Introduction - Moving to Cloud, but with a real Strategy — part 1
Today, I want to take you with me down the rabbit hole to dive into some examples of the most popular Cloud Transformation strategies, which are no strategies at all. I will share the toughest challenges you will face throughout your journey and how you can overcome them.
In this first part, I will explain how a lousy strategy and lacking vision will very likely directly lead you in a world of pain. This part is undoubtedly generalizing some topics, but does, unfortunately, contain more truth than most managers would like to believe.
Don´t also miss the upcoming parts addressing topics around Cloud and preconditions to transform your company. I am definitely pro Cloud, but as with every technology, you need to make proper use of it.
In my past years, I have been working with several Enterprise customers from different sectors. And — you all know that there is currently an intense hype about Digital Transformation. We all have been hearing from many customers and companies severely struggling with the Transformation, but I want to raise a provocative question here — Isn´t this obvious?
A transformation would not just be a transformation if you could reach it by taking risk-averse incremental steps without having any disruption. And, as many executives think they can break this simple rule, you can see a lot of “interesting” strategies, which create even worse results.
The easiest way to fail in a Transformation is to treat it as an incremental change.
Cloud as one solution
The migration to the Cloud is heavily being used as a driver to push the internal Digital Transformation. (Reference link) Companies start evaluating and moving their existent infrastructure, the applications, and data to the Cloud to keep up with the pace of Start-ups and strong competitors.
But why are you moving to the Cloud?
For every strategy, a good argumentation should be in place, well-documented, and communicated — throughout your company — today more than ever before. And even more important for your company — you should have a vision in mind and share it with everyone:
- Where do you want to be in two, three, five, and ten years?
- How do you want to improve your way of working and your culture?
- What value do you want to provide to your customers, but also to your employees and partners?
- How do you want to drive and change your business in the next years?
- What is your plan to motivate your employees to accompany you on this journey?
- And how can IT help you with this significant change?
And — in reality:
“We want to reduce costs.” — Ah, really? Who doesn´t!
Frequently, I see CTOs and CIOs having a strong influence by their CFOs, or speaking more general: IT is just being taken as a service provider delivering the same value continuously, and the only room for improvement lies in decreasing costs. (Reference link)
As a result, customers usually start with Lift & Shift approaches into the Cloud, moving dumb IaaS infrastructure from left to right. (Reference link) The argumentation for such movements is very often explained in a reduction of costs and are being sold with similar diagrams like the following:
And? Is it as easy as this?
Well done, salespeople — well done ;)
Surely not, and it is usually a sign for an unmatured Cloud strategy. But let us now dive into the different facets of this simple topic about TCO.
Total Cost of Ownership
The sum of all direct and indirect costs of the IT estate including all application development, maintenance and support, operations, data center, network, and BC/DR.
Interestingly, many customers already have their challenges by retrieving these simple data points. You should invest some time when inspecting the following diagram and the study from Gartner. (Reference link)
Here is the first thing the viewers recognize:
“Oh, hey! — We can save money!”
And what they happily ignore:
“Hey btw, there are some heavy assumptions made, which will be very hard to accomplish.”
When having a closer look at this picture, you should be able to identify the significantly increased costs due to the migration itself, the initial high Cloud costs, and the long persistence of data centers (on-premise costs) till their final deprecation. And, these on-premise costs usually come hand in hand with long-term support and licensing costs creating duplicated values for on-premises and in the Cloud.
In addition, you plan for a full replacement of the on-premises environment. It will, though, take two up to four times the time which you have spent for the migration to amortize your additional migration and overlapping hybrid costs.
“Okay, nevermind — can we start calculating?“
Well—, the diagram from Gartner does not even include all costs. Have a read through this reference link — In the meanwhile, the most impactful topics/costs which were left out:
- The continuous training requirements, which the application and infrastructure teams need for being able to manage their environments in the Cloud and even being able to secure and optimize it.
- It also entirely ignores any additional licensing costs or any other technical complexity, e.g., due to service dependencies that start communicating between on-prem and in the Cloud.
- Remember as well that a partial replacement of the on-premises environment will not provide the same results, as general costs like data center infrastructure, networking, any appliances, etc. sitting on-prem cannot be shut off.
- And you should also evaluate the cost of risk around compliance issues.
But the worst issue when just merely moving to the Cloud without a dedicated strategy: You still don´t use the real value of Cloud, as VMs in the Cloud still keep being VMs. Or — in a more fitting analogy:
If you have trash on-premises and you move trash to the Cloud, it still keeps being trash. And there is still no one who wants to take care of it.
This pseudo modernization is frequently being communicated as the first step in the right direction, where you can follow up with the refactoring process afterward. But as your teams don’t need to learn about the new world, this is very unlikely to happen.
Change when changing is required and stop delaying.
Making the transition to my introduction about Digital Transformation, the task of moving IaaS blindly to Cloud is the perfect example of a small risk-averse step.
You are still not running through the tough challenge of cultural and technical change, but furthermore, you try to keep as close to your current implementation as possible. You are only keep focusing on potential short-term gains, which we already led to absurdum with the Gartner example.
Does this mean moving to the Cloud doesn´t make any sense in terms of cost-savings?
I didn´t say that! But —
Serverless computing definitely has massive potential for cost savings, as you only pay for a serverless app when it’s running.
This quote is from a useful article that concludes this topic in further aspects and provides general tips, but especially pin-points to PaaS and Serverless. (Reference link)
Besides, I have found an article on making the cost calculation solely for Serverless, which demonstrates of what cost differences we are speaking about. (Reference link)
A more simplified way to see it can be achieved by having a look at the shared responsibility model, which demonstrates the missed potential without PaaS and SaaS/FaaS — not even touching the additional benefits:
And yes — sure. There are a lot of ways how to improve IaaS, and sometimes you will not be able to get rid of some IaaS.
But then you should check for similar approaches as these from Azure:
B-Servies VMs, idle resources, right-sizing, reducing unused resources/disks, and SQL elastic pools. (Reference link)
ROI — Return on Invest
The financial gain from an investment in cloud divided by the cost of that investment.
When moving to the Cloud, you have two ways of viewing it.
The first one is reducing Cost, and the second one is gaining Business Value*
* A theoretical value which is hard to catch and in the first place frequently ignored or besmiled from Enterprise companies. (Reference link)
Typical Business Values with Cloud:
- Business / IT agility
You enable application teams to deploy faster, automated, and, therefore, react as a company faster to market changes.
- New business models
You evaluate new business models, e.g., by using AI or Data Science on your now centralized data warehouse on the Cloud
- Less operational issues and effort
This is true if you are using PaaS or SaaS. (friendly pinpointing) Otherwise, you will very likely face the same unpleasant challenges as before.
These Business Values are, in fact, the more exciting scope for most companies, but not everyone includes them in their strategy. You will not be able to use those if you don’t drive a Digital Transformation within your company and address also the Cultural Change. (Reference link)
Everything Cloud, aka. “World of Pain”
Typical strategies define their companies are going to move with xxx% of their IT infrastructure to the Cloud by 202x. Having a look at the study from Gartner before, you would be well-advised to move as much IaaS into the Cloud as fast as possible to possibly shut down your DCs. But here you are now running into even more trouble.
At first, there is silence.
Big companies usually first hesitated to even evaluate Cloud. After some time, they got convinced and recognized that competitors have also been moving to the Cloud. Initially, the first challenges have always been Security and Compliance. Due to the lack of skills and the missing trust towards the Cloud providers, many struggled. A fascinating phenomenon of that was the "Security bunker." A Cloud Architecture which was tremendously and often unnecessarily restricted, and many companies are still facing these issues today.
Hesitation over - now migrate!
Depending on how your organization is structured, the migration work is being delegated to the application owners. After time and a lot of emails pressuring them, they might want to have a look at the migration. But be careful — the application teams usually don´t have the Cloud skills yet, and they will, as they are forced to move, take the easiest path.
By directly or indirectly pushing IaaS, you don´t motivate your teams for change, because everyone will stick to their roots.
Therefore, these teams will usually try to replicate the Infrastructure and Application environment into the Cloud, similar to as it has been done on-prem. And when trying doing so, they will face some challenges, because
On-Premises ≠ Cloud
Moving to the Cloud is not only a technical challenge (and it is different!) but also a cultural one. You will directly learn what cultural resistance means when the teams don´t instantly understand why they should now migrate, and they also recognize that it is much harder than they expected. Due to this issue, most of the landscape will be moved into IaaS instead of applying a dedicated and qualified evaluation of the workloads and potentially transitioning it directly into PaaS, SaaS, FaaS.
Another issue which you will very likely face is when you are having a core team or IT as a service provider, that tries to build up a perfect solution for the Cloud. And eventually, but very probably, they will start to (over)engineer a solution that fits Multi-Hybrid-(Ultra-Secure)-Cloud and work on this implementation for years. When they have finished their exercise, and the CTO now tries to motivate all application holders to transition to Cloud, it feels like you have started 50 downloads in parallel.
All migrations run
For small and midsized companies, there are also additional challenges. These companies do only rarely have the trained and dedicated resources to build up a robust architecture and fulfill the migration. Lacking security controls and a resulting Cloud environment, which is heavily differentiating to the best practice implementations, can become the final state. Without a clear focus, you will very likely exhaust your current engineering teams.
Not having a clear strategy is also a strategy. It is the worst.
Well - If you are trying to move a dedicated percentage of your infrastructure to the Cloud in a dedicated period without a further strategy, this will force people to oversee the main challenges in the Cloud Transformation. You will very likely struggle with Cultural Change topics and end up in a legacy environment sitting in the Cloud facing severe security risks because it has not been built in a cloud-native approach. You will also very likely waste a lot of money because your migrated landscape does not respect the best-practice Cloud-design, as your teams are using the Cloud as an extension to your datacenter and still didn´t learn anything new. Or — as my fantastic colleague, Omar Alva, said:
“Buying a Formula 1 car will not make you a Formula 1 race driver.”
And as you did not have the dedicated time to evaluate your application landscape beforehand correctly, you did not pick the best first movers that could have created huge savings or dedicated business value. (Reference link)
And — what about you?
Here the first part ends — opening the discussion, feedback, and experience round.
Check also the next articles addressing various topics that are addressing various topics around Cloud and preconditions to transform your company!
All the best,
David das Neves