Should I go to the Cloud?
How to answer this question confidently
This is a question I get asked a lot and despite data showing there’s an upward trend in Cloud adoption, I still come across groups of people questioning the effort, security, cost, value etc and staying away, while others engage reluctantly or from a safe distance with light workloads. I can categorize these adoption concerns in 5 buckets, and I’ll provide an assessment guide on whether to jump in, learn more, or wait it out.
1. What we have works, we’re happy
What I’ve quickly realized with this group, is that they are really good at what they do. They have systems they trust and rely upon. They also tend to be Cloud-curious and they are likely to have vetted a Cloud solution and evaluated initial cost. However what I’ve found out, very consistently, is that they are wanting to improve on what they have, which is somewhat expected, given their diligence in putting in place what they have in place today.
If you find yourself in this bucket, identify your top 3–5 areas of improvement and evaluate Cloud offerings related to this. To illustrate, I came across an SMB that was happily running a fleet of internal web apps 24–7, but their usage was very limited at night and on weekends, they also kept lots of files on disk and had to maintain a robust backup system. A limited-access-time policy, scheduled availability and migration of artifacts from disks to Cloud storage, drove costs down significantly in monitoring efforts and resources by leveraging S3, smaller volumes, serverless (really? no servers? lol this deserves another post), automation that starts and stops instances on schedule. All this, done within days!
2. Our information will be out there, we’ll loose control!
This also comes in this version: We are so eager to adopt, security wont approve. This by far is the most common I run into: You have a group of very enthusiastic individuals but face a hard wall with security or compliance. Well, everyone is trying to do their job and I truly understand the deadlock. Change management is an important process of any reputable organization, it requires buy-in and diligence when implementing. There are definitely very valid cases for this, for instance, when regulations require data to be stored within specified regions or countries, and your preferred Cloud Service provider doesn’t have a presence there. Well, keep doing what you do but keep yourself informed, major service providers are expanding to new markets and becoming compliant with major regulations.
When it comes to general IT Security, audit of what security measures in place today is warranted, and more often than not, there’ll be gaps identified. But most importantly, having a well defined list of security and compliance concerns, makes scoring against Cloud services, very easy. For instance, if you are keeping unencrypted data on-premises, you’d be concerned migrating it to the cloud as is, where hackers are very active, and it’s unlikely they’ll pursue “smaller” targets e.g. an on-prem data center. But are you really secure just because you are not a target? Cloud providers have security as their number one concern above anything else, and they provide resources for you to do your part. It’s always a Shared Responsibility model and understanding this should be the first step. What this audit will also reveal, is that there’s also non-sensitive parts of your organization that could run in the Cloud with minimal effort and without compromising regulations: a great place to start. For instance, you can leverage an image-to-text extract service, where you store all artifacts locally and purge them after processing them in Cloud, paying only for what you use.
3. We work locally/regionally, global connectivity isn’t what we need.
While this is a big selling point for Cloud, it’s not the only one, and my advise is to quickly shift focus back to your customers, systems, processes and find areas of improvement. You can do so many things in a single region without ever having to venture out to other regions. And in the future, who knows, you may need a presence in another region. But here’s a good blog on what to consider before deciding to go multi-region. Disaster Recovery is a good reason to go multi-region, and you can do this on the cheap.
Majority of what I’ve seen, keep their solutions simply in one region, for a very good reason: proximity. Proximity to your customers, proximity to other integrations (and oh by the way most newer online business apps run on the Cloud as well), but most importantly, proximity of your own data, processes and services in the cloud. Think about running analytics: You’d source your data in the cloud, compute in the cloud, store results in the cloud, where your cloud-hosted apps access it. All this is super fast, leaving you to do more .. of your thing I guess?
4. We simply don’t understand the cloud, not much sense investing in it.
Never invest in a business you cannot understand
— Warren Buffet
I very much agree you shouldn’t invest in “strange” business. But trust me when I say, there’s gold in learning what the Cloud can for you. It’s not only a trend, but a very convincing and valuable trend! You may not have time for it, totally understandable, but someone on your team might. I’ve written a blog on how to get started with Cloud skills, check it out here. If not, plan to spend some time with a Cloud expert, they’ll help you understand your options.
5. It seems expensive, both the adoption process and operations.
Unfortunately, this is very true. For a successful cloud adoption, commitment is a key ingredient in multiple areas including financial, human capital, and time. There are quick and dirty ways to do it and while they might appear cheap in the beginning, might end up being expensive. I wouldn’t recommend the alternative either, where you are ultra careful and plan every little detail before acting. This will cost you as well.
There’s a happy medium: iterate and do it fast. Now I don’t mean trying out things you are not familiar with, have an understanding of how they work and what they are meant to be used for. Best investment is knowledge. Do this first and fast. Then follow up with a swift execution, get a feel of the process, and start a review process immediately, and keep asking yourself these questions:
- Does it work as intended?
- How much is it costing me? Cloud costs balloon quickly!
- Is it optimized/any recent updates to the service/better alternatives?
- Are our skills current/are we engaging with an up-to-date partner?
This will keep your costs down just by keeping an eye on things and questioning everything.
“You cannot swim for new horizons until you have courage to lose sight of the shore.”
— William Faulkner
Make it a great day!
