The Drivers for Customer Assurance
Traditionally, network information was used by network engineers and operations, and business data were used separately by marketing, finance, and sales. Also, in traditional models, phone companies sold dial tone. As can be seen in Figure 1, the traditional model is a distant memory. More services, more complexity, more partnerships, and increased focus on the need to be truly customer centric are all drivers in today's market. With the introduction of multi-play convergence, Telcos now have more data than ever before from a wide range of BSS source systems. Getting all this divergent data into a single repository to enable convergent business intelligence is becoming the norm with leading Telcos.
The opportunity is now to take convergent business intelligence to the next level, and integrate network OSS data with BSS data, for all networks and services. This will be required to become truly customer centric as customers will use voice, data, and content services over circuit switched and IP networks. Understanding the customer's experience for all services, over all networks, is key.
Adding relevant information, such as customer value, service level agreement (SLA) performance, and churn score to customer experience records provides a deeper level of understanding of the customer and enables faster responses to situations. This intelligence can only be sourced from network OSS and BSS systems. Bringing these data together, turning them into information, and presenting them to the end user as business intelligence enables companies like yours to make better decisions. This is the role of the enterprise data warehouse (EDW), which provides the intelligence layer for all services, for all networks, and for all customers to give a single view of the business, particularly as related to customer experience.
Using signaling data as the source for OSS network data provides a richer set of information than switched data. Combining this OSS information with the BSS information that resides on an enterprise data warehouse provides Telcos with enhanced intelligence – the ability to get a single view of their business from the network edge all the way to customer care and finance. This enhanced intelligence enables Telcos to take a proactive approach to customer assurance. By using this end-to- end information, Telcos can create an enterprise customer profile and measure the customer's experience at the relevant customer interaction touchpoints – from when the customer cannot make a phone call and gets a network busy message on his phone to that time when the customer responds to an advertising campaign.
Providing proactive customer assurance using enhanced intelligence requires capturing bad news as well as good news. Often Telcos use mediated and/or billed data for churn prediction models and are, in fact, missing the negative elements of the customer experience. This is because there are cases when an attempted call doesn't even make it to the switch. Getting a fast busy signal on a call indicates network congestion. As such the switch doesn't record this call attempt. Also the memory in switches can become congested, and then they stop recording and switching calls. The impact the call from the CSR will have to the end user is related to the time the contact was made. If the information on the failed downloads is not available for several days to the CSR, then a phone call to explain how to download music may not stimulate the user to use the service, as they may have already written it off as a waste of time.
This paper will discuss sample business uses for Telcos to utilize convergent OSS/BSS business intelligence.
For telecommunication companies to get an end-to-end picture of their business, the relevant information needs to be in a single repository that is accessed by a wide range of business users. Some older Telcos rely on departmental data marts. As well as the duplication of resources required to run and maintain numerous data marts, this approach leads to duplication of data and multiple versions of the truth. For example, a simple question about new subscribers could result in numerous different answers, dependent on the department asked. The leading Telcos use an EDW approach where all the data are stored in a central repository. This leads to data consistency, the ability to reuse the same data for different purposes (store once, use many), and reduced operating costs. Figure 2 highlights this centralized EDW taking OSS signaling data, as well as BSS data, and making it available for all users in the Telco.
Sample Business Applications for Convergent OSS/BSS Business Intelligence
Customer Retention and Acquisition
SLA Management and Performance Monitoring
Many of the most valuable customers a Telco has use telecom services to transmit mission critical data. For example, a utility company may use GPRS or 3G-based networks to schedule all work orders for their engineers. Dispatch companies may use mobile telemetry to monitor shipping and delivery performance. Network downtime or slow transmission speeds can cost these corporate customers millions of euros. When a Telco is looking to retain these valuable customers, they know network performance is critical. Many large corporate customers are now demanding SLAs as part of their contract with their Telco. Conversely, when such corporate customers are issuing tenders for their mobile business SLAs are now becoming an important element in their decision making. Having this capability enhances a Telco's competitive offerings, especially in saturated markets.
Telcos need to be able to capture the measured network experience of customers and integrate this with the business intelligence produced by the EDW. By combining network information, usage, and customer intelligence, operators are now in a better position, ideally based on customer value and not just revenue, to know which customers to offer SLAs to, as well as have the ability to track and monitor performance against SLAs. Using only switched data to monitor SLAs would be risky because the Telco wants to be able to quickly identify the unsuccessful calls and be able to drill down to establish the root cause. This level of detail required for identification and root cause analysis is found in signaling data.
Figure 3 shows that current BSS intelligence practices may only be looking at the tip of the iceberg, and the addition of OSS data provides a much deeper and insightful profile of a customer.
The more variables that are measured to score a customer's propensity to churn, the more accurate the churn model will be. However, most churn models take data from the switch and then from BSS. Adding the missing dimension of OSS data to churn models can increase accuracy of churn modeling. Adding OSS data, and being able to quickly model and update churn propensity scoring within hours, will enable corrective action to be taken more quickly. When one considers that the average monthly churn rate in mobile is 2.3%1 and that operators are reporting multi-billion euro revenues, the financial impact of an annual percentage point increase in effectiveness of customer retention through better identifying potential churners is significant.
Customer Experience – New Customer Monitoring and Viral Marketing
It is often discussed in the mobile industry that their first thirty days of the life of a new subscriber affect how long they are going to stay with their Telco and the uptake of new services. Customer experience for new customers and users of new services needs careful tracking in the first thirty days to enable any potential problems to be quickly identified and corrective action taken. An example could be a new 3G user trying to download music from a Telco portal. Due to user error, the user tries to download music three times, and then gives up. Identification of these failed downloads is available from OSS data. A customer service rep (CSR) can call the user and talk them through the process of downloading music from the Telco's portal. In this example, it can be seen that timing is important. The impact the call from the CSR will have to the end user is related to the time the contact was made. If the information about the failed downloads is not available for several days to the CSR, then a phone call to explain how to download music may not stimulate the user to use the service, as they may have already written it off as a waste of time. However, if the call was made within minutes of the attempted download, the user would be more willing to try again, and become a regular consumer of mobile content. Figure 4 demonstrates the need for data to be fresh so the actions taken are relevant to the end user.
All too often customer experience is limited to network experience. Dropped call, unable to connect, and incomplete download equals bad experience, whereas completed call and, successful download with acceptable QoS equals good experience. As well as network OSS measures (download successful in so many seconds, etc.), Telcos can extend customer experience to identify subsequent usage patterns of service – for example, a customer downloads a game. Did they get their friends to download the same game? Did they purchase similar games from the same content provider? With content, the measure of customer experience will be more dynamic than with voice. Identification and analysis of customer experiences for social networks will be important for Telcos to control viral churn and influence viral marketing.
One of the problems with the wide range of handsets and devices on the market today is that they are not all equal. Some handsets perform better than others. For example, some content cannot be accommodated on certain handset models. Telcos need to know the best and worst performing handsets. Also, in many customer records, the handset listed is often the initial handset purchased from the Telco. If a customer purchased a new handset from a third party, then the marketing department of the Telco often doesn’t know about this purchase. By integrating OSS data with the BSS data that resides on the EDW, the handset information for every customer can be kept up to date. Telcos now know who has which handset and the capabilities of this handset. This can lead to improved targeted marketing and better customer experience.
If a Telco can quickly identify nonperforming equipment and produce business intelligence from the EDW, they see the truth. For example, a Telco launched a new 3G data card and heavily subsidized its supply to corporate customers. In eagerness to increase non-voice ARPU, the Telco signed up to strict network performance SLAs. As soon as the OSS data showed problems with these 3G data cards, the Telco was able to identify the corporate customers affected, measure their SLA exposure, calculate how this negative experience would affect churn scores, and calculate what the immediate revenue impact is. As this Telco had an end-to-end view of the required OSS and BSS information, they were able to prioritize the remedial action by the customer’s churn propensity score and profitability. Many handsets can also be misconfigured or not work at all with, for example, GPRS. Typically when a customer swaps the SIM to a newly acquired handset of a different brand and or capability, the handset is not configured for the services the customer has used on their old handset. Access point names (APN) – mobile URLs – can be mistyped, leading to customers not being able to access the content services that the operator has spent millions of euros promoting. This leads to customer frustration and lost revenue for the operator. By analyzing mobile data access attempts in the EDW, Telcos can take corrective action.
There are many marketing campaigns running that encourage outbound roamers to select a specific partner’s network when they arrive in a foreign country. Yet, how many customers actually bother to go to the effort of manually selecting a network when they switch on their handset at the airport. In most cases, they are left at the mercy of the strongest signal strength wins. Telcos need to know, as soon as possible, when their home subscribers are roaming and on which network they’re roaming. OSS data can pinpoint, in real time, who is roaming and on which network. Having this information available within the EDW enables Telcos to take the necessary corrective action – e.g., sending the customer an SMS reminding them to manually select network so they’ll get better call value – and it also provides valuable information when renegotiating contracts with roaming partners.
OSS data can provide information about the best (and worst) in bound roaming cell site capture points. However, using this information for network management provides only half the picture. Knowing both the historical and current behavior of inbound roamers also helps operators analyze the most valuable inbound capture points and allocate network spend accordingly.
Relating customer behavior to content is critical as operators look to increase non-voice ARPU. Being able to measure the success of specific content for specific segments in near real time gives operators the degree of flexibility they will need. Content is much more dynamic than voice calls. Telcos need to think like retailers of fast moving consumer goods (FMCGs) and look for immediate bundling, cross-sell and up-sell opportunities tied into a specific customer’s behavior. OSS network data provides Service Usage Records (SURs), which provide the required information for a Telco to carry out click stream analysis for both on-portal and off-portal traffic. Having this information available and being able to cross reference this with the network experience for each content partner’s site visited is important. This information can then be provided to the Telco content partners to enable intelligence about who is visiting their sites, what their experience is, who is browsing, who is purchasing, what is being purchased, and so on. This detail of information also opens up opportunities for very targeted mobile advertising, thus developing new revenue sources for Telcos.
Using signaling information as the basis of truth for all usage and building CDRs and SURs from these data enables a new layer to be added to many revenue management applications. In retail revenue assurance, there is a general view that most revenue assurance systems measure switch to bill. However, as discussed in the introduction, using signaling data as the source for OSS network information provides more detail than switched data. This can provide a new dimension to revenue management as Telcos will be losing revenue due to switches not recording calls. This paper discusses two uses of OSS network data and BSS data in an EDW for revenue management purposes.
Interconnect Billing Verification
As a demonstration of the store once, use many approach to EDW, Telcos can demonstrate a very quick ROI from the investment in adding OSS data to the EDW. Comparing OSS signaling records by interconnect group, with mediated records by interconnect group, enables Telcos to see the off-net calls that they wouldn’t be able to bill. This enables corrective action to be taken and the calls but quickly back into the billing cycle. So, when the interconnect bills come in, the operator has the basis of truth for every call connecting with an interconnecting network. OSS signaling records can also be used to validate out-going interconnect bills to demonstrate billing accuracy for dispute management.
Content Settlement Assurance
For content partners, operators will need to ensure that they are paying out the correct amount of money and not opening themselves up for disputes. This is similar to traditional interconnect settlement assurance, in that it will be the party who can prove that their wholesale bill for third-party financial reconciliation is accurate that will save money.
Looking at billing data on its own is not going to solve such disputes. Take the example of a user downloading a game. He clicks on the start download icon and starts the download. After thirty seconds, the download is not complete, so he stops the download and starts again. Same story. Only on the third attempt does the user manage to successfully download the game. The content provider views this as three downloads and will send a wholesale bill to the Telco, which shows and charges for these three downloads. The Telco will pay this bill. Meanwhile, the retail user has been charged three times for the game he successfully downloaded once. He refuses to pay the bill, and his satisfaction with his Telco plummets.
If the Telco had the OSS data available to show that the user started three downloads but only one was successful, it would be able to dispute the content partner’s wholesale bill, and also use these data when discussing SLAs with the content partner. The Telco would also have been able to take preemptive action with the retail customer – check his bill, then call and, if he’s a valuable customer, offer him a free game.
These are just some sample business uses for OSS/ BSS integrated into an enterprise data warehouse.
The business benefits to be gained from the integration of OSS and BSS information on a single EDW are substantial. What we’ve already described is only a sample, and new business applications and uses will emerge as users discover the possibilities of customer insight by combining OSS and BSS data. It is important to reflect on the requirements that achieving true customer centricity will place on a Telco’s management information systems. Along with the increased rollout of IP data networks, this will result in increased demands for processing performance in terms of data throughput and also in terms of minimizing data latency. Only high-performance information systems will deliver the results that the Telcos will expect.
1Wireless Intelligence, reported in Telecommunications Magazine, October 2006.
About the Authors
Martin Morgan, of Teradata, has worked in OSS and BSS for more than 20 years. He is a regular speaker at conferences and has published more than 40 articles in the telecommunications trade press. Martin was a founding member of the GBA (Global Billing Association), and his areas of expertise include billing and revenue management.
Tony Whale, Teradata Industry Consultant, has worked for a number of operators and also vendors in the network, BSS and OSS space over the past 20 years. He is Teradata’s SME for network applications for operators built around enterprise data warehouse capabilities.
Alan Martin, of Agilent, has specialized in the area of Service Management for the past 15 years, traveling around the world working with service providers to solve their critical needs in characterizing service experience issues, assuring service revenues, and troubleshooting customer related problems. Prior to joining Agilent Technologies, Alan worked in the OpenView Software Division at Hewlett-Packard.