Monthly Archives: August 2019

Digital Transformation -Stay Ahead Of The Curve

Digital transformation is real. If we do nothing it will result in what could be a digital stagnation. Transformation is not a final goal but a continuous process of improvement. The word digital was first used in the 15th century. It took a different meaning with the invention of computers and the first electronic digital systems in the early 1930s and 1940s. Digital transformation has been happening over the last 100 years. 

So, what does the digital transformation in the 21st century really mean?

The digital transformation at this time stands for how we gather data using modern technologies to gain insight in order to better serve and engage customers, employees, and businesses. We could extend that to serve humanity as a whole.

The undeniable fact is that we produce and gather more data than we are able to comprehend and analyze without the aid of machines. Is this good or bad? It depends on the quality of the data that we collect and the benefit that we derive out of it to overall serve people better. Excessive data collection and consumption could become an issue just like the plastic waste that we produce. People are losing their lives while trying to capture the very moment of their lives in the selfie culture.

Digital transformation consists of various components. Some of them are IoT, 5G, AI, blockchain, NLP, ML, big data, analytics and cloud native applications. The rest of the article will primarily focus on the software application aspect of this transformation.

Technology Leaders

Amazon, Apple, Google, Microsoft, and Facebook are continuously inventing to solve their business problems. What we call digital transformation is something the big companies have been doing for a long time. Digital transformation is a process and not a final product. Most of us are only catching up with the pioneers in the field. If we are too late to catch up, it could become the dreaded legacy that no one is proud of. As an example, Netflix OSS dropped their own tool Hystrix for Reslience4J. Another example is that of Google which stopped selling their Google Search Appliance (GSA) product in 2016. Many organizations that invested in GSA had to quickly transform their businesses to use Elastic, Solr or other proprietary search engines.

We need to understand that these companies are not trying to build frameworks for the sake of building frameworks. They are solving the business problem that they are facing. If the solution is irrelevant they move on.

Digital Transformation Failures

Many digital transformation projects fail according to a Forbes article in 2016. The same sentiment held true in a Harvard Business Review article in 2018 and even today among business leaders.

A major university in the east coast with tens of thousands of users spent three years implementing a digital transformation solution for their students. By the time they finished their product, they were already in a legacy system that approached the end of support and needed an upgrade. The choices that they were facing were either spend another year or more upgrading the system, or build a new system. The transformation product that they choose already became a legacy and not the right solution. Three years is a lot of time in the era of digital transformation. In that timeframe, anything we build could become a legacy.

It is inevitable that all organizations need to transform. It is a challenge that many are not able to succeed in.

Fortune magazine did a special investigation report in early 2019. According to that report more than $36 billion dollars were spent over a 10 year period to digitize health records. It was a major failure. It is unfortunate that some of those errors resulted in the loss of lives.

Digital Transformation Success

There is a centuries old proverb in Tamil which says “Even when throwing (spend money) in a river, measure what you throw.” It applies greatly in the age of digital transformation. Many organizations throw their money into the river hoping that their misfortune with technology will change. It does not work that way. Intuit  is one of the companies that has seen some great success with their digital transformation. A company that helps people keep their books does seem to know how to keep their books right.

There are a few things that companies can do to have a successful digital transformation.

• Usability. Anything that we build needs to intuitive. It should not require 4000 mouse clicks a day for a physician to perform a shift in ER. We can’t expect our users to be developers who know how to work around the system to get their job done.
• Accountability. Often teams blame others including the product they spent months choosing, or the vendor that they vetted through RFP process for the failure. We need to own the responsibility. 
• Measure as we spend. It is important to look at the returns whether it is short term or long term and be able to justify the cost. 
• Small steps. We need to learn to walk before we can learn to run. Identify some areas that can be transformed and make it a model for the company.
• Assess before adopting. A lot of the technologies out there are to some extent a fad or hype. Don’t be afraid to look under the hood. If you need help, choose a trustworthy partner. Organizations often choose the wrong product or platform and then try to find a partner who could help them transform.

Stay Ahead Of The Technology Curve

It is important that we always look at the problem that we are trying to solve and understand why we are doing what we are doing.

Many organizations may not have the necessary means to invent their own transformation tool. It is perfectly fine. Everyone doesn’t have to invent their own plane in order to fly in it. The key is to understand the purpose and limitations of the tools. 

REST API became very popular since the early 2000s. Along with that came the Javascript frameworks which made it easy to build applications. Teams started building apps that made dozens of request to update various sections of a single page. This soon became a major issue with all the overhead that comes with every single request. Solutions such as Websockets were proposed to reduce some of the overhead. Even though the data can travel at the speed of light, we share bandwidth just like roads and bridges. 

How did some organizations solve this problem? Some organizations built view optimized tables, caches and even relied on flat schema search engines such as Elastic to reduce the number of requests. Facebook solved this problem by building their own solution and called it GraphQL. 

Is GraphQL a final solution to all problems? It is definitely not. Now you have to scale the GraphQL server in a similar way you would have to scale any of your database and applications to support all the traffic. We just introduced one more layer to the problem.

What did AWS do with GraphQL? They found an opportunity to take this and make it into AWS AppSync. AppSync relieves the end users of the pain and effort of maintaining another layer. Organizations that are early adopters often may face the challenge of doing everything themselves. The companies that are at this juncture should evaluate which path to choose for a successful digital transformation. Should you spend years building you own or find one that saves you years of effort?

The same applies with Kubernetes. Within a short time of releasing Kubernetes, a whole bunch of companies popped up to tell you how Kubernetes can be made easy and painless so that you can focus on solving your business problems. 

Choose wisely a platform or a cloud solution that lets you focus on the business rather than building massive technology teams. At some point an organization may find itself at the crossroads of becoming the pioneer. If it happens, let it be so. Don’t be afraid to pick up the baton. Thirty years ago, some of the major players transforming the field right now were either too young or yet to be born.

Conclusion

We are in a time where a team may be relying on tools and contributions from Microsoft, Google, Amazon, Facebook or Netflix to build a single solution. It is perfectly fine to work with multiple technologies. We are solving business needs. If we keep our focus on the goal, the technology becomes as an asset rather than a liability. Digital transformation is not a goal but a continuous process.