good business insights

Abstract:

How effective data management and mining can deliver good business insights

Main Article:

Today the life of the average person has become almost frenetic with activities that seem to demand his attention all day and every day. While it has made them look very active it has also made their lives complex. Many opportunities have arisen in just about every domain. For instance providing personal services has literally exploded. Similarly, in the B2B space, needs that are specific to each trade have also appeared and as a consequence the purchasing behavior and pattern of individuals have also diversified. Once upon a time we used to classify household as ‘under 40 or above 40’ but today, that doesn’t belong to any reality. In many areas even the gender factor is not a strong enough criterion to differentiate the marketing messages though it used to be sufficient in the past.

 

This evolution has not come without consequences in the world of data mining. Reality has become more and more complex and data-mining techniques have had to adapt towards it. Many criteria such as the socio-professional category, the age or the gender which were the main criteria used to define marketing targets in the past, can no longer suffice today because it doesn’t make sense any more! For example, a couple with a somewhat low income can make unusually high value purchase even if it means high debts, because it may correspond to a strong desire at a point in their life. These behaviors depending on their stage in life do not correspond to the ‘cliché’ type of behavior that one would traditionally expect.

It becomes therefore harder and harder to use just intuition in marketing. A marketing analyst needs landmarks in the middle of all the data available. The role of descriptive data mining is to build an accurate “map” with the data available.

|

To make an analogy, would you venture into the ocean without a proper chart that would show you how to reach your destination?

Nowadays, the same question applies with all the marketing databases that are available to the advertiser. We just need to cite a few to be convinced: CRM databases, transactional databases, e-commerce platforms, web analytics, call center databases, email database, SMS, etc. each of these databases (whether they are interconnected or not) can represent a different – but mostly complementary – view of the same customer. Extrapolating insights from the mountain of data available is critical for maximizing profits. In this respect, you won’t be able to do predictive analysis if you didn’t complete a good descriptive analysis earlier. Data must be ‘understood’ before trying to create a model. An organization must put itself through an often advanced and detailed descriptive stage for any analysis to be worth the trouble and expense. The complexity will depend on the multiplicity of data one wants to process.

Descriptive data mining must answer the question “Why did it happen?”

Raw data most of the time is unfortunately impossible to understand or utilize. Not only are they very large and composed of numerous different formats, but the information that is really useful may be diluted or even hidden. You can change this data into information, which give scales, dashboards, ideas but you still may not be able to understand ‘Why’.

Descriptive data mining can bring you the tools, which will really help you to reveal the critical and actionable knowledge that lies in your databases.

|

|

Sometimes it would mean using specific technical knowledge, like in finance or marketing, when we have to wear a hat of technical expert along with our hat of a statistician. The broadcasting and sharing of the results of predictive marketing happens mainly at that level. Descriptive analysis enables the creation of true models and all the subsequent predictive computing. Let’s forget the erroneous idea that only predictive marketing adds value. Actually the opposite is usually true. The length of the descriptive data mining stage is sometimes gigantic compared to the length of the predictive stage. It is more or less a Pareto principle: 80% of the time is used in descriptive data mining and only 20% doing predictive data mining. In the end, the predictive data mining is a bit like the cherry on top of the cake, the last result issued from data mining, or the visible part of the iceberg that is published. More often than not the things we learn about our clients, markets, products and campaigns mainly come from the descriptive part.

In an increasingly competitive market, banks that can transform vast amounts of data into valuable information can help generate sales and create differentiated customer experiences.

Overview

The ability to use data to shape decisions and outcomes has become a key factor for banks to maintain a competitive advantage over others and over the passage of time. Yet, for many banks, data and data mining remains an underused and underappreciated activity. As customers demand more and more qualified information and expect more personalized experiences, it is critical for banks to be able to segment clients to unlock their potential.

|

Banks can leverage data analytics to:

  1. Enhance customer acquisition, retention and cross-selling.
  2. Optimize pricing structures.
  3. Gain customer insights through integration with relationship management architecture.
  4. Implement real-time event management to improve contact rates and redemption rates.

In its simplest form data mining can start with the information available on an average ‘Invoice’. For instance just by examining his invoice a medical-store owner can monitor his regular clients, their average purchase amounts, the frequency of their purchase, the medicines that they buy on a regular basis and the quantity they buy. Armed with just this much detail the owner can ensure that he is holding just sufficient stock of the required drugs, he can make a forecast of his sales for the week, make arrangements for payments as they fall due. Over a period of time he can even eliminate the stocking of certain medicines that have poor sales and only occupy valuable shelve space. He can identify those medicines that combine good off take and profits to optimize his sales.

In its most complex form through data mining, a company can streamline its entire activities right from the purchasing of raw material, up to the shipment of the manufactured product. It can give valuable insights into the purchase and stocking pattern of raw material, the time taken to manufacture the finished product, manage its working capital needs etc.

|

|

Sometimes over production of a product or the under production of a particular album would cause a loss to a company. Take for instance the sale of CDs of a film starring a popular actor. Though they would be expected to sell in large quantities, by analyzing earlier sales patterns the sales staff would be able to keep just enough stock to cater to the demand. By studying the past pattern of sales the staff could store them in regional stock rooms to respond quickly and thereby beat the pirates.

Data mining could prove invaluable in the manufacture and sale of perishable products too. Often holding on to stocks when their expiry date is approaching can cause huge losses. Information from analyzing the past will help the management to offer trade discounts to clear stocks that might remain unsold otherwise. These are analysis in the short term but it is in the long term that data mining will be of critical importance.

BUDDING MANAGERS
SEPTEMBER 2014 ISSUE


Rate This Article:  
Average: 5 (1 vote)
Author:  buddingmanagers
Posted On:  Tuesday, 23 September, 2014 - 12:30

Leave Your Comment

CAPTCHA
This question is for testing whether you are a human visitor and to prevent automated spam submissions.
13 + 3 =
Solve this simple math problem and enter the result. E.g. for 1+3, enter 4.