Never Let the Future Disturb You — My Journey to the Cloud

Never let the future disturb you. You will meet it with the same weapons of reason which today arm you against the present. — Marcus Aurelius

How do you stop the bleeding? That was the question we had to ask ourselves.

One of our business units had experienced a decline in revenue for many months and something had to be done. Membership numbers weren’t where we wanted them to be. Our customer experience needed improvement. Our site wasn’t as reliable as we would have liked. Releasing new features and products simply took too long.

This is the situation I experienced while leading global marketing technology and web development at a FTSE 100 company. I have been a technologist for more than 20 years (since before the World Wide Web) and felt that I had to do something, even if only to offer advice. I teamed up with a business executive and we began to research the root causes of these challenges.

Several things became clear:

1. While the business model had evolved over more than 10 years, the technology had not kept pace and was being used for things for which it had not originally been intended.

2. Business priorities precluded work to address mounting technical debt and growing technical drift.

3. Complexities in the platform and technical debt caused unexpected problems during development and resulted in bugs, outages, and unplanned work.

4. To combat these mounting pressures, the business invested in a large QA effort, which added days or weeks to the already extended release schedule.

5. In combination, these realities severely hampered the business’ ability to deliver. New products required months of development and testing and limited the business to making two or three product bets per year.

The system had been very profitable for years and was highly optimized in many areas. We considered refactoring the existing platform, though we quickly realized that it would take several years of concerted effort to effectively evolve the legacy systems. Given competing priorities and budget limitations, this was simply unrealistic.

Instead, we proposed a greenfield project to develop a cloud-native platform leveraging Amazon Web Services (AWS). This would provide faster time-to-market for new products and features while meeting existing requirements. Initially, the business unit executive team and the global CISO received our proposal with trepidation. So we took great care to clarify assumptions, to explain capabilities, and to mitigate risks. Our leadership paid off and the project was approved.

To limit the risk to the business, we created a small cross-functional team to develop a functional prototype that would demonstrate the effectiveness of this new approach. We received funding for 90 days and a (mostly) free hand to make rapid decisions within the constraints of compliance and security.

90 days later, the 13 members of our team demonstrated the prototype. Not only had we met the business requirements, but we could now make and release simple feature changes and bug fixes within minutes. We had developed self-service capabilities for the business that previously required IT involvement, and we had demonstrated that we could conceive, develop, and release new product features within days or weeks, instead of the typical weeks or months. We could also scale on demand without having to rely on over-provisioned data centers for the first time.

After this initial success, we received additional funding and an aggressive timeline to expand our services and capabilities. Our roadmap primarily focused on user stories for customer acquisition, fulfillment, customer support, and sustained engagement. We also wanted to create additional self-service capabilities that would enable the business to directly manage the product experience. And, finally, we wanted to expand the platform based on DevOps principles to enable developers to better develop, maintain, and enhance products and features.

We learned by doing — none of us had had significant prior experience with AWS. Members of the original team eventually formed a permanent platform team with the objective of creating self-service infrastructure tools that included systems to build, test, deploy, scale, and maintain applications. Within months of deploying the prototype, this team was getting requests for help from other business units in multiple geographic regions, and it’s now well on its way to becoming a cloud center of excellence (CCoE).

We also gained ancillary benefits that addressed internal needs. We realized, for instance, that we now had a more fine-grained understanding of the cost of developing and operating our product than had previously been possible. We could use metrics to independently optimize each service for performance and cost. We made improvements to security operations, so that findings identified by the InfoSec team could be remedied in a more repeatable automated form. Business operations, too, could be optimized using an extensible framework that made it easy to instrument and automate processes. Finally, technology changes created an opportunity to explore ways to improve communications, gain greater engagement in our agile practice, and invigorate the culture by emphasizing innovation and experimentation.

Our foray into the cloud was a success and established a navigational chart loosely based on four stages of adoption for other teams and business units to follow and improve upon. The journey was not easy and there are many things I wish we had known beforehand. However, over the course of those two years, I became thoroughly convinced of the strategic importance of the Journey to the Cloud. I consider Amazon Web Services to be THE force-multiplier that gave us the ability to accomplish our “overly aggressive” or “impossible” goals, as some stakeholders had called them.

Which is why I jumped on the opportunity to join Amazon Web Services as Enterprise Evangelist for Europe, the Middle East, and Africa. I grew up in Germany, where I attended high school, and now, more than a quarter century later, I look forward to returning to my European roots. What gets me most fired up is how to evolve processes and technology to unlock unrealized potential — whether that’s to grow the bottom line or make life easier for customers and employees. In the past, we were locked into IT decisions and investments for years before we’d get an opportunity to revisit a solution and make it better and/or cheaper; but, today, with the emergence of cloud services, we have an opportunity to constantly improve, or even re-invent, our businesses. AWS was transformational in helping us provide superior services to our customers. Now I would like to help you do the same.

So check in from time to time and let me know how your journey is unfolding. What can AWS do to help you realize your vision? What would have to be true for you to take your enterprise to the cloud?

Until next time,

-Thomas
thoblood@amazon.com
@groberstiefel
http://aws.amazon.com/enterprise/