By Gerardo Barranquero, CEO.
Both Forrester and Gartner tell us that customer experience will be the key differentiator, even over price, when deciding to buy goods and services. In the age of the customer, with disrupters around every corner, who will succeed and who will fail comes down to one simple concept: Code Wins. Period! This axiom is supported by two basic principles – trust nobody and trust nothing.
The “fast and buggy or slow and clean” dichotomy is a thing of the past.
Step 1: Code Wins: It’s that simple
The best idea, service or product does not translate into success in the same way it used to. If you are still talking about your disrupter whilst your competitor is going to market with theirs, you are already dead – you just don’t know it!
First mover advantage, coupled with the ability to quickly cut clean code, will always defeat a superior product delivered late to the market. But how do we develop quality code quickly? Historically, we have been taught you can deliver fast and buggy or slow and clean, but never fast and clean. So clearly we need to change our traditional approach by adopting two basic principles: trust nobody and trust nothing!
Step 2: Trust Nobody
As humans we all make mistakes, no matter how good we think we are. Businesses need to reduce as much of the human factor from our delivery pipeline as possible. We do this by creating an automated continuous integration, delivery and deployment capability. This automation extends from the developer’s machine right through to production environments supported by operations.
Continuous integration allows developers to continuously commit code changes to a single code base that is fully automation tested. As each code delta is small, bugs are easily identified which materially reduces our mean time to resolution. We can continuously deliver new functionality or bug fixes across multiple test environments and eventually continuously deploy the tested code to production ready for customers to use.
This capability means we can roll out new functionality or bug fixes in a timely manner with a low risk profile. You may recognise this approach by its popular name of CI/CD or DevOps.
Step 3: Trust Nothing
The recent #censusfail proved that, despite an effective DevOps delivery pipeline, we can severely impact the customer experience by trusting hardware. Just as we cannot trust humans not to make mistakes, we cannot trust hardware not to fail. We must build in redundancy and scalability into our delivery pipeline.
We need to let go of traditional views that treat hardware as pets: to be cared for, patched and nurtured. Hardware should be disposable, replaceable units leveraged across our solution as needed. Better still, we want someone else to manage it for us offsite, maintain it with patches as needed and have unlimited amounts to enable us to ramp up and down as needed without having to pay for wasteful excess capacity. You may recognise this approach by its popular name: cloud computing.
Think of it like this, it is much better to show up to a Formula 1 race with a car that is competitive and change the tyres on the first pit stop, the brakes on the second, the engine on the third and eventually have the fastest car going around than to show up with the fastest car one week after the race was run!
Contact Avocado to uplift your delivery pipeline, bringing together everything you need to operate a durable but scalable release management system. Because in the end Code Wins!