Serverless Efficacy - Travel & Hospitality Industry
It's no secret that the travel and hospitality industry has been constantly changing and adapting over the past few years.
However, has the industry gone far enough?
In my opinion, they haven't.
In this article we will talk about how a cloud initiative can be supercharged to not only take advantage of "the cloud", but leverage fully managed (serverless) solutions to streamline development workflow, lower TCO (Total-cost of ownership), and help build systems that can automatically scale up to match high customer demand and scale down to match an uncertain customer demand.
The Impact of 2020
In the beginning of 2020, like many of you, I was traveling around the country and to other countries. My travels were focused on helping some of our enterprise clients take advantage of serverless in an onsite fashion, when things took a turn.
The House Of Cards Collapses
Off the back of an onsite training session outside the US for a large client where Serverless Guru helped train 30+ developers on how to utilize serverless development tooling and started building the initial blueprints for that team to migrate and build new serverless services in parallel.
The news broke of something called "Coronavirus" or "Covid-19".
At first, it wasn't fully understood if this Coronavirus would become a full blown pandemic or not.
As I was traveling back from Canada towards the end of January 2020, there were hundreds of people wearing masks who appeared to be traveling in from overseas.
It was unsettling. At the time there had been whispers about Coronavirus and you could start to see some panic setting in for the travel and hospitality industry. No one had a full picture of how things would develop, but for those in the travel and hospitality industry it appeared to be more clear how quickly something could spread globally.
Fast forward to early March and the US Stock Market tanks, lockdowns go into effect, air travel gets suspended, and almost every industry runs straight into a brick wall.
Shut It Down
When the shutdowns took effect and people weren't traveling, staying in hotels, etc. something really interesting happened to some of our clients and their cloud bills.
The enterprise clients that we worked with who had invested in 2019 and early 2020, saw a magical moment take place where their costs dropped off a cliff matching the user demand dropping off a cliff.
What we are looking at above is what we saw reflected in some of the cloud bills that our clients had after the lockdowns took effect.
These clients invested into modernizing their applications with serverless technologies many months before this event happened.
Their customer base was crushed, the traffic to the apps was decimated, and as an interesting result their cloud costs shrunk to match the new crippled traffic.
Meaning MASSIVE cost savings. The best part, they didn't do anything special on their end when this event took place. They just watched the cloud bill tank, freeing up resources for other critical operations to combat the new changes. Including refocusing developers towards features that might help generate revenue. Something that every company wanted to do, but didn't have the capacity too while trying to fight wildfires.
They didn't adapt in months, they adapted instantly.
Saving Millions Of Dollars Automatically
When the lockdowns took place the travel and hospitality industry was one of the worst to be hit. Flights couldn't take off, customers didn't feel safe at hotels, and companies were scrambling to figure out how to cut costs quickly.
However, in reality when you're panicked, you're not doing your best work. You are pushing buttons in a rush and hoping that the changes you make will have an impact. You're not prepared and unfortunately this was the reality for most businesses impacted by the Coronavirus.
Now Imagine Your Company Had Been Prepared?
And when the lockdowns happened your cloud costs adjusted automatically to match the demand drop.
Leading to millions of dollars in real savings without lifting a finger:
- No need to turn off servers manually
- No need to reduce capacity on databases
- No need to invest hundreds to thousands of engineering hours to try and react
And not only the RAW resource cost savings, but because you made the shift to serverless and were relying on fully managed services. You didn't need to pay engineers and other staff for thousands of hours of work to try to adapt in real-time.
At the scale of our enterprise clients this would have resulted in millions of dollars in further investment towards tearing down, reducing, and optimizing workloads.
Because our clients didn't have to focus on these activities, they were able to put their full attention towards how to pivot, how to optimize other areas of their business, and strategize for the new reality we found ourselves in.
Hindsight is 2020
As the saying goes, we can see clearly looking backwards, but have a hard time seeing into the future.
The idea that a global pandemic could rock the entire world and cause this level of disruption would have been nearly impossible to fathom in 2019. Everything on the surface appeared to be booming.
Our enterprise clients who moved to serverless before the crash of 2020 didn't make this shift to safeguard from a pandemic, but they did so knowing that fully managed services (serverless) would allow for the following:
- Faster Delivery Of Services
- Less Overhead & Maintenance
- Less TCO (Total Cost Of Ownership) With Pay-Per-Use Pricing
- More Focus On Their Product
As a bi-product of taking advantage of serverless, they were not only prepared for the when a global pandemic overtook the world. But, had put the pieces in place to focus on delivering more to their customers, faster, and with less overhead.
Your Customers Don't Care About What's Under The Hood
Your customers don't care if under-the-hood you're using servers, containers, or cloud functions. They don't care if it's a fully managed database or some internal hand-rolled database that scales to the moon and you won an award for it.
They don't care and why should they?
It's not important if your team built the best server or container in the world. Your customers care about the product and unless you're a virtual machine hosting company or a cloud provider then your customers simply don't care what makes the product usable.
They do care about outages and how long it takes to build new services.
By keeping your focus on building a world class server setup or container setup you're focusing on the wrong thing. It's great for resumes, it's great for internal high-fives, but as a result the number of hours spent on building new services is diminished.
Sprinkle in an outage incident and your customers may jump ship, but at least you have a world class server or container setup, right?
Hear Me Out
Let's do a thought experiment.
When we think about existing workloads that are running on servers and containers. Let's ask the question.
Can our workloads leverage serverless solutions and achieve less overhead, less cost, faster delivery, and still provide the same level of quality to customers?
In some cases the answer to this question may be a flat NO. There are cases where the number of requests per second that your applications need to service your customer traffic may be difficult to match in a serverless architecture.
However, this is not the most common case.
From our experience, most workloads fit into a serverless context and the ones that don't are constantly being challenged by cloud providers like AWS (Amazon Web Services) who are consistently improving things under-the-hood without any cost to you.
Think about that for a second.
You didn't do anything, you didn't pay someone to do anything, and still your applications get stronger
It's the magic of serverless (fully managed) services and why it has been and will continue to revolutionize the software industry into the future.
The Savings No One Focuses On
When we think about cloud costs, we often are talking about how much a database costs per hour, per month, and per year.
We are NOT talking about how much money it costs to pay humans to do right-sizing, provision planning, etc.
We are NOT talking about how much money it costs to pay humans to handle fires that happen with the database in real-time while customers are facing outages.
We are NOT talking about how much money it costs our businesses when our applications fail and we lose revenue that could have been realized.
We are NOT talking about missed opportunity cost because our focus was in the wrong area.
However, in each one of the above scenarios companies are still spending truck loads of money and ultimately it's a diminishing return.
Why Serverless Makes Sense
Serverless (fully managed) services allow for you to offload the responsibility to the cloud provider. By doing this, you are effectively solving for the costs which often don't get brought up and aren't seen or tracked as often.
And these costs are WAY more expensive than your bill to keep the application running.
Building applications that leverage as many serverless (fully managed) services as possible is the key ingredient to having fault tolerant, highly scalable, and cost optimized applications by default which are hardened over time automatically in the background.
When your company is constantly focused on everything, but your actual product e.g. the underlying infrastructure that needs to be managed and coordinated. It makes sense that the product isn't where it should be or why new competition with less resources is biting at your heels.
In my opinion, this leads to the worst case scenario.
Leaving something on the table. Leaving a rock unturned. Not hitting the true potential of your business, your product, and your team.
You've heard me rant about why your company should take serverless seriously and why your competition absolutely is taking serverless seriously, but still how do you take one step forward?
You need a guide.
The same way you would plan to have a guide if taking a trip across the Amazon Rain Forest or taking on a climb up Mount Everest.
When taking on a serverless initiative with many moving parts in an uncertain environment which is our new reality.
You can't afford to take a misstep, the stakes are too high.
Let us be your guide.
Serverless Guru is a cloud consulting company that specializes in serverless adoption for enterprise. We are on the forefront of the industry and our team is unparalleled in the realm of serverless application development.
Serverless Guru has worked with some of the largest brands in the world helping leadership map the future of their company with serverless and development teams transition to serverless.
If you want to simply get an idea for what this might look like for your company, please fill out the form below and get in touch with us.