are therefore progressively more targeted at a mobile technology, to make easier for the customer to get predictive analysis, algorithms and real-time insights, guiding their daily decision-making, bringing them the knowledge to take the next best action.