IBM and Stream Computing

 CEM, Mediation  Comments Off on IBM and Stream Computing
May 032012

Normally I do not write about vendor products but since today’s topic seems to be a new driver for the industry, I will make an exception. Today’s topic is about IBM Infosphere Streams product which introduces a new term to our industry: Stream Computing.

Stream Computing concept stems from the fact that todays’ OSS/BSS environment is composed of “streams” of data flowing between the systems. Each stream serve different purposes. The stream that fetches CDR data to the mediation system over the ftp protocol serves to the Order to Cash process. Another stream that looksup the segment information of a given MSISDN from the
Campaing Management Application could serve to a different end-to-end process.

Stream computing allows us to “intercept” and do additional actions over the traditional streams that we use in our operations.

Take mediation example: We fetched the CDR from the switch to the mediation system. This is a standard ftp operation and until it finalizes no system has a control over its payload. If we put a stream computing system between these two, we can intercept the data and play with the payload. Here is how it works:

You create the main stream in the Infosphere Streams system, which does the real thing: ftp, from one place to the other. This operation is transparent to the OSS systems in the chain: The mediation system “thinks” that it is getting the data from switch. And since we do not “touch” this main stream, there’s little or no latency in the mediation process.

The magic, however, lies within the Infosphere Streams. With this product you can “clone” the stream to serve parallel different purpose. Following the same example, I clone the main stream and I have two output data now: CDR information. The second stream, goes to another OSS system which checks the MSISDN with the campaign mgmt system to see if this is a VIP activity. If so, the VIP customer can be SMSed after the call for example.

Streams reaches billions of events per second in the data processing speeds, which can easily cope with Telecom’s moving data speed and volume.

As you may know, IBM is spending too much time for the research an development of AI systems. AI studies go in parallel with big data studies. The most recent outcome of these studies was the introduction of Watson, which is an AI program equipped with big data processing algorithms.

It will be wise to combine stream computing with these AI study outcomes and IBM seems to be moving in that direction.

Mar 222012

Today, I want to talk about a new trend that seems to popped up in the SQM/CEM field: Mobile Device Agents.

Mobile Device agents are software components that reside on user devices and collect statistics about the quality of user experience which will enable the operator to act upon service degradation. Operator can also have the same data correlated with service quality data to plan future service improvements.

Device agent term is fairly new for the mobile industry. However, this is not the case for the fixed line. In fixed line, operators have been collecting metrics about the given end-to-end service for years. These metrics are collected from CPE(Customer Premises Equipment) devices (mostly routers and L2 switches) that reside on the customer premises. Ideally, but not necessarily, these devices are also managed by the operator, taking the name Managed CPE. By utilizing data coming from these CPEs, operators are able to measure not only the core network health, but also the Access side.

In order to increase user perceived quality,  service providers continuously seek new datasources that will give clues about the customer’s service perception. Customer usage data can be collected in several places:

–          Probe systems

–          DPI systems

–          Device Agent systems

Probe system and DPI system can provide the top most visited URLs, throughput/speed kind of statistics that will give clues about the service usage. Probe systems can additionaly provide call drop statatistics and catch device configuration errors.

Device agents can do both. But, they also provide device related information such as signal strength or battery status. They even can tell which software along with their versions are installed on the phone.

If we collect all this data (usage + device + signalling) and correlate successfully, we can do lot’s of customer experience related analysis with it. We can detect that a specific service usage drop from the DPI system and correlate this with mobile phone configuration errors.  The dropped calls can be correlated with device battery information to see if the dropped call has occurred because of a device problem. In some cases where the operator has not done any investment to DPI and probe systems, just the Device Agent system can provide all that data.

But why device agents are not so popular? First answer is the privacy. Most people will not want agents on their phones that are sending their usage patterns to somewhere else. There are not so many regulations around this but we should expect to see them soon.

The second answer is more technical. The agents consume processing power and drain the batteries soon. In order to get rid of this, agents should not always be on-line, and collected statistics should be uploaded in relatively longer intervals (a couple hours). That late data cannot be utilized by SQM systems so it can only be used for late correlation and planning purposes.

Device agents use push mechanism and upload their statistics to a central server where further correlation and reporting functions can be executed. However, because of the reasons I have provided, they cannot be real-time data sources which are required by most SQM/CEM systems.

Who are the VIP Customers?

 CEM  Comments Off on Who are the VIP Customers?
Dec 122011

VIP customers of the service providers have always got the attention and put first in the retention and churn related initiatives. But who are the VIP customers?

In the early days of telecommunications, where there are only primitive tools available to the sector, service providers started to search for their VIP customers in their CRM systems. CRM systems were able to provide demographic information, along with the purchasing history. By looking at this data, SPs were able to identify most spending customers. The spending behavior is then further be analyzed to identify the customer lifetime value (CLV) which looks at the spending behavior from the first day customer has activated his service.

CLV is a valuable information but not enough to identify the VIP customer segment. There is a second type of users that also needs to be taken into account. These users are called influencers. Influencer term comes from marketing context and refers to people who has strong impact on the spending behavior of other users.

If an influencer is not happy with the service and starts to talk negative about the service to his/her “followers”, the perception will fall gradually. In the worst case, if the influencer churns, the followers may follow. Because of these reasons, influencers should also be identified and treated as VIP customers.

Identification of influencers is done by a technique called link analysis. Link analysis analyzes all the entities (customers) in a system and counts their connections between them. The more connections one entity has means it is a better candidate for being an influencer.

In the telco world, link analysis started first by looking at the call patterns. An entity who is called by a diverse community is a good candidate for an influencer. “Call link analysis” is done by using CDR information and most mature service providers run it for marketing purposes.

The new trend is social media and link analysis technique should also be applied to this area as well. If a user has lots of connections in his/her Facebook, Linkedin or Twitter profile, a score could easily generated for this user. Looking at blogs to collect the comment counts could also be another data source for social media link analysis. For crawling such online resources an enterprise content management system can be utilized.

In order to do the social media link analysis effectively, the service provider should also match it’s customers with their social media fingerprint. At last, there are multiple Murat Balkan’s on social media but which one is the customer of that specific service provider?

The service provider should enrich CRM’s and self care application’s data models to include this kind of information. The entrance of this information can then be promoted by providing free service units, gifts etc.

The ultimate goal should then to reach a unified influencer score that combines the usage and social media link analysis results. The scores above a certain level then can be segmented as VIP customers along with the high CLV customers.

Mobile Device Repositories

 CEM, Device Management  Comments Off on Mobile Device Repositories
Nov 212011

One of the first requirement for mobile portals is to provide customized and optimized content which will enable a consistent customer experience throughout the mobile channel. Mobile portals should convert the content to the desired format on the fly before the content hits the consumer device.

In order to do this conversion, portal (or conversion enabling tool underneath it) needs to “know” the device. The device information should be listed in a central repository where it can easily be searched and extracted. Another important point is to have this information stored up-to-date at all times.

Most mobile device management software provide this kind of central repository. Some mobile portal products are also packed with this kind of information. If you are using a commercial and supported product, you can rely on the vendor to maintain this information. (It could be a nightmare to maintain this information by yourself.)

There is a nice, open-source initiative named Wurfl if you haven’t heard about it yet. You can reach it at here.

From this platform, you can download the latest mobile device repository in XML format and use it in your applications.

It also has a web interface where you can search a specific device.

After locating  the device you are looking for, you can reach information categories such as:

  • Product Information
  • Display Information
  • Markup Language/CSS/Ajax Support
  • Ajax Support
  • Playback/Streaming Protocol Support
  • Chips (NFC, GPS etc.)

Having and maintaining an up-to-date device information repository is a must for todays’ mobile telecommunication operator. Delivering a consistent customer experience in the mobile channel, effective support processes, effective device campaigns all require you to have this kind of repository.