Sorry, you need to enable JavaScript to visit this website.

Delivering Agile DevOps in a BI world

Delivering Agile DevOps in a BI world
June 28, 2017

Businesses are changing at an unprecedented rate, the current life of a company on the S&P500 is now just 15 years. With the downward trajectory of this tenure and the exponential growth of digital footprints, it is unlikely that this tenure will hold for long.

DevOps in


To stay competitive amidst global disruptors, enterprises are scrambling to harness strategic and tactical insight from their ever-growing data footprint. Corporate data reservoirs which now hold massive volumes of data and continue to grow at an exponential rate with emerging technologies such as the Internet of Things (IoT) and multi-cloud frameworks being adopted. Today’s world requires a digital platform that enables business agility. Business decisions in the digital age require radical shifts to operations – turning traditional business silos on their side, starting up new business units in a matter of weeks, with constant innovate, test, learn cycles to make better, more informed decisions and out maneuver the competition

T-Mobile is an excellent example of how a laggard in the industry radically transformed their organization to focus on a target market segment and become the “Uncarrier”. In order to do this, T-Mobile had to create a bank, eliminate contracts, and work with partners to provide free streaming data – in their “Binge On” campaign, and much, much more.  


No company can drive this level of transformation with conventional means – an unconventional way of thinking, on a next generation platform, with a massive focus on experimenting and rapidly introducing new services and market strategies. The results are dramatic with T-Mobile increasing market share from under 9% to over 15% in 4 years


The Need for a Next Generation Platform

Agility is of utmost importance across the value chain – yet, legacy platforms often fail to deliver the kind of agility demanded by the modern enterprises. Additionally, the BI revolution is rendering the traditional approach supported by legacy platforms inadequate.

Dynamic organizations have embraced the tectonic shift in BI and adopted next-generation platforms – promoting agility and enabling highly automated interactions. Features of this new architecture are:

  • Seamless management of new and non-traditional data sources using concepts such as data lakes, data virtualization, and data blending
  • Movement to Artificial Intelligence and Autonomous Decision Frameworks
  • Real-time streaming analysis
  • Enhanced data visualization & improved deployment options such as the cloud, mobile devices, and growing interest in augmented reality
  • Data Quality is always paramount, having a robust business led quality view on this data for transactional and master data sets

Enterprises are now compelled to generate immediate and accurate insights – yet, they must also contain IT spends while operating the new, data-driven architecture.

Running this Next Gen Platform in an Agile Model

A key benefit of the new data platforms is that there is inherent flexibility with a “structure on read” format (NoSQL) vs. a “structure on write” format (conventional data warehouse - SQL).  Many times, you do not know the analysis that you are looking for until you dig into the data, when you do find something interesting, or want to explore a specific aspect, you need to restructure the conventional data warehouse that is highly normalized with layers and keys.  In contrast with NoSQL databases, data is stored at its lowest level (it is denormalized), the structuring of the data occurs on the “read” and you only structure what you need for the read, not the entire data set – this allows a much more flexible and agile model for businesses to operate within – you no longer need to structure an entire date warehouse to get your answer, just the information you’re targeting.  The turn-around for requests is dramatically shorter as there are not any fundamental architecture or structure changes that occur with this approach.

Now we have the stage set for truly iterative development – with agile sprints on this new platform. The business can now be fully engaged – we typically see a 2-week sprint cycle as the most effective tempo to engage the business fully and allow focused development with the right level of feedback and connectivity with the business users.

Testing and deployment automation

Testing and deployment automation

In order to operate at a 2-week pace, the need for automated testing and deployment is essential, otherwise too much time would be consumed in these processes vs adding value. It is important to implement automation to the necessary processes to operate at this velocity, by doing this, the error rate decreases as the time on actual code and solution development increases making the team more effective and efficient overall. Automated testing gives testers fast feedback, allowing them to quickly pinpoint and redress problems and allow them to move towards Continuous Delivery.


With the very fast pace of development, there is now no longer a need to separate the development and support folks – in fact you want the developers to be in charge of the support for their releases, there are many benefits to this as they bake in components to the design and solution that make trouble shooting and support more effective. There is also a greater sense of ownership and quality with this approach.  Once again, automation is a key component to enabling this, moving folks as much as possible to self-service and self-diagnosis with easy to use tools – this makes the support more responsive and effective.

DevOps and Agile enable enterprises to out-learn and out-adapt their competition. Some of the key advantages of adapting to DevOps model are:

  • Building what the business needs: DevOps methodology allows innovation and iteration in close proximity to the business with frequent progress touch points
  • Early and repeatable delivery: companies realize results much faster than conventional methods
  • Timely error detection: automation plays a key role in accelerating testing and deployment. Being Agile leads to quicker resolutions – reducing costs, time and effort with an improved time to market
  • Collaboration among disparate business lines: Agile & DevOps methods ensure better working relationships among teams, bridging the business divide that results in silos and producing higher quality output
  • Greater transparency and visibility: Agile teams working on projects produce results consistent, coherent and regular results – allowing stakeholders to integrate business planning to maximize results

The way forward

To sustainably compete in today’s world, companies need a platform that facilitates agility – a next generation data platform, where they then need to execute programs in an Agile and DevOps manner.  With these tools, not only will enterprises be able to keep pace with the onslaught of data but they will be able to leverage it an out maneuver their current and future competitors.