SaaS Infrastructure Automation Considerations
I recently read an article trying to clear up the difference between foundation settling and subsidence. It goes without saying: One is OK, and the other is really, really bad.
Early on in a building’s life, it is often difficult to tell what is settling and what is subsidence. Although settlement is often irritating and sometimes troubling, the adjustment period is temporary and will end. While it may look the same at the start, subsidence lets you know if your foundation cannot support the structure and if it will ultimately fail. The article did a good job of providing guidance on how to tell which you have, but I got to thinking: I would rather know how to build a strong foundation from the start.
SaaS and PaaS delivery is similar in that regard. The SaaS product strategy needs to support the overall business strategy and, in turn, the foundational SaaS infrastructure strategy needs to support the overall business strategies. Without a good foundation, the entire structure is at risk.
SaaS business strategies are typically dominated by a couple of basic principles:
Increase customer lifetime value
Key to customer lifetime value (CLV) is customer retention. Customers that use a SaaS product regularly (and, ideally, habitually) are less likely to leave. With frequent usage, customers become more invested, and the product becomes a part of their daily routines. Frequent and habitual customer usage results from great user experiences: Building solutions that deliver fresh experiences. Delivering new functionality to customers regularly. Keeping a focus on the rapid translation of feedback, and then translating that insight into functionality for an improved customer experience.
To deliver on these guiding principles, many organizations are looking to agile, continuous integration and deployment (CI/CD) methodologies. But to succeed at it means coming to the realization that the tools, methodologies, and code are only as good as the infrastructure at their core.
SaaS and other organizations see the benefit of CI/CD to help them quickly deliver, but where many run into issues with their CI/CD transformation is this first step: defining and implementing the infrastructure required for success.
Because CI/CD requires a core commitment to automation in the technical processes, operational culture, and organizational thinking, this early step can feel like boiling the ocean to get started with large investments in infrastructure design, deployment, and staff training before being able to realize any benefit. This is partially due to the fact that implementing automated infrastructure has traditionally involved either: 1.) adopting traditional infrastructure components to suit a purpose outside of their original intent or 2.) building a custom solution from the ground up. In both cases, a large portion of the burden has fallen on the development engineering organization tasked to focus in other areas or shorthanded IT organizations.
One piece of good news is that automated infrastructures designed specifically for CI/CD are readily available and easily consumable. There’s also a thriving community with a wealth of information and experiences for deploying an automated infrastructure to support CI/CD, both organization-wide as well as departmentally. So whether you are just kicking the tires or going all in, there is no need to work without a net.
With the the advances in infrastructure scaling and automation, getting to a production CI/CD model is feasible for more and more organizations to deliver on the following promises:
Engage better with customers through delivery of responsive and consistent performance, customizable user-defined experiences, and delivery of new capabilities without outages — building trust and yielding greater customer satisfaction.
Reduce customer acquisition cost (CAC) by delivering functionality that includes easy, self-service sign-up; a 24-hour service day; and premium upgrades, allowing customers to decide when, where, how, and if to communicate.
This provides the vehicle and encouragement for customers to share and invite, turning contacts into customers and customers into advocates that drive additional customers.
Utilize a model that scales much faster than staffing requirements grow. From 2010 to 2020, the data managed by enterprise IT organizations is expected to grow 50x, while IT head count will only grow by 1.5x.
Keep costs in line with revenue. Traditionally when deploying, infrastructure organizations try to look far into the future and anticipate upcoming growth and capability needs. This inherently mean over-purchasing; on the front end, tying up space and capital with hopes of growing into it. Instead, try to select and design infrastructure that facilitates nondisruptive performance and capacity growth, eliminating the need to purchase long in advance. For more on developing a sound SaaS infrastructure strategy to support overall business strategy, download Automation Considerations for your SaaS infrastructure and be sure to sign up for your SaaS with More Cowbell kit.
| 0 comments
Posted in Agile Infrastructure.