Let’s face it – the process of selecting the right technology to address next-generation solutions like enterprise mobility, BYOD, or big data, is exciting for sure but also implies undertaking a massive research project. Researching the optimum solution needs to be strategic to one’s organizational requirements as well as very tactical to ensure faster adoption right from the beginning. It is required, by design, that the technology stack of choice not only solves the current pain point but can also be scalable.

It is not uncommon in this research project for the core team members to sweat the small stuff and find themselves as victims of what we call “the whiteboard effect.

It essentially means that the team spends a vast amount of time, going in spirals and over-analyzing the problem, before drawing that first line on a clean board. Interestingly, there are boatloads of vendors, and counting, in the market today. There is also an immense amount of marketing data and statistics of the few that ventured down this path and how their sales and savings skyrocketed.

After having established the numerous distractions that exist today in hindering the optimum technology selection process, here is a suggested framework that can guide you to achieve the desired goal of enterprise mobility and beyond.

We typically, begin this exercise by deconstructing the problem and then build a solution framework in a broader sense that could apply to multiple situations.

Problem Deconstruct

The first step to attain the right solution is to identify and document a well-defined problem statement. The definition of a problem statement in an emerging technology should, in our opinion, always include in a succinct paragraph:

• Business problem
• Ideal success criteria

For example, here is a problem statement from an organization we worked with and that we shall use in this article:

“We are a retail business, and our sales reps are on the field making decisions out of few months’ old sales data and gut feel. For deal specific data, our reps have to be at their desk for secure access. There is no real-time data to support a decision on the field. Our reports are generated every few weeks and are at the mercy of our reporting software speed. Ideally, we would like our sales team to be mobile and selling products and not tied to their desks. We want our workforce to be able to access data that is closer to real time and drill down as needed and make a data-driven decision.”

These statements are not very far from a typical enterprise mobility problem in the global market. On further analysis, this business requirement can be broken down into:

• The right device for the right situation
• Secure access to corporate data
• Visualization application
• Fast reporting system

These result in a list of technology bullets for the CIO/ CTO:

•  Device platform strategy
• Enterprise connectivity
• Wireless strategy
• Carrier-based strategy
• Enterprise boundary
• Asset management
• Process support
• Expense management
• Mobile work for cloud strategy

They are all valid concerns. However, it is interesting to note how we block ourselves from moving towards the next steps and end-up creating a laundry list of action items before even addressing the real problem at hand.

In our experience, any business-driven initiative in an emerging space should be agile and focus on fulfilling basic requirements, rather than focusing on strategic technology initiatives from the first day. Refactoring and reconsidering a solution during implementation is normal and expected in an emerging technology space.

On the other hand, the CTO’s list of items leading to a mobility roadmap is a valid approach for a technology-driven initiative. In this article, we focus on the business-driven solution building only.

Solution Building

After analyzing the problem and deriving a set of focused and complete business requirements, we figured out a truly innovative, one-of-its-kind, scalable, and simple to understand the solution.

The result was to build a data aggregation solution, based Extract-Load-Transform (ELT) concept, which could stream data rapidly from all possible data sources of the organization:

• Social media
• Web analytics
• Sales data marts
• ERP
• POS Data

The entire data set was streamed into a repository, normalized, transformed, and further loaded into a massively parallel database cluster. We hooked up a visualization platform for advanced analytics at the very end. This architecture is not an invention, but one that is well known and been around for a while. However, a large number of enterprises even today do not have it implemented but are instead still mulling over the next steps.

This architecture is unique and low-cost with a virtualized footprint. It is one of the first few “Big Data Lake” cluster on a virtualized cloud environment, providing advanced analytics for drilling down to the right device at the right time.

Additionally, a Role-based-access (RBAC) model was added to create a multi-tenant environment. It enabled each department to leverage their BI team to analyze data and direct it towards their respective department visualization dashboards.

There may be performance questions on the architecture. However, it is not essential to notice that the solution needs to be optimized to meet the requirements, but it need not break the bank. More importantly, it needs to be easy to recreate and scale up as and when requirements grow.

To succeed, the solution, especially in emerging technology space, should at a minimum, always resolve the pain point and increase user adoption.

Another subtle yet essential aspect of this solution is the – Apps > Data > Analytics lifecycle. In this case, within a few weeks of implementation, the team was tasked to build a mobile app driven by the data in the big data lake. The next step is to go about this were in line with the agile software development strategy utilizing Micro-services and Cloud Foundry for rapid development and deployment.

In essence, not all environments are the same, and each of you might have a different set of environments, requirements, and concerns. However, if an architected solution is revenue-generating rather than cost-saving and quick-to-operationalize, rest assured that there will be a lot of innovative and exciting work for the entire enterprise.