In the first part of this blog two-parter we discussed who shouldn't embark on a Cloud Native transformation right now. In part two, we'll talk about who should.
The problem that the cloud helps solve for businesses is commercial growth. It’s a tool for a growth phase, not a cost cutting or consolidation phase. Is growth the number one item on your leadership team’s agenda? For example:
If so, then your company is trying to get bigger. That’s a good thing and fundamentally an investment. That’s the right mindset with which to approach the cloud.
So, how does the cloud support you to invest for growth?
Your company board will be familiar with the Joe Wanamaker quote, first coined nearly 100 years ago, “Half the money I spend on advertising is wasted; the trouble is I don't know which half”. Solving this problem by monitoring results has been a key goal of marketing ever since.
Fundamentally, it is pretty much impossible to predict exactly what’s going to work in marketing - what customers are going to respond to. Great ideas tank and “stupid” ones underpin every successful company. Good marketeers understand this. They set up lots of channels, try loads of things and monitor like crazy down to the smallest level to see what’s working. Then they tweak (aka evolve) their campaigns. They don’t usually run one, single shot, massively expensive campaign and cross their fingers.
But we do in tech.
Usually the biggest win for a cloud native transformation is making tech systems look more like marketing ones. We want to swap infrequent, single shot, massively expensive product releases (and crossed fingers) for small, fast delivery, iterative ones with high degrees of monitoring, measuring and response. In tech we call this feedback loops, in marketing it’s analysis.
This is what Cloud Native, i.e. automation, on-demand (cloud) infrastructure, microservices, continuous delivery, and monitoring, is for. The purpose is to get the same improvements we got in marketing for tech too. It’s about accepting that we can’t always predict in advance what’s going to work. We need to “suck it and see”: make small changes, analyse what’s successful and then keep improving. As my colleague Jamie says, when things are uncertain, the range of possible futures is too broad. The challenge is to narrow the range to the options that fit your needs as quickly as possible.
This works. Over five years, international newspaper The FT have used this technique to move from “big bang” releases every six months to highly monitored releases every 15 minutes. As a result, they became a pioneer of content paywalls and the first mainstream UK newspaper to report earning more from digital subscriptions than print sales.
The iterative techniques above don’t just work to improve what you have. They can also help you build from scratch - as long as you start small. The idea behind a minimum viable product or MVP is that iteration works, so you should start doing it as early as possible. That means getting bare-bones products in front of customers ASAP, then keep improving them based on real-world analysis of their performance. Google have three stages of public release before they even consider a product to be ready: early access, alpha and beta. If that’s good enough for Google, it’s good enough for us.
Many forms of business expansion involve trying an idea out to see if it works. That might be new sales regions or new technologies like devices or tools. Cloud services are designed to be cheap to start projects with. That doesn’t mean they aren’t expensive later - they often are. What it does mean is they are optimized for early experimentation and a good place to try out new markets. You can worry about cost cutting later when you have a much better understanding of the ROI (and have thus de-risked the project).
Again, it’s not cheap to hire machines on demand in the cloud. However, it’s faster, less risky, and less expensive than buying the maximum number you need for your peak load and then having them sit around idle for the rest of the year. Buy what you need when you need it and hand it back when you don’t.
Disaster Recovery, or DR, is the one exception to the growth rule. If your business has to maintain extremely high reliability for your systems, that is an area where the cloud excels and a DIY approach can be prohibitively expensive. Infrastructure from the major cloud providers is highly distributed and resilient by design. If that’s something you need it’s generally available more cheaply in the cloud.
In general, the cloud is an excellent long term growth investment for companies that are expanding. It’s also great for companies that need specialist high availability disaster recovery.
If you want to know if cloud is worth it for you, it is time to ask "are any of these true?":
If your company is battening down the hatches to get through a tough period, is already maxed out handling difficult projects, or has little desire or need to change, then now is not the right time for an attention-consuming digital transformation. Delay that for a year until the situation changes and then come back and read this again.
If you need to find ways to grow faster and bigger, if you understand the challenges and are willing to invest, then the cloud can help you and the sooner the better.
However, be warned, adopting Cloud Native is not the kind of project that will complete in six months with an easily projected cost and ROI. It is not a normal candidate for a technology business case. This is a far more wide ranging endeavour, which will move your whole company in new directions by de-risking ideas that you’d never previously have contemplated.
Are you ready?
If you think Cloud Native might be right for you and you want help with the business case, give us a call.
Photo by eberhard grossgasteiger on Unsplash
Want to learn more about Kubernetes? Check out our training course below: