italiano | español


Business Cases

Optimized customer base management

Autori: A. Zampella, D. Ceresa

Advanced analytics for TELCO

Authors: L.Scendrate

- Social Network Analysis(2012)
- Fast Development Analytical Environment (2013)
- Self-updating Data Structure (2013)
- Web Crawling and Competitor Monitoring (2013)

Download the document (.pdf 1.1 MB)

Churn Forecasting

Event: SAS Forum ITALIA 2014
Authors: A. Zampella, D.Ceresa

Download the document (.pdf 1.2 MB)

Innovare nel Credito: dalla gestione dell’insoluto alla gestione del cliente

Authors: F. Cavallo, E. Pallini

Download the document (.pdf 1.3 MB)

Cross Selling Financial products

Authors: M. Labi, D.Maurizio, S. Piccarreta

Download the document (.pdf 310.9 KB)

LGD Risk Parameters Development

Authors: M. Barka, A. Fichera, P. Bauce, S. Piccarreta, E. Pallini, V. Spinelli

Download the document (.pdf 383.4 KB)

Credit rating Decisional Engine

Event: SAS Forum ITALIA 2012
Authors: F. Cavallo, S. Broscritto, B. Marchisio, S. Ghisleni, E. Pallini, G. Passerone 

Download the document (.pdf 761.3 KB)

Wake them up before it's too late! Predicting dormant status to prevent banking churn

Event: A2009 Analytics Conference
Authors: M. Trotta, E. Pallini, I. Frattini, M. Montagna
Award: Best Poster

For a banking institute, forecasting the state of clients’ dissatisfaction is regarded as bottom-line issue to curb the alarming phenomenon of churn of its clients.
The main problem of standard churn models is that the warning doesn’t often come in at the right time. The signal is available when the relationship between the client and the bank is already seriously compromised; therefore the marketing campaigns for customer retention are often inefficient.
What is our idea? The abandonment is often preceded by implicit signals (reduction of the liquidity of current account, investment reduction, etc.), as a sort of dormant status. Analyzing the permanent dormant status in several consecutive months, we pointed out that there is a strong connection between the dormant status and the phenomenon of churn.
We have developed a distinctive churn model for the individual clients of an Italian bank called Banca Popolare di Milano (BPM). The model singles out and predicts the entry into the dormant status prior to the loss of the client.
BPM has a Data Warehouse designed with the SAS system and the entire process of the above model has been implemented with SAS.
The bank will be able, in due time, to offer service proposals that will meet those needs which were not met until that moment and it will therefore be able to prevent the phenomenon of churn.
This innovation represents a crucial element for a bank as BPM whose final aim is to establish a privileged and long-standing relationship with its clients.

Download the document (.pdf 630.9 KB)

A map is worth a thousand words: a Geomarketing business application

Event: A2009 Analytics Conference
A. Saccardi, M. Tosi, P.Vigliercio, A.Carrassi

Geomarketing consist of the use of Geographical Information Systems (GIS) in companies in order to physically analyse questions relevant to marketing, sales and service issues.
GIS takes into account company-owned data as well as population and economic data that have all a spatial component which can be displayed and analysed by means of digital maps.
What is the distribution of customers and prospects?
Where is the concentration of targeted customers?
Therefore, where can an immediate growth develop?
Do “points of attraction” exist?
These and many more business-related questions can be answered efficiently and graphically by means of GIS tools. GIS makes it easier for sales and marketing staff to make a decision through its visual appeal and expressiveness.
We present a territorial analysis concerning the estimation of territorial potential growth of an economic newspaper sales. Territorial analyses give an overview of the market organization within the context of existing sales structures. We stress the importance of performing territory planning through a goal-oriented collection of smaller geographical units in higher territorial units in order to provide the market with the best possible sales performances.


Download the document (.pdf 1.1 MB)

Putting orders in the right order

Events: A2009 Analytics Conference (2009)
Authors: Maurizio Mendaia, Mauro Cicchetti, Salvina Piccarreta, Valeria Nale

“Il Sole24Ore” is a publishing company created through the merger of quite a few corporations. Hence its customer base is heterogeneous and difficult to analyze: the information is available within the company but it is hidden and scattered among many different business systems.

Nevertheless the company does need to know its customers since its goals are:
• to organize increasingly effective and efficient marketing campaigns
• to evaluate the potential of specific areas of the territory
• to check that its own offer is suitable enough
in order to cope with a more and more competitive market.

A complex data mining project has therefore been implemented by using the SAS System. The project has provided a unique customer recognition, detailed analyses and concise reports, the set-up of synthetic indicators, and at last the recognition of behavioral clusters by means of a segmentation analysis carried out with the SAS Enterprise Miner. The results have been used both to support strategic decisions and to implement marketing actions to pursue set objectives.

Download the document (.pdf 772.4 KB)

18 months to make an Analytical CRM dream come true: results obtained by Vodafone Omnitel in B2B churn prevention

Event: SEUGI (2002)
Authors: Flavia Cavallo, Alessandro Bernorio, Guido Cuzzocrea, Paola Bauce

Implementing Analytical CRM, are you ready to go? You need money, commitment, technology, right skills and management . . . ., just like John Belushi had sun glasses, half a tank of gas and above all a strong and determined will to go.

Are you worried about the monsters you will encounter, before getting a possible return on your investment? We would be glad to share what the following points actually mean:

  • Allow the IT resources the necessary time to prepare the infrastructure for “the big project”, but do not wait until the CRM environment is available;
  • transform your business goal into a non trivial analytical design;
  • build a predictive model for churn prevention, considering the future data production process, the continuously changing characteristics of the offer, the timing and details of the customer caring activity,
  • implement in a new production environment a data mining process, developed off-line;
  • do not reach paradise, but achieve a reasonable quality standard to start with successful.

Download the document (.pdf 215.9 KB)

Forewarned is forearmed: Preventing defections of our best customers

Event: SEUGI (2002)
Authors: Fabio Marchetti, Filippo Avigo, Alberto Saccardi, Salvina Piccarreta

BIPOP-CARIRE operates through a network of over 600 among branches and financial shops mainly located in Northern and Central Italy.

In the spring of 2000, under the supervision of the Marketing Department, Bipop-Carire started its CRM project.

In September 2000, in cooperation with Nunatac, a company focused on building decision support systems, took place the start-up of the activities concerning the development of Data Mining environments, analysis models and tools to interface with branches and financial planners during marketing campaigns.

Our presentation will illustrate one of the most significant actions adopted in the winter of 2001-2002 aimed at preventing customer defections. Specifically, the operational process we implemented included the following steps:

Development of a churn model to identify the customers most likely to defect;
Creation of a model to single out, based on their profile, customers who may be targeted as likely buyers of insurance products, which have proved to reinforce customer loyalty;
Customer assessment based on their current and potential values;
Development of a three-dimension matrix - the three dimensions being the three measures generated by the models above - to identify marketing actions targeted to preventing best customers defections.
Enterprise Miner™ allowed us to accomplish the project in few months. SAS/AF® was the front end application we used as the tool for communication to branches and financial planners and for easy and immediate reporting of results.

Download the document (.pdf 232.3 KB)

KALEIDOS: a splash of colour for the “silvery haired”

Event: SEUGI (2001)
Authors: Linda Pimpinelli, Ignazia Vitali, Guido Cuzzocrea, Salvina Piccarreta

Where did those people heading “for the top” actually get to? Have the “career singles” decided to settle down? Finally, are the “movers, shakers and VIPs” still in their respective positions? Nearly 10 years after the ISTAT census and nearly five after the SEAT Geo-demographic Cluster Segmentation, based on these data, were put on the market, Giallo Dat@ has carried out a project to refresh the colouring of the 320,000 Italian census micro zones.

The available data sources used to reach this ambitious objective are heterogeneous in terms of detail. They also cover diverse information categories, and for comparison purposes, a number of extremely consistent samples have been utilised, with, however, no overlapping.

This paper presents the results of the project, it describes the methodology used which enabled us to successfully use a wide range of sources and to construct a classification process which was at the same time sound and rich with information: SAS users can follow a procedure which leads to an optimal utilisation of the integration potential and the completeness of the available analysis techniques in the system. For marketing users, surprises abound, a splash of colour for the “silvery haired”.

Download the document (.pdf 263.6 KB)

From Data Modelling to Business-to-Business Application Development: the Targa Services / FIAT Auto Case

Event: SEUGI (2000)
Authors: Giovanni Lux, Andrea Dorigo
Award: Best Paper (Data Mining Stream)

Database Driven Direct Marketing is not just a matter of implementing an efficient Customer Database, but involves the process of developing customized analysis tools and sharing marketing intelligence in a Business to Business perspective.

To meet these requirements you need:

  • A dedicated Customer Database (CD), structured for data analysis, containing information on identity and buying behaviour of all your customers;
  • Skills and tools for producing relevant market and customer intelligence from the CD;
  • An application for managing outbound and inbound information flows while delivering up to date campaign reporting;
  • A system which integrates Central Sales Dept. with Dealers for sharing marketing strategies.

Targa Sevices is a FIAT Auto Group Company whose mission is to introduce innovative customer oriented marketing solutions.

In this paper we describe the joint Targa/Nunatac effort to achieve the above mentioned objectives with the purpose of showing how this approach gives sounder knowledge of your customers and leads to the development of better performing Business to Business marketing tools.

Download the document (.pdf 298.3 KB)

E = (Dm)2

Event: SEUGI (2000)
Authors: Italo Cattaneo, Riccardo Burtulla, Guido Cuzzocrea, Massimo Saputo

Banca Popolare di Milano’s objective for this project is the rapid implementation of a “listen-and-answer” approach to customers’ needs and behaviour.

You could define this approach as Database Driven Direct Marketing; i.e. an efficient management of Customer Relationships based upon the following tenets:

  1. a dedicated data infrastructure, organized in a customer centric fashion, containing all the key anagraphic, demographic and behavioral information upon which you can differentiate your customers in segments;
  2. the tools and the competences required to analyze your customers, drive and measure your CRM Campaigns and produce a Profit and Loss statement for your marketing activities;
  3. the set-up and training of a dedicated team for CRM Management, driving the Marketing Campaigns and customer interaction;
  4. the integration of this application with the current CRM application in order to deliver all of the above to 1.2 Million Customer via a network of 450 Branches and over 4.000 strong Sales Organisation in the most Efficient and Effective manner.

The Marketing Datamart for Datamining is one of the Bank key components for adding value to our on-going implementation of a new Customer Centric Sales re-organisation.

The Datamart is based on SAS Warehouse Administrator, the strategic platform used in BPM for data management, which seamlessly integrates with SAS Enterprise Miner, the tool of choice for creating our "Business Intelligence" Marketing factory.

Download the document (.pdf 281.8 KB)

Data Mining Methodology Case Study: Customer’s Profitability Predictive Model

Event: SAS Data Mining Methodology (1999)
Authors: Guido Cuzzocrea, Federico Ambrogi
Rivista scientifica: SAS Data Mining Methodology

Customer situation/problem description

The Mondadori Group is a collection of subsidiary companies operating in the publishing industry. One of these, Club degli Editori (CDE - Publisher's Club) was founded in 1960 to sell books by mail order using the book club formula. It is currently a leader in Italy with a turnover of approximately 100 billion lire, and an annual profit of approximately 10 billion lire. CDE has some 900,000 members, to whom it sells some 8 million books annually from all publishers.

The Gruppo Mondadori counts about two thousand staff. CDE is a small unit of less than one hundred people, ten of which are SAS Users. The introduction of SAS System into the company, in 1994, represented quite a turn-around, giving rise to the use of the most advanced statistical techniques.

Download the document (.pdf 216.0 KB)

Data Mining Methodology Case Study: Behavioural Segmentation of Retail Customers

Event: SAS Data Mining Methodology (1999)
Authors: Susanna Ercoli, Guido Cuzzocrea, Massimo Saputo
Rivista scientifica: SAS Data Mining Methodology

Customer situation/problem description
The Banca Popolare di Lodi is head of the banking group of the same name. The group is composed of 11 banks, 280 bank counters, 500 financial backers, about 2,500 staff and it serves about 600,000 customers.

The Banca Popolare di Lodi counts 140 bank counters, 1,400 employees and about 320,000 customers. SAS is used in the department: Divisione Pianificazione Studi, which is staffed by 10 people.

The strategy of expansion that the Banca Popolare di Lodi group has pursued in recent years has led to the acquisition of banking properties which are deeply rooted in their place of origin. Consequently, it has become aware of the need to gain in-depth knowledge of the bank's customers in each particular region, so as to offer products and services that best meet their needs. The growth of credit and the stimulus towards new forms of saving and investment cut across the knowledge of the characteristics, preferences, aspirations and specific needs of each user of the services offered by the bank.

Download the document (.pdf 211.4 KB)

Database Marketing Applications in the Italian Banking Sector

Event: SEUGI (1998)
Authors: Guido Cuzzocrea, Alberto Saccardi 

Many database marketing projects in Italian banks have been started before the development of a Data Warehouse and without a specific Data Mining expertise within the company: these projects have the goal of obtaining operative tools for marketing activities, such as cross-selling.

In our experience, these projects are the best way to test costs and benefits derivable from larger investments in Data Warehousing and Data Mining: dealing with the selection of a work group, the design of a Marketing Data Model, the implementation of data analysis results and, finally, with the evaluation of the return on investment, the company has the opportunity to validate and justify the strategic evolution towards Data Warehousing and Data Mining.

To gain a competitive advantage, banks have the opportunity to exploit the enormous amount of data at their disposal. Data Mining should supply the marketing decision makers with operative tools for the evaluation of client potential, the planning of an integrated offer and the selection of campaign targets.

Two case histories will be presented:

  • the behavioural segmentation of retail customers in a mid-sized bank,
  • the development of a scoring system to select the best campaign target in a large-sized bank.

Download the document (.pdf 329.8 KB)