How to Eat an Airplane: A Crash Course in Cloud
Feel like everyone’s got this cloud thing figured out but you? Here’s an explanation of what it is, why it matters to your business, and how to make your move.
The cloud in its most basic form — networked computing and remote operations and storage — is as old as the internet itself. What’s (relatively) new is the business model that calls for things like servers and basic IT infrastructure to be purchased from cloud providers. Usually one of the big three: AWS, Azure, and Google.
The basic pitch: Because you pay only for what you use, you can invest in exactly the IT resources you need, and no more. Free from the need for in-house expertise in tech, your organization can focus on core business functions. And the infrastructure (and servers, and whatever else you get via the cloud) will not only be more reliable, it will also give you access to emerging technologies like AI that are being developed at a level and speed by cloud providers far beyond what you could afford on your own. And the time it takes to bring apps and other new products to market will be a fraction of what it would be if you had to deal with every aspect of deploying them yourself.
Is it all hype? The short answer: no. But the pressure to “get on the cloud” can cause unnecessary anxiety when what we mean by that term is not clearly articulated. Demystifying the process will give you the confidence to get started. Spoiler alert: “getting on the cloud” is not really the goal anyway. Achieving your business outcomes is.
But we’re getting ahead of ourselves.
What is the cloud?
No, really. What is it?
Let’s go back to the very basics. (If you find this too basic, please skip to next section.) Think of computing in layers. The top layers are what’s closest to the user. If you use a traditional PC, creating and saving files onto your hard drive in the manner of our ancestors, and do nothing else, congratulations! You are 100% not on the cloud. The second you connect to the internet and get data in the form of email and webpages served through your browser, you are technically (though minimally) “on the cloud.”
On the other end of the spectrum is a Chromebook — a device even more in the cloud than a mobile phone, which generally still requires you to download and install apps. Software physically installed on a Chromebook is minimal. Just enough to access and run web-based apps. The apps themselves, the computing power that runs the apps, the location of your saved files — all that stuff is out there somewhere in the salt mines of Utah, the wilds of Australia, or wherever the servers are.
There are many versions of this chart floating around the web. Some of them show where in the chain the cutoff is for Software as a Service (SaaS), Platform as a Service(PaaS), and other aaSes, but those terms aren’t necessarily great for thinking clearly about the cloud. This one is simplified to give a quick conceptual understanding:
Creating and storing files on a traditional web-connected PC
Using a PC to work with files on OneDrive or Google Drive
Working on files on a Chromebook
If there was a one-size-fits-all, off-the-shelf solution like Chromebooks to meet the needs of businesses, we wouldn’t be having this talk. The needs of a telecom giant, healthcare provider, or whatever industry you’re in, are infinitely more complex than those of an individual consumer. But the promise of the cloud is analogous to the promise of a Chromebook: you can make a minimal investment in hardware, and access — via the cloud — all the necessary IT infrastructure you need. Infrastructure far more advanced than you could build yourself, maintained by a vast team of experts who do nothing else.
But what does that mean for me?
What all that actually means for businesses falls into three broad categories:
1. New, cloud-native capabilities
Think: anything you haven’t done before that you’re building in the cloud. This can include things for your internal operations and things for your customer.
Take some work Slalom did with Qlik, for example. Qlik is a data visualization and analysis company. We developed Qlik Sense App Sharing — a new visual dashboard — using Amazon Web Services. As Qlik CTO Anthony Deighton put it, “We wanted people to be able to try Qlik Sense without having to go to the trouble of installing it.” This cloud-based system lets people collaborate on Qlik technology using any browser.
This project showed the speed that’s possible using AWS. In the old days, spinning up a dev environment was an arduous process that could take up to a month and included fun things like negotiating data center contracts. With AWS and other cloud services, it can be done in a matter of hours. The cloud means getting new products out faster than your competitors, and the power to respond to the market in something close to real time.
There’s also scalability. In that same Qlik project, Slalom helped set up a system that automatically scales regional servers based on demand. That demand can come from a desktop, iPad, or mobile device. The cloud-based nature of the Qlik offering makes it agnostic with regard to devices and OS.
Cloud also offers the opportunity to start fresh instead of migrating old systems just because you’re used to them. “This is a genie that you can’t put back in the bottle, so we wanted to make sure that we did it right,” says Deighton. Putting it on the cloud will make Qlik’s offering more widely available to customers who have lots of data to parse through but don’t necessarily want to host Qlik technology on their own servers.
2. Migrating existing assets to cloud
This category includes all the things you want to do now but can do much better in the cloud. Perhaps you want to improve mobility (e.g., making customer records web-based and device- and platform-agnostic for field service personnel), or reduce the need for in-house IT expertise. Both would fall under this category.
Analytics firm Teradata, seeing its customers shift their data from on-premises to cloud, wanted to make its services more widely available on AWS and Azure. The goal was to give customers greater access to self-service analytics, on a flexible, pay-as-you-go basis. Part of the solution was called IntelliCloud Management Console, which we built in close partnership with Teradata, at first on AWS, and later extended to Azure. Says Abhishek Lal, the company’s director of product management and engineering, “Being serverless reduces the operations cost. There are no servers to manage and compliance is much simpler. Everything is more event-based, and so much easier to operate.”
The Teradata example also highlights the importance of shifting to the cloud because that’s where your customers are already headed. But there can be pitfalls. In the rush to “get it on the cloud,” companies can give in to the temptation to dump existing problems into the cloud rather than seizing the more difficult but ultimately more valuable opportunity to fix them. A process is not inherently better because it’s done in the cloud rather than on prem. It’s only better if it’s better — i.e., it offers automation, better scalability and security, etc. If you approach the cloud as simply a more convenient hosting environment, you’re likely to miss out on most of the benefits.
3. Integrate, secure, and automate the environment
These three buckets of cloud stuff are not in chronological order. You can migrate and build net new capabilities at the same time. But this last category can be thought of as the final stage (insofar as anything is ever truly “final”), in which legacy processes are not only migrated but made truly cloud-native, security protocols are codified, and as much of the tech environment as possible is automated. New projects and products are developed on 100% cloud platforms, and can be released with a speed far beyond what you can achieve currently. In this stage, the full promise of the cloud is unlocked. You have connected the individual cloud projects from our first two buckets into a greater whole. Benefits are compounded in a greater-than-the-sum-of-their parts “1+1=3” effect.
You may have started with something small, but in this phase, you are free like never before, with the full power of the cloud behind you, to focus on your core business, and expand into new ventures that we may not have even imagined yet.
How to eat an airplane: Meet Mr. Eats It All
The question is how to get started. The list of issues and dependencies is formidable: Security. Data policies. Rolling out major organizational change while existing work continues unimpeded. Accounting for all the legacy operations currently done on-premises or on traditional servers. Getting strategic alignment across the enterprise. And, most importantly, doing it all in a way that will still make sense when your cloud transformation is complete. It’s a lot!
So how to approach this impenetrable thicket of issues? Consider Michel Lotito. Between 1978 and 1980, this Frenchman ate a Cessna 150 airplane. He drank a lot of mineral oil and water to smooth the way through his digestive system. But, more importantly, he broke the plane down into one-centimeter cubes. Lotito — better known as “Monsieur Mangetout,” or “Mr. Eats It All” — then ingested a few of these chunks every day until the plane was completely gone.
The lesson: impossible-seeming tasks are best approached by breaking them down into manageable increments. Don’t try to come up with a perfect plan. Trust that you know where you want to go: a world in which your customers are connected to you in new ways, you have seamless access to the latest innovations, and you can focus on your core business instead of maintaining your own IT infrastructure. Then pick something and start.
But how to pick something? Which part of the Cessna is first?
You first need to generate a list of possible cloud projects. Let’s not skim past this part as if it’s easy. Chances are, you are already doing cloud initiatives of some kind, though you may not call them that. You’ll need to account for those, as well as business needs throughout your organization, and figure out a place for them in your strategy. Trying to map these needs to even a simplified list of AWS or Azure services can be daunting. CloudWatch, Lambda, or Device Farm, anyone? How about AppStream, Application Gateway, or Table/Blob Service?
Once you have your list of possible cloud projects, rank them in terms of complexity and value to your business. That will enable you to come up with a chart like this:
Now you have your starting point: the upper right quadrant of lowest complexity, highest value projects. The idea is to get a win, even a small one, and build on it. Starting with the most valuable/doable thing not only adds value faster, it helps you build credibility within your organization, and create momentum for the longer-term project: eating that airplane.
The human element of cloud transformation
The cloud has the power to change the way you do business. To put it even more plainly, it will change the way you do business. Profoundly. As the very epitome of a disruptive technology, cloud computing will inevitably generate its own barriers. While an exhaustive exploration of change management as it relates to the cloud is a subject for another paper, here’s a quick round-up of some of the keys to success in our experience:
• Bring key stakeholders along for the ride: A core principle of Slalom’s agile approach to change management is keeping stakeholders engaged throughout the process, and not springing change on them in a single massive release. Involving the people who will be most affected by the change in every stage of the process will make it more likely that those stakeholders will be invested in your success.
Building with serverless technologies requires a cultural shift.
Align your executives: Slalom’s agile approach of rapid iteration cycles and getting quick wins as early as possible is — not coincidentally — also ideal for driving executive alignment. If you’re the point person for cloud transformation in your company, you may have support in principle at a high level, but there’s nothing like a quick success to solidify that support.
• Set realistic expectations: Overselling what cloud projects can do will hurt you long-term. An accurate cost/benefit analysis is thus a foundation for a successful cloud transformation.
• Manage fear and anxiety: Psychological barriers around cloud computing can arise when data loss and privacy risks are perceived as major challenges, and be must be addressed as part of your advocacy for cloud adoption.
• Do some people planning: Limited resources and expertise is an issue whenever you’re pushing into a new technological area. And though you will ideally be making less of an investment in in-house IT talent, your journey to the cloud will eventually need a new approach to recruitment and retention. Understanding the talent impact associated with new ways of working and securing appropriate expertise is key to sustain the cloud adoption and realize the benefits.
• Prepare for IT team changes: Adjustments to internal IT culture are almost certain to be part of your journey. Duties may shift, and people may be forced out of their comfort zone. In our experience, this kind of change is best achieved by example rather than edicts from on high. We embed with pilot teams and do the work with you. These teams, empowered by their first-hand experience, are then released into the larger IT ecosystem and charged with helping facilitate and advocate for the new ways of doing things.
Hope not fear
There’s anxiety around the cloud. A cloud of anxiety, as it were. The thought is, “if we don’t adapt, we’re going to die.” And that’s… well, it’s actually kind of true. The cloud is where things are headed. It’s certainly where your competition is headed.
And maybe you’ve waited a little bit longer than you should’ve to get started. The great news is that it’s not too late. Not by a long shot. The process is both daunting and highly doable. If you’re getting a late start, the great news is that the tools and resources available to you now are more advanced and accessible than they’ve ever been.
There will be bumps along the way, but that’s to be expected. And that’s another lesson we can take from Mr Eats It All: Rolling with adversity and moving forward. Once he was stabbed and underwent surgery for life-threatening injuries. Three weeks later, he ate a robot.
Special thanks to Kevin McClelland, managing director at Slalom, for his contributions to this article.
The Slowdown is brought to you by Slalom, a modern consulting firm focused on strategy, technology, and business transformation.
