Thinking aloud

One morning I stepped into the development centre of a large software company which has just proudly transformed itself from an old age, obsolete Waterfall software delivery model to a dynamic, nextGen Agile model. The company takes a huge pride in this transformation – it is there in its new brochure, on the intranet and is the talk of the town! So, all excited, with coffee in my hand and huge anticipation in heart, I started my journey to understand this methodology and learn from this huge transformation -

“As I was stepping inside, I saw the Project Manager (who still doesn’t like to be called a Scrum Master) walking out of a two hour long meeting with a team of 15 engineers and, while walking into his office room, tells his team “Please be on time for tomorrow’s daily scrum meeting”

He proudly told me that the engineers of the scrum team are working close to 12hrs a day as the deadline of the sprint is only two weeks away and they need to deliver a workable product by then. I asked “Two weeks away? What is the duration of the sprint then”, I got the answer that the sprint is actually eight weeks long!! The team was also asked to create a suite of documents which can be shipped to the professional services team as a baseline for their customer facing documents, hence two members of the team were burning their midnight oil to get these lengthy documents in shape.

Finally, the spring deadline date arrived but there was no sprint review meeting, Why? The deadline was not met and hence it was conveniently delayed by four working days! Four days later, finally, sprint was finished and hence there was the sprint review meeting (sigh!). But there was no product owner, no customer and the participants were same as those in the daily scrum meeting: the team and the project manager!! Where is the product owner, where is the customer? Later I learnt that the delivery was dependent on some other module which was not to be ready until next month, so workable delivery was workable only on paper!

Next day started with a six hour long sprint planning meeting to decide the next set of user stories — which were strangely just the copy-paste of the requirements from the requirement document (also conveniently called as product backlog by our agile team). Strangely 50% of the user stories picked were those which were from the last sprint as they are still not completely incorporated then.

Bit dissapointed, I decided to move around the office to have a quick chat with the team members but “Where is the team?”  I got to know that this sprint team comprised members coming from different groups within the company and hence is spread in two separate buildings and three floors (?). These members communicated through emails and met once every morning for two hrs for the scrum meeting. A quick peep on the daily scrum meeting revealed that each member of the team started by describing, in detail, the bugs he was working on, explained to the whole team how many lines he has coded and how many test cases he has done, when his module is supposed to be ready plus any other social thing he wanted to discuss (?).  Also, to my surprise, all the regression test cases in every sprint were done manually and there was a dedicated team for that.

Finally, I decided to look for artifacts but bit surprised to see no metaphor, backlog graphs etc. After inquiring a bit about it I got to know that instead, the each member created his own low level design document, class design document, unit test document and the team together created interface spec, high level design document and architecture document with every sprint release”

Perplexed with my understanding of Agile methodology, I drove back home still not able to get my head around it and it was not until I opened the dictionary looking for word Agile, I got my answer. I found following antonyms of word – clumsy, stiff, inept, lazy etc and realized “Why would anyone not like to be called Agile even when one is not really agile”? After all it means dynamic, nimble, fast … and, above all, is the buzz word in the new generation software industry. It has spread like a wildfire and companies are embracing it with every passing day, but wait a minute, are they really embracing it? Are they actually following the key tenets of Agile methodology which takes its pride in key concepts such as team collaboration, 40-hr week, open communication, continuous integration etc?

It is very easy to adopt Agile methodology but equally hard to follow it. Let’s stop fooling ourselves and introspect are we really Agile? Companies should regularly ask the most important question – “Are we really following the Agile manifesto (http://agilemanifesto.org/) or is it forgotten in the midst of our daily work”? By simply using the terms such as sprint, scrum, product backlog, product owner, scrum master, sprint review, backlog chart, TDD etc etc, one cannot become Agile

Let’s recall the manifesto again and promise together we will follow it religiously -

We are uncovering better ways of developing software by doing it and helping others do it. Through this work we have come to value:

1. Individuals and interactions over processes and tools
2. Working software over comprehensive documentation
3. Customer collaboration over contract negotiation
4. Responding to change over following a plan

That is, while there is value in the items on the right, we value the items on the left more.

From left to right – Amit Limaye, Saujanya Shrivastava, Karthikeyan Rajasekharan and Ali Almossawi

On March 11th 2011, we SDM’11 students (Amit, Ali, Karthik and Saujanya), presented our idea of Samsung’s future platform strategy to its executive who traveled from Seoul looking for innovative solutions. Our proposal was of a truly connected home, which can take Samsung towards a truly converged home appliances provider. After doing our research on Samsung’s growth strategy, its key products and portfolio, its future ambitions, we felt that it needs to capitalise on its strong vertical integration of products and the breadth of portfolio and use these as instruments future growth.

Below is a snippet of our solution, Towards a truly connected home!

• Home device landscape

Before solving the problem, we tried to find the current status quo and for this we started looking into current device landscape in a typical home. We divided all the home appliances into four broad categories based on the internet connectivity and interoperability. The former suggests the appliances which are capable of low/high internet connectivity such as washing machine (low) or laptops (high). The latter are the devices which can interact with each other, such as exchanging media or transfering data, an example of such devices will be refrigerators (low) or cameras (high).  Our vision was to have something that had both high level of connectivity and equally good interoperability – suite of devices which can talk in our next generation connected home

• Scenario

A typical scenario would be TV getting mute as soon as phone rings (and back on voice when the call is completed) or washing machine sending message to the TV when wash cycle is completed.

• Architecture

How we can achieve this? Our vision of architecture is depicted below

We proposed Samsung to create a common protocol which all the devices can share and using which they can talk. Also, we didn’t want all the devices (or even app developers) to know each and every bit of this protocol, so get an abstraction layer in between. This layer will also be used by the app developers to develop innovative apps which can be uploaded on Samsung App store (controlled by Samsung). Users can download these apps and load them on their devices. Also, we proposed a concept of master and slave devices: master is the one having high processing capabilities such as smart TV or laptop which will control the devices in connected home while slave will be the thin processing device (such as washing machine) just taking commands and sending responses. This will prevent incorporating unnecessary complexities on the slave devices.

• Our Proposal

1. Samsung should provide a subsidized product development environment for app developers
2. Samsung should seed this marketplace with useful apps
3. Samsung should also develop an abstraction layer for each family of devices
4. Samsung should provide the apps with a familiar and consistent interface for each of the above device
5. Samsung should simplify its product families
6. Samsung should leverage on user centric innovation through open specification
7. Samsung should encourage third party devices to adopt Samsung’s platform

• Evolution

This platform cannot be evolved overnight and will require some strategic decision making from Samsung. We proposed to start by identifying different types of users

1. Active users – The lead users who will drive the innovation forward
2. Aspiring users – The users who are not the first one to use but not way behind either
3. Early trend followers – The users who take the technology early and start using them. These are the first buyers of technology from the market
4. Trailing users -The users who are circumspect and generally late in using the technology
5. The laggards – and finally the laggards, the ones who despice the new technology.

Our proposal is to target the lead users, early followers and let them drive the market place. We intend to create a diffusion affect in the consumer’s home so that he will be more propelled to buy Samsung appliances. This will not only increase the market share of Samsung products but also help it to stand apart from the current home applicances providers.

Last but not the least, it was a great experience to showcase our learnings of different subjects taught during spring semester of SDM course. User centric innovation, Technology strategy, Marketing management and above all the Systems thinking concepts helped us to carve a solution for the innovators of Samsung. We are only 3 months into the SDM program and are already ready to provide innovative ideas to leading technical firms. This truly stands with key mission of SDM program: to inculcate leadership, innovative ideas and systems thinking to its students.

Watch this space…

Over the years, communication service providers have invested huge amount of money to leverage on the increasing customer demand. Communications industry is one of the most competitive industries in today’s era and the need for innovation is almost inevitable for its survival. It needs to think about innovative ways to increase its ARPU (Average Revenue per User) and think about ways to reduce OPEX (Operating Expenditure) and CAPEX (Capital Expenditure) to maximize the profits. It needs to do a deep introspection to find out best ways to use its existing resources and see how it can use them optimally to gain profits.

Data by far is one of the most important assets of any communication provider and it is imperative for the success of these providers to use this data intelligently. Most of these communication providers (like mobile operators) are sitting on huge pile of subscriber data which, if analyzed properly, can be sufficient to completely define the behaviour, e.g. eating habits, hobbies, education etc, of the end customer. It is up to these companies to leverage this data to their full capacity and derive useful information from it.

Hence, a mechanism is needed to transform this dead, raw and practically useless data to live, transformed and useful information. This mechanism should help in strategically storing and transforming this raw data into business intelligence information which could be used to research and development of new services into the network. Looking this problem from the lens of systems thinking - an approach to problem solving by considering the complete ecosystem as a whole rather than reacting to individual elements – lead us towards a key value proposition, strategic data consolidation.

Strategic Data Consolidation is a technique in which telecom operators can store their data at a centralized location and then develop different views to provide the data access to different elements in their network.

Historically, most of the operators just use their network infrastructure as a “dump-pipe” (rather than an “intelligent pipe”) to transmit the data and never bothered to look introspect to see how they could use this data to their advantage. For e.g. most of the operators have Equipment identification number (IMEI, a unique id which can tell the mobile handset make, brand etc) as well they also store subscriber’s complete profile. However, both of these data are stored at different places and just lie there in silos without interacting with each other. If both of them can be intelligently consolidated, the operator can get sight on the capability of the handset of the subscriber and then can advice him of particular services which he can use. This will help the operator to generate more revenue from the customer and hence increase the overall ARPU. This is just one of the examples in which data consolidation can help the mobile operators and there are many such use-cases.

Various telecommunication standardizations institutes have worked on standards which can define how data can be effectively stored and accessed. Also, they give guidance on how data can be selectively fetched with the concept of views. One example of this is ITU-T Technical Recommendation X.500 which talks about the LDAP (Light-weight Directory Access Protocol) and DAP (Directory Access Protocol) which can be used for the hierarchal database architecture and provides a set of recommendations of data access and recovery.

Also, various data mining techniques can be used to transform data into intelligent information. One of them which is quite effective in this case is ETL – Extract, Transform and Load. This can be used by operators by extracting the distributed data, transforming it to useable information and then loading it at the centralized location for getting information out of it.

I will like to conclude by presenting a hypothetical (but not unrealistic) case. This will provide more insight on the importants of Strategic data consolidation

Two major European mobile operators (MyFone and OurFone ), both having customer base of more than 20 million subscribers, decided to launch their 4G (LTE) business to provide high speed data access to the customers. These are pan country operators and as the cost of new infrastructure is very huge, both want to start the operation from the region where higher end subscriber density is high. This is a practical decision as it can be realistically assumed that these subscribers will be having high end data devices and hence the return on investment will be fast if the service is launched in this region. Both devised a high level strategic team to do the above analysis and asked them to provide the results as soon as possible.

MyFone has a complex distributed data storage system, with data spread across the country. The team asked the managers of these data centres to fetch the raw data and send it across to a central location. Once the data arrived at the central location, another person wrote tools/scripts to transform it into something which can be read and load it into central database and then finally third person executed database queries to get the business useful information out of it. Due to huge legacy data (in terabytes), it took 2 months to get it out of the data centres, another month to transform and load it and then another 2 months read and understand it. In summary, to take the business critical decision of launching the service the total lead time for MyFone company was about 5 months.

On the other hand, OurFone’s data was consolidated and centralised already. They have already transformed the data on which business intelligence search can easily be done (and was already ongoing). They wrote a set of new tools as per the requirement and got the result in 15 days!

With the competitive nature of mobile telecom industry, by the time MyFone’s 4G services were launched (i.e. 4.5 months after the launch of OurFone’s services), OurFone has already captured the high end customer market and its revenue has increased manifold. Apart from revenue, the overall image of OurFone has now grown and it was started to be considered as a cutting edge leading mobile network operator which brings service to market before anyone else!

The above shows the importance of strategic data consolidation.