FORE School of Management

BUSINESS INTELLIGENCE FOR THE TELECOMMUNICATIONS INDUSTRY: IMPLEMENTATION AT AIRTEL

Course Name: Business Intelligence

Group membersMayank Jain Pradeep Sahoo Rahul Singh Shounak Mondal FMG 19-A

191033 191042 191047 191055

ACKNOWLEDGEMENT

Table of Contents
ACKNOWLEDGEMENT .............................................................................................................. 2 EXECUTIVE SUMMARY ............................................................................................................ 5 INTRODUCTION .......................................................................................................................... 6 DEFINITION .............................................................................................................................. 6 TELECOM ENVIRONMENT ................................................................................................... 6 SHIFTING BUSINESS LANDSCAPE .......................................................................................... 8 THE EVOLUTION OF THE TELECOM INDUSTRY ................................................................. 9 The Challenge ........................................................................................................................... 10 BUSINESS INTELLIGENCE CAPABILITIES .......................................................................... 13 STRATEGIC DECISION SUPPORT ...................................................................................... 13 SCORING AND SEGMENTATION ....................................................................................... 14 CAMPAIGN ASSIGNMENT AND MANAGEMENT ........................................................... 15 PUTTING THE CAPABILITIES TO WORK ......................................................................... 16 DATA WAREHOUSING FUNDAMENTALS ........................................................................... 17 METHODOLOGIES AND PROCESSES................................................................................ 18 ARCHITECTURE: ENTERPRISE VERSUS DATA MARTS ........................................... 18 CRITICAL SUCCESS FACTORS ........................................................................................... 20 INDUSTRY CASE: AIRTEL ....................................................................................................... 22 INFORMATION TECHNOLOGY IN BHARTI AIRTEL ...................................................... 22 THE DATA MANAGEMENT INITIATIVE .......................................................................... 23 CHALLENGES IN IMPLEMENTATION .............................................................................. 25 FIRST CHALLENGE: UNCOVERING PROCESS ISSUES ............................................. 25 SECOND CHALLENGE: MANAGING CHANGE............................................................ 26 THIRD CHALLENGE: BUSINESS INTELLIGENCE ....................................................... 28

FOURTH CHALLENGE: IT IN BUSINESS....................................................................... 28 FUTURE DIRECTION AT AIRTEL ....................................................................................... 30 CONCLUSION ............................................................................................................................. 31

EXECUTIVE SUMMARY

INTRODUCTION

DEFINITION
Data warehousing is the process of integrating enterprise- wide corporate data into a single repository. The resulting data warehouse may then support a variety of decision analysis functions as well as strategic operational functions. This data often originates from a variety of sources, formats, and types and is generally consolidated, transformed, and loaded into one or more instances of a database management systemusually a routing database management system (RDBMS)to facilitate a broad range of analytical applications. The data warehouse may consist of a single large enterprise-wide database, to which users and administrators connect directly, or it may incorporate several smaller systems, called data marts, each of which addresses a specific subject area within the overall warehouse. As a technology, data warehousing is the foundation of the business intelligence capabilities that enable customer acquisition, customer care, and fraud prevention.

TELECOM ENVIRONMENT
In competitive telecommunications environments, customers choose their service providers. Today, this is a reality for long-distance, Internet service provider (ISP), wireless, and some local markets in the United States. Internationally, competition exists in countries such as the United Kingdom, Australia, and New Zealand. However, deregulation of the communications market is occurring in the U.S. local telephone market, and privatization/deregulation is occurring across Europe, Latin America, and Asia. Under competitive conditions, the customer becomes the central focus of the carriers activities. Customer requirements not only determine service offerings but also shape the network and impact the organizational structure of the carrier to focus on particular types of customers.

Competition (and the threat of competition) has resulted in two distinct trends in the telecommunications industry: Industry consolidation In an effort to compete, carriers are looking for ways to enter new markets and offer new services. The most logical and expedient way to do this is through mergers and acquisitions. Convergence The integration of multiple lines of telecommunications services (i.e., video, voice, data, etc.) has become an economic reality, and carriers are beginning to offer multiple services, often in the form of packages or bundles of products.

Telecommunications companies worldwide are exploring business intelligence solutions to achieve competitive advantage.

The key solutions for which telecommunications companies are looking involve marketing: customer retention target marketing campaign management Customer-relationship management Network business intelligence to streamline network assets.

SHIFTING BUSINESS LANDSCAPE

The Strategy

The business landscape of the telecommunications industry is quickly evolving. The previous model, shaped by a handful of competitors in each country, is being replaced by a model shaped by hundreds of competitors vying for a global presence. How do companies survive in this environment? There are two basic strategies that can be pursued:

Product Strategy Companies can continue marketing their products to the masses. This market-share strategy has been very popular in the telecommunications industry in the past. To compete, companies are driven to increase advertising and marketing costs aggressively while discounting their products. Unfortunately, this strategy has driven customer loyalty to an all-time low. For example, it is not unusual for a consumer to switch from Company A to Company B to Company C and back to Company A in the period of a single year. This may be why companies can report a 40 percent Disconnect Rate over the period of a year and still show an increase in market-share. Clearly, this model for doing business presents significant challenges and over time threatens to drive profit margins unacceptably low. The ultimate evolution may be similar to what has been seen in the retail industry, where a good year produces profit margins in the two to three percent range.

Customer Strategy Companies can focus on tailoring products to the individual customer. In this "share of customer" environment, customers are differentiated in addition to products. Corporate resources are efficiently allocated to customer care in relation to the customers lifetime value. Those customers whose loyalty can be earned and whose lifetime value to the company is high will receive a majority of the attention. The result will be an environment that optimizes profits by nurturing valued customer relationships.

Essential to this strategy will be the ability to leverage evolving technologies to accomplish the following: understand the customers needs and behaviors leverage this understanding to identify, develop, and deliver relevant products and services

THE EVOLUTION OF THE TELECOM INDUSTRY

As we review the evolution of the telecommunications industry, it is clear that many companies are aggressively moving (or have already moved) from a business model based on a product strategy to a business model based on a customer strategy. This environment is characterized by customer relationships, product customization, and profitability, and is in response to pressures transforming the business landscape throughout the telecommunications industry.

Growing Consumer Demand Customers (or consumers) are expecting companies to understand and respect their needs and desires. In this world, the customer drives the relationship. It is the role of the business to hear what the customer has to say and respond by delivering relevant products and services (what they want) on their terms (how they want it). Companies can no longer expect to sell several products and services to the masses (mass marketing) but must tailor many products and services (i.e., mass customization) to the individual. This is generally referred to as mass customization.

Growing Competition The ability to refocus a product mix in response to evolving competition is a critical success factor for any business. The key is to be able to anticipate the needs of the marketplace before ones competitors. It is this ability to outpace competitors that most companies find difficult or impossible to do, given todays amalgamation of technologies and architectures. Why is this important? Corporations today are facing more and more deregulation; mergers and acquisitions are blurring the relationships to customers; and globalization of the marketplace and

consumer is opening up businesses to new avenues for expansion and, as a result, new competitors. Therefore, it is mandatory for a corporation to restructure it quickly without losing the ability to compete.

Optimization The ability to measure and predict return on investment (ROI) is something that corporations find difficult to perform rapidly. These measurements indicate the health of the corporation, and the ability to determine them rapidly allows a corporation to change its direction with minimal loss. Other examples of the need for optimization include the ability to determine the most efficient channels for contacting customers; target the appropriate customers for a corporations product mix, and identify new product opportunities before the competition.

The Challenge
Most information systems are built around the product strategy business model. This has resulted in a variety of product-oriented systems that effectively run day-to-day operations. Moving forward, additional systems are needed to augment capabilities produced by these business operation systems with capabilities to deliver best-of-breed business intelligence and business management. These capabilities enable companies to accomplish the following: understand the needs of their business (business intelligence) manage actions based on those needs (business management) effectively run day-to-day operations (business operations)

Figure 1: Additional Information Systems Are Needed to Enable New Business Capabilities

These capabilities will enable companies to realize the opportunity of a business landscape characterized by customer relationships, customized product delivery, and opportunity-driven profit. One of the key enabling technologies to this evolution is the data warehouse. Figure 1 illustrates how the operational data store, as an enabling technology, delivers business management capabilities to complete the information solution.

Figure 2 illustrates how the data warehouse, as an enabling technology, delivers business intelligence capabilities to support business functions in the telecommunications industry.

Figure 2: A Data Warehouse Delivers Business Intelligence Capabilities to Support Intelligence Applications

BUSINESS INTELLIGENCE CAPABILITIES

STRATEGIC DECISION SUPPORT

This is the cornerstone to business intelligence. In this model, end users are provided with intuitive tools to distill information about corporate assets and their performance. Corporate assets include customers, products and services, network infrastructure, and employees. Typical performance measurements include profitability, availability, usage, sales, and lifetime value. Companies can now track key performance measurements, refine customer segments and scores, and optimize campaign strategies. Some of the typical strategic decision-support capabilities in the telecommunications industry include the following: Develop simple reporting capabilities that allow one to measure and trend key performance metrics; these metrics include the following: o install and disconnect rates o call-center average sales per hour o call-center average talk time o campaign performance o customer segment lifetime value o peak network volumes o uncollected receivables o customer satisfaction Develop complex reporting capabilities that allow one to uncover problems and discover new opportunities; typical areas for analysis include the following: o market assessment o channel planning o competition assessment o strategy and pricing o customer penetration and profitability o customer segmentation

o program definition o recognition of patterns relative to customer behavior and needs

Develop statistical models that predict customer needs and behaviors; for example, one can build models that predict a customers likelihood to do the following: o buy a new product o generate high profitability o Respond to contacts through specific channels (e.g., direct mail, telemarketing, email, etc.) o not pay their bill

In addition, models can be built that predict network growth and fraud based on traffic patterns in the network.

SCORING AND SEGMENTATION

These provide the mechanisms for deploying score and segmentation rules developed through strategic decision support. Scoring provides processes that apply statistical models to each customer (or prospect). A score from one to 100 is then assigned to indicate how well the customer fits the model. For example, suppose that a model predicted who was likely to be a high-usage customer. This model would be applied to each customer and a resulting score would be assigned. A score of 100 would indicate a near-perfect match to the model, as opposed to a score of one, which would indicate that the customer did not fit the model at all. Segmentation provides a means for grouping similar customers. For example, one may segment the customer base between residential and commercial markets. In addition, one may decide to provide further granularity by defining segmentation within these sub-segments. Defining customer segments is the first key step toward defining a customer management strategy.

CAMPAIGN ASSIGNMENT AND MANAGEMENT

These start where strategic decision support and scoring and segmentation leave off. Now that we understand what products to deliver, to whom and how, it is time to set up a campaign to orchestrate the contact activity. Generally, campaigns contain six key elements: the list of customers to be contacted as part of the campaign the channel to be used in reaching the customer the product, program, and service to be offered the incentive to be used in selling the relationship relative to other campaigns the priority relative to other campaigns

Once the campaign has been defined, it is executed via the contact management capability. Business intelligenceand subsequently the data warehouseplays a key role in this process by providing capabilities to generate customer lists (element one). Elements two through six are handled as business management activities, as a result of different data, access, and availability requirements. Attempting to perform elements two through six as business intelligence activities (via the data warehouse) is likely to constrain ones information architecture over time.

PUTTING THE CAPABILITIES TO WORK

Now that we have discussed business intelligence capabilities that will be crucial to surviving in tomorrows business landscape, we will examine how these capabilities align to the needs of the business. Customer retention will serve as an example. 1. Strategic decision support would be used to track key performance metrics relative to

customer install and disconnect activity. This would provide early warning of increasing disconnect activity. 2. If disconnect activity began to grow outside of acceptable limits, strategic decision support would be used to analyze why customers were disconnecting and extrapolate impact to profitability. 3. If the profitability impacts were not acceptable, strategic decision support would be used to formulate strategies for retention. 4. Once strategies were formulated, strategic decision support would be used to develop

predictive models that would align retention strategies to the appropriate customers. 5. Scoring would apply these predictive models to the entire base of customers, assigning a score value between one and 100. 6. Campaign assignment would use these scores and other relevant data to assign customer lists to the appropriate retention campaigns. 7. Business management would initiate these campaigns and manage their execution. 8. As feedback is being returned from business management, input would be used by strategic decision support to refine retention strategies. As may be deduced, a number of capabilities are needed to support a single business need (e.g., retention), and these capabilities are integrated through the business process. What may not be quite as evident is that these capabilities can be reused to support other business needs, such as customer care or fraud. Capabilities are essential to providing telecommunications companies with the ability to respond to the changing needs of their customers and the marketplace quickly and cost effectively. The data warehouse is a fundamental enabler to delivering business intelligence capabilities.

DATA WAREHOUSING FUNDAMENTALS

Data may be viewed and analyzed from the warehouse in a number of ways. One very common retrieval paradigm involves the use of an on-line analytical processing (OLAP) engine. OLAP is the term most often used to describe the broad range of analysis for which the data warehouse is used. It is different from on-line transaction processing (OLTP) in a variety of ways, but the key difference is that OLAP focuses on processing queries to the database, whereas OLTP systems process transactions in which information is entered into the database. OLTP applications are rarely, if ever, built on data warehouses. OLAP products can offer a range of advanced analysis capabilities, such as Top N, data pivoting for multidimensional analysis, statistical functionality, the ability to drill from high-level data to successive levels of detail for iterative analysis, etc. OLAP engines can be leveraged to deliver high performance and extended analytics for ad hoc query and reporting tools, as well as for specialized packaged applications or custom executive information systems (EIS). Data warehouses can also be used to support sophisticated operational analysis functions such as customer scoring systems (described earlier) and fraud detection applications. All of these facilities may be used in concert with a separate process known as data mining. Data mining is the practice of polling data for interesting elements or anomalies without actually having to pose specific questions or queries. It usually references sampled data and is often customized to an individual organizations needs.

Figure 3: Data Warehouse Architecture

METHODOLOGIES AND PROCESSES

ARCHITECTURE: ENTERPRISE VERSUS DATA MARTS

As previously mentioned, the choice between single enterprise-wide warehouse architecture and an architecture consisting of many smaller data marts is a point of some contention. The difference is not necessarily one of cost; although a well- planned enterprise solution may require a larger investment initially to account for the more extensive data analysis and modeling up front, as well as to avoid continual hardware upgrades as the warehouse grows. A data mart solution is typically more modular, thus exhibiting a more linear cost curve. Over the long term, however, costs tend to even out, and both systems can allow for expansion and growth. The choice of architecture usually depends more heavily on organizational factors. Depending on the situation, strong arguments can be made for both sides.

Pros and Cons

Decision-support systems existed long before the term data warehouse was coined, often starting in the marketing, sales, and finance departments, with each group pulling the data they required down from the legacy systems into small databases on their local servers. These solutions were often implemented independently and without assistance from information services (IS). Consequently, results tended to be inconsistent, particularly as compared to results obtained by sister organizations with similar but separate decision- support systems of their own. Thus, early wisdom in the data warehousing market favored a single, central enterprise system that would serve all organizations needs. Although some compromises might be necessary in serving the needs of the many, this type of solution would at least ensure consistency of data and scope across organizations. In addition, the information was typically more reliable, as the process of acquiring and cleansing data was more deliberate and more carefully managed. However, the enterprise data warehouse approach has a drawback. It takes time, and it takes coordination between many

groups across potentially many different organizations within a company. Some pioneers of this approach ran into long delays in their warehouse implementations as well as coordination stumbling blocks that either terminated or significantly set back their deployments. In particular, companies that are very compartmentalized in their structure do not lend themselves easily to an enterprise-wide solution. In these cases, integrated data marts can offer greater customization, less contention for system resources, and greater independence for participating organizations. It is important, however, for the reasons stated earlier, that particular care be taken in planning the data acquisition, transformation, and consolidation stages for a data mart. Centralized consolidation can often help minimize data inconsistency between data marts. A sound deployment strategy is to harness the benefits of both approaches by engaging a few key end-user groups as well as the IS organization and designing a skeleton architecture for the broad enterprise data warehouse. However, rather than begin by trying to load, cleanse, deploy, and manage all of the enterprise data, select one or two key applications to focus on first, define the data marts required to support them, and start with a manageable project. Then, as new applications are brought on-line, merge them into the enterprise warehouse architecture and implement them one at a time. This strategy is sometimes referred to as the architected data mart approach.

CRITICAL SUCCESS FACTORS

1. Know Your Goals Perhaps the most important success factor in the data warehouse is to keep ones longterm goals in mind. It is easy to let a warehouse project explode beyond the initial boundaries set for it. One user group finds out that it has been excluded from initial rollout and demands to be included. The initial target user group decides it would be even better if it could do churn management and pricing analysis during phase one, rather than wait until phase two to bring the pricing analysis system on-line. The IS director decides it makes sense to define, build, and load the entire warehouse infrastructure before trying to deliver data to end users. Clearly, it is almost impossible to eliminate scope creep completely. Valid issues arise and must be dealt with. However, it is generally a bad sign when objectives change significantly after final sign-offor, worse, after implementation has begun. Whenever possible, initial goals to which all are agreed should be maintained. Exceptions will always occur, but it is an important rule of thumb.

2. Performance

The data warehouse will fail if users do not get answers back in a timely fashion. This is why parallel database technology, advanced indexing, data modeling, aggregation, and sampling are so critical to the success of the data warehouse project. This is an area in which investment makes sensenot only in technology, but in experts who know how to maximize the utility of the technology.

3. Data Volumes: Loads and Management

One of the characteristic challenges of a telecommunications business is managing enormous volumes of data. Not only do telecom systems generate huge numbers of

detailed call records (and others), but the records themselves are very large, and it is not always possible to know what, if anything, can be filtered out of the warehouse. Complicating matters is that common telecom applications, like fraud detection, require real-time data feeds for optimal effectiveness. As a result, telecom data warehouses must be able to handle tremendous insertion rates, as well as manage databases that will likely grow into the multiterabyte range. Building a system with technology that will not scale effectively as data volumes grow can be one of the most frustrating (and expensive) experiences a company can endure.

4. Relationship between IS and Users

In any data warehouse implementation, it is critical that IS and the user community be in agreement about the objectives of the warehouse and be in constant communication throughout the project. A project driven solely by IS runs a significant risk of slow adoption, low return on ROI, or even outright failure. A project solely driven by the user community risks a duplication of effort, a lack of integration with other critical systems across the organization, and an unmanageable enterprise environment.

5. Executive Sponsorship

Data warehouses are very large, expensive, complex systems. All warehouse projects will encounter tumultuous periods, whether due to technological setbacks, staffing issues, insufficient infrastructure, or poor communication. And this is not an exhaustive list. All problems are resolvable, but if there are no sponsors at the senior management level, these problems can quickly take on monumental proportions. By sponsors, we mean people who understand the business pain being addressed by the project and believe wholeheartedly that building the data warehouse is the best route to resolving that pain. A powerful executive sponsor can help knife through challenges before they impact project momentum. A lack of executive support raises constant doubts about a project and can lead to multiple starts and stops in the implementation, badly demoralizing the project team and the business community.

INDUSTRY CASE: AIRTEL

INFORMATION TECHNOLOGY IN BHARTI AIRTEL

Airtel was one of the first mobile operators in India to adopt sophisticated data based decision making processes and policies. Bharti Airtels evolution to an IT savvy and customer-centric company started in the late 90s. It is not just IT at Airtel said Mr. Rupinder Goel, it is ITi where i stands for innovation and that's how we refer to technology at Airtel! Here IT is not treated as an expense but as a strategic investment. The companys first major IT investment was in the year 1995, when they automated their customer billing systems. Realizing that managing the breakneck growth needed intense focus on core business strategies, Airtel started outsourcing its non-core activities. In the year 2004, the company signed contracts worth $400 million with Ericsson, Siemens and Nokia a deal that effectively outsourced Airtel's entire phone network operations to these vendors. The vendor fee was dependent on the customer traffic and the quality of service provided. Subsequently, Mittal signed a ten-year, $750 million deal with IBM to handle customer management, billing systems and even the companys internal network systems. In May 2007, Bharti Airtel won the Best First Steps award for their managed services agreement in the 2007 Outsourcing Excellence Awards. The annual award, presented by the Everest Group, a global outsourcing consulting firm, recognizes excellence in outsourcing partnerships. According to them, Nortel and Bharti Airtel won the Best First Steps Award after the first year of their managed services contract, reflecting Airtel's customer care vision, Nortel's understanding of that vision, and the technology solution put in place to deliver a best in class call handling experience for subscribers. Airtel currently employs 182 staff members in its IT department and more than 1400 IBM consultants support Airtels IT infrastructure.

THE DATA MANAGEMENT INITIATIVE

Digitization at Airtel started right at its inception as Sunil Mittal believed that data based decision making would offer the company a sustainable strategic advantage. The data management initiative started in the year 1995 with the firm starting to utilize primary data available in the form of CDRs coming right out of a call switch box in a crude technical format to gather meaningful information.

This data provided to Airtel by its network vendors was primarily used for billing, network capacity planning and reviewing basic performance statistics. This was followed by the implementation of an ERP system Oracle Express in 1996, which served as the billing system. This system took in the CDR data and used it to generate automated billing records. The primary goals of such initiatives were improved process coordination, automation of basic operations and better control of information flow. Bharti Airtels first major strategic IT investment phase began in 1997, when Arthur Anderson, Airtels consultant, reviewed their existing ERP system (Oracle Financial) and advised them to start considering implementing data warehousing, which they believed would integrate their existing data and pave the way for mining rich customer information and converting it to a source of competitive advantage. As Rupinder Goel, CIO, Bharti Airtel explained, We realized that as the industry was maturing, the high quality of service delivery, driven by our knowledge of the customer was going to differentiate us from our competitors. Telecom is an industry where understanding the customer is the key to long-term success. It is not like say, FMCG where a customer walks in, buys products and the transaction ends there. There are separate plans for each of the three different aspects of the customer focused business model and as the customers were getting more demanding, the focus clearly needed to shift to retention and enhancement. Bharti Airtel therefore, based on Arthur Andersons suggestions, believed that data warehouse would prove to be a powerful tool that would help them build effective churn prediction models. On the enhancement front, the company wanted to identify the usage patterns at an individual

level and offer each customer customized plans based on his/her usage enhancing the value of the services offered via schemes such as corridor calling There were two different parts to Airtels data warehousing initiative The phase when the management at Airtel was made aware of it in 1997 The phase in which data warehousing was actually implemented in the company more than a year and a half later. Having bought the idea in 97, Airtel started gathering requirements to build the data warehousing platform.

CHALLENGES IN IMPLEMENTATION
The initiative began with a marketing representative from Telecom Italia, which had a twenty percent stake in Bharti Airtel, leading the effort. The person was well informed on the implementation aspects of multi-dimensional analytics and data warehousing and was involved in the initial phase. However, he quit when Telecom Italia divested its holdings in Airtel and the project was still in the requirement gathering phase. The internal IT team took over the project and stated analyzing and evaluating the technology options available. The choices finally narrowed to Oracle Express and SAS. At that point in time, SAS churn management system was already being used by many major cellular service providers including Hutch and BPL5. But Airtel had already implemented the ERP system with Oracle and were comfortable with the vendor and the product. Also, Airtel believed that by choosing the same vendor (Oracle), they would not have to confront many compliance issues from the vendor perspective during the implementation of the data warehousing system. Once the vendor was chosen, Airtel zeroed in on KPIT, a technology solution provider based in Pune, to assist in the implementation of the data warehousing solutions. The system went live in 1998.

FIRST CHALLENGE: UNCOVERING PROCESS ISSUES The first three months after the implementation were crucial during which the internal IT team monitored the system closely. The IT environment was in a flux as internal policies, processes and workflows were being scrutinized and streamlined where necessary. During this initial phase however, the decision making process was still based on the pre-data-warehouse information as the credibility of the new system was yet to be established and there was reluctance among the business users to switch to alternate sources of information. The acceptance by the users came as a result of a surprising discovery, according to Abhay Gupta, IT manager at Bharti Airtel. There was a daily usage report generated from the telecom switch, to which the CEO paid close attention. The report summarized the usage volumes based on the crude switch data. The IT manager attempted to reconcile this data with MOU (Minutes of Usage) from the billing data obtained from the data warehousing system.

The results were startling with the data warehousing reporting a 15% 16% higher customer billing revenue compared to switch data. The manager reckoned that customers would not have been billed more than what the usage warranted because during those days, the cost per unit talk time was high and such an erroneous overprice would have certainly resulted in a large number of customer complaints, which was not the case. Further analysis of data obtained in next three days revealed that the discrepancy was not an isolated incident and had a consistent pattern. We headed straight to the CEO and informed him that the reports provided to him were incorrect! recollected Gupta. The CEO immediately spoke with Ericsson, Airtels network vendor who provided the daily data that was being analyzed. Ericsson clarified that the data provided was based on the average line usage and not the actual call usage. During that time, the ASR (Answer/Seizure ratio) was quite low, since the incoming calls were chargeable, resulting in many subscribers literally using their mobile phones as pagers. They would make a note of the incoming call number and return calls using a landline. This led to a large discrepancy between the total network usage and the number of actual calls made. The daily switch report that Ericson provided was originally meant to be used only for network planning exercise and not for making business decisions. Thus, a key insight that Airtel obtained within a couple of months of using the data warehouse was that critical business and marketing decisions were being based on daily revenues that were around 15% 16% less than the actual amount. From that point forward, the firm used only reports that were generated out of the data warehouse system to make business decisions.

SECOND CHALLENGE: MANAGING CHANGE Within the next few months, Airtel made numerous changes to both internal control systems and business processes. For example Airtel had four monthly billing cycles. The customers were billed on the 2nd, 8th, 16th and the 23rd of every month for their previous month use. However for the reporting purposes, the finance department was using an end-of-month report generated on the last business day of every month (which contained both billed and unbilled data) where as marketing division was using only the actual billed data i.e. from the 1st to the 23rd of the month. This led to a discrepancy of about 15%, which caught the attention of both the

management and the IT department. After careful deliberation, heads of both departments decided to collate pertinent information from the 1st to the end of the month, but only after the next billing date, which was the 2nd of the subsequent month. Thereafter the IT department took another 4-5 days to consolidate the data and got the final report out on the 6th or the 7th. Hence, Airtel made a conscious decision to use the previous months data to make business decisions for the coming month.

THIRD CHALLENGE: BUSINESS INTELLIGENCE In 1999 Airtel was operating in only 1 circle Delhi. But with the mobile market growing at a rapid rate, they acquired fourteen more circle licenses within a few years. The Oracle data warehouse, though quite robust was difficult to scale up. By now the business users had also realized that though the system managed to integrate data across systems, it had limited capabilities when it came to information output. They switched the billing system to another customized solution by a well known vendor. However, this system lasted only for a brief period from September 2002 March 2003. During this phase, getting any insight from the data was tedious and the data extracted was inconsistent with the actual observations. In addition, the expansion plans and the potential growth called for more focus on achieving greater operational consistency and Airtel decided to implement a centralized billing system and a comprehensive and full scale data warehouse. We were evaluating our options among SAS, Oracle, Teradata and DB2. After carrying out the necessary due diligence, Bharti Airtel finalized the deal with Teradata.

FOURTH CHALLENGE: IT IN BUSINESS When Bharti Airtel was using Oracle database, the management decided that business users primarily in the marketing department would write their own Structured Query Language (SQL). There were 400 business people who have been trained in SQL. Airtel wanted to enable the business users to be able to write ad hoc queries based on their information requirements. As Munish Kantor, marketing manager, Airtel, observed, IT staff would assist them when needed, but we decided to ensure that the users were able to write queries, extract useful information, store the same in a suitable format and understand how the data was organized.

Bharti Airtel also conducted a three-day training program on the effective use of SQL. As Munish recalls, they even went to the extent of educating them on the technical aspects of Relational Database Systems (RDBMS), certain key tables and data organization. Airtel invested on instruction modules so that the users could be trained easily. Our justification for this initiative was that IT could assist the users, but that certainly would not be their priority. The

IT department believed this priority conflict could lead to a 15-day lag, which could prove costly. Airtel does have an IT helpdesk to assist the users when needed. The business people however were not very competent in SQL and the queries they could write are suboptimal. Hence the queries load the system substantially and the outputs are slow. Our approach is a business strategy and not a technology strategy We gather the business requirements and then proceed with our implementation plans. We recognize that technology is important, but ensure that business objectives are chalked out clearly and met using technology as a tool. Initiatives such as data warehouse implementation change the business dynamics and therefore, need to be driven top-down with stakeholders and sponsorship from business.

FUTURE DIRECTION AT AIRTEL

We started with using the standard reports and then moved on to using ad hoc queries. The future direction is towards understanding and serving the customer better. The data required to do so is currently generated from queries and reports. mentioned Abhay Gupta, IT manager at Bharti Airtel, Currently, on 80% of the occasions, we use standard reports and on the remaining 20%, rely on ad hoc queries. We are moving towards 30% standard reports and 70% queries and the goal finally is to base the entire churn prediction model and marketing management on real-time analysis

Another area in which Airtel could possibly leverage its data management expertise is the Value Added Services (VAS). In the telecommunication industry parlance, VAS refers to all services beyond the standard voice calls. VAS is seen as key to drive an increase in ARPU as it paves the way for customers using the operators services more extensively than their traditional voice-only needs. Mobile VAS industry, estimated at $0.55 billion in 2007 and 5% 10% of the total revenues for a company, was expected to increase to more than fifteen fold to over $ 9 billion by 2010. We see immense potential in VAS, but they can only be targeted towards the high end customers as of now. Identifying these customers is hence the key.

Targeted marketing is also becoming important as it allows the service providers to market their offerings to target customer segments effectively. There has been a paradigm shift in our operational model. Now, we were making informed decisions rather than just decisions. We are gaining insights into the underlying customer behavior patterns for every service that we offer. The decision-making process has effectively moved from an observation to a prediction model.

CONCLUSION
As telecommunications markets become increasingly competitive, the ability to react quickly and decisively to market trends and to tailor products and services to individual customers is more critical than ever. Although data volumes continue to increase at an astounding rate, the problem is no longer simply one of quantity; at the heart of the issue is how companies are using their information. Increasingly, particularly in the telecommunications industry, it is important to understand customer preferences and behaviors. A data warehouse can be a very effective means of organizing and analyzing the complex barrage of information generated in ones business and helping to generate a more effective business model for keeping ones customer base happy and profitable.