Have you noticed lately how, despite technology advances and a huge number of connected cloud services, you aren’t reducing costs in digital? What if I told you that there was a way to reduce what you spend on digital by a third? Would you take five minutes to find out what it is?
In the early years of the web, every man-and-his-dog could publish a web page. Web publishing was easy and as simple as sticking a file on a server. We had HTML, CSS and FTP and that was all we needed. The barrier to entry was low, because the technology was simple.
Fast-forward to today and the landscape is unrecognisable: the technology required to power a high-performance website in today’s world requires a highly-skilled, always-up-to-date workforce with knowledge across a stack of complex, highly-tunable services that are not only expensive to manage but require your business to be competent in skill set’s from infrastructure management, to configuration management, to network technologies, to Agile delivery methods. It’s not surprising that faced with this challenge most business either spend an awful lot of money or simply fall behind (and sometimes both).
But it turns out there’s a way out of this conundrum, and the secret is working less.
The 2016 Puppet State of DevOps Report surveyed over 4500 developers and drew some very interesting conclusions about how your development and deployment processes affect your ability to deliver. Puppet wanted to find out what makes a high-performing development team, and what they found was very interesting. Development teams who were truly deploying quickly (e.g. once or more per day) – and thus in smaller increments – had lower failure rates, faster recovery times, and spent more time working on new features. For teams deploying once a week or less, recovery times shot up and failure rates were anywhere up to 45x greater.
Forty. Five. Times. More. Bugs.
Take a moment to consider that. By deploying quicker and in smaller increments, the average team reduced its rework rate by a factor of 45. Think about what you could do if you even achieved a fraction of that?
This practice of building in smaller increments forces you to build quality into each increment, a practice known as Shifting Left.
In a traditional development model, we move from left to right across design, development, testing and production. Shifting Left means moving activities which were previously on the right over to the left of that process, in order to bring them together. It is often focused on when, how and who does testing, and what they test, as the aim is to test the right things early enough to make a difference, and in increments that are manageable.
There are a few things you can do to shift your development process left, however moving to a Continuous Integration and Deployment model is one of the key ways a development team can make this happen. However, that comes with its own challenges: suddenly you also need to be an expert in complex toolsets and infrastructure management. This is not a simple or easy thing to do.
I am blessed to be working with some pretty damn smart people at Platform.sh that are trying to solve these issues day in and day out. They are obsessed with how to get all of these gains while keeping stuff simple and lightweight.
What this bunch came up with is a hosting environment that completely removes the need even to know what DevOps means, let alone understand how to configure a server cluster, scale a database, or seamlessly upscale on Black Friday. Their system enables developers to deploy automatically by linking environments to their version control commits, and automatically configures and deploys complete Enterprise web services based on a simple, human-readable configuration file.
This enables your team to focus on delivering quickly, iterating fast, and getting some of the benefits of a shifting left, without worrying about whether or not your next release is going to fail, or take too long, or be to complicated to deploy, or any other number of headaches which have simply gone away. It makes shifting left much, much easier.
It turns out there is a pretty big payoff from moving quickly. We regularly ask our customers how they are going and they are reporting significant cost savings as a result of switching to Platform.sh. I’m not talking about the cost of their web hosting (though we’re pretty competitive there too) but the savings across their overall digital budget as they “shift left”.
On top of reduced errors, greater customer satisfaction, near zero downtime and faster time-to-market, Platform.sh customers report average cost savings of 30%. And they never needed to learn how to deploy a thing.
Even my dog could do that.
Are you interested in finding out more about how Platform.sh can make your life easier? Get in touch with us today.
Christopher Skene is Platform.sh’s outreach lead in the Asia-Pacific region. With over a decade’s experience in project delivery and technical roles, Chris knows the day-to-day challenges faced by companies delivering digital projects. Contact Chris…