Everything is more important than IT infrastructure. Or is it?


When you start a new business, almost everything seems to be more important than the IT infrastructure that your product will run on. It's the least of the new CEO's concerns. “We will just buy a bunch of servers and our IT guys will take care of everything else”- he thinks. “First I have to take care of my target audience, frontend and UX of the app, graphic design, contact network, marketing, hiring… Oh, and let's not forget about great coffee and a ping-pong table”

And then suddenly a huge f*ck up happens.

The application crashes during traffic peaks. Becomes unavailable.

At this point, it won't matter how strong your brand is or how elaborate the app's design is.

Your customers are upset, and would rather spend their money with your competitor. Your revenue rapidly goes down. Not to mention that the whole IT department is on the verge of a mental breakdown. Looks like these few servers were not enough in the end.

But all this could have been avoided. You can prevent this from happening to your business. You can even squeeze more revenue out of a properly functioning infrastructure.

I am not saying, of course, that all the above-mentioned things such as the design or marketing are not important. They are. But I want to show you that you should not take infrastructure issues lightly and pay at least as much attention to it as anything else.

You might be thinking:

“But I have people for that. Why should I understand how the backend of my digital products works? That's why I hire developers!.”

This is, of course, true but you are the one making business decisions.

Knowing what is happening behind the scenes and understanding “what” your products run on (or could be running on) will make your decisions so much easier.

  • It will be easier for you to be able to tell if someone tries to pull your leg.

  • You'll be able to make well-informed decisions on what type of technology will serve your needs best.

  • You'll be able to react to the changes in market dynamics and customer's needs quickly.

  • But most of all, you will have an overview of the investments in your IT and how much sense they make.

Once you stop seeing the IT infrastructure as “a necessary evil” and gain a better understanding of it, you'll achieve much better business results.

Let's take a look at some examples of the most common challenges that companies selling a digital product face.

Slow time to market

You are entering new functional requirements, but developers are unable to implement them at all or are implementing them too slowly. This, of course, annoys you, because there is a risk that your competition will be able to offer a better product sooner and you will lose your customers.

Solution? Microservices.

If you split your application into multiple building blocks, the so-called Microservices, you can split your development team into smaller agile groups. Each of them can work on a different part of the application and will not have to wait for the others. Individual parts of the application will create independent parts, so there is no need to synchronize much. In addition, if you properly use containers and Kubernetes, you can easily deploy hundreds of updates to your application every day and always deliver the best possible product to your clients.

The app struggles or crashes completely

Your application becomes unavailable often.

Your service often crashes and struggles. Whatever the reason (and that there may be a lot of them) the customer is absolutely not interested. When the service doesn't work, he/she goes elsewhere.

Solution: Microservices approach + Kubernetes

The most common reason for the application malfunction is the release of a new (faulty) version. Combining the Kubernetes and Microservices approach will help you to respond to such errors very quickly and roll back the backed-up version. Another common reason for an application crash is a sudden increase in the number of users of your product. With Kubernetes you can automatically “inflate” or “reduce” your computing power according to the demand.

The operation is expensive

Operation of proper infrastructure can be quite expensive. Not only do you have to buy enough servers, some of which you use only at the time of the peak (and even those are sometimes not enough), but you also have to cover the costs of renting the building where they are stored, their maintenance and the workforce who takes care of everything.

Solution: Scalable, cloud-based infrastructure (with containers and Kubernetes)

The problem with high-cost won't be solved by Kubernetes only. However, using these two technologies combined can help you maximize the use of cloud infrastructure (but in some cases, it could be an on-premises solution too). Pay-as-you-go model that is common for the Cloud. You only pay for the computing power you are actually using, you always have enough computing power and you no “empty” servers lying around when you only need extra power for the peak times.

My team is not big enough

One of the biggest problems companies face nowadays is the so-called “operations overhead” in their IT departments where folks don't have time to manage work on a product because they keep extinguishing fires in operations.

Solution? Again microservices, containers and Kubernetes

A well-designed infrastructure built around microservices, containers and Kubernetes can be managed by a very small team. Because these technologies offer a great deal of automation, you don't need dozens of specialists just to keep your application alive. A small team is enough - automation for the win!

Because Kubernetes and microservices approach help you to minimize downtime, you can use your best people to develop your product instead of wasting their potential on fighting fires. Your developers will not be hampered by poorly functioning infrastructure and will be able to deliver their work quickly.