by Bryan Firestone
While in general it’s not a good idea to have one’s head in the clouds… what about your data? Increasingly it appears so. Software as a Service (SaaS), aka cloud-based software, and Platform as a Service (PaaS), aka cloud-based hardware has become mainstream. SaaS allows applications and data to be delivered via the web, allowing vendors to host and maintain servers, databases, and code, and letting end users easily access those apps through a standard browser. I like how an article in Software Advice sums it up:
From massive corporations to tiny mom and pop shops, cloud-based software is the default deployment method in 2020 for nearly every type of business technology in the world. And for good reason. Business applications delivered via a web browser lack painful installations or strict contracts and are paid for over time instead of all at once.
As I tell people, instead of installing hardware on your premises to run a program, you’re basically ‘renting’ a server offsite to run it.
Yet despite how Software Advice presents it, it’s not for all organizations.
Determining what path is best for your lab is a journey that will be unique to your situation. For instance, we recently worked with an organization that needed a lot of IT help: infrastructure, operability, methodology, data exchange with hospitals – everything. There wasn’t going to be any one thing that fits the bill. As we often see, the solutions are multiple ones and yet all must align with their five-year goal. SaaS can often be the answer to these situations.
Forms of SaaS started in the 1990s and took flight as major data centers adopted it. From then until today, security was always a concern. As security issues were addressed, redundancy was increased, the price decreased, and you saw more companies moving to it in the early 2000s. The biggest movement we can all relate to was when Microsoft stopped selling their software on CDs and moved Office to a subscription-based product in 2001. Salesforce, the world’s largest CRM platform, was another early adapter of cloud computing, and Customer Relationship Management (CRM) programs went this route as well, including our own fulCRM.
Our team is certainly familiar with SaaS – President/CEO Cristy Reiter had experience working with it from the early 2000s, in her words, “when SaaS was not even a buzzword yet, when very few people actually understood the concept.” We’re proud to say that she and the companies she was working with had the foresight to lead the path into the cloud for our industry.
SaaS vs. In-House Data Centers
At least on the surface, in-house systems are less expensive. If a lab client is buying a server, it comes with the operating system and a firewall and almost everything you need to keep it on the premises. For this hardware you might pay around $40,000, plus the software licenses. A SaaS (PaaS) solution could be at that same price annually and needs to be paid on a recurring basis. But when you dig a little deeper, you’ll realize that you need to pay for maintenance and updates every year for the in-house hardware system. Anything that goes wrong is a problem that needs to be serviced by your people. Then in three to five years it becomes out-of-date and needs to be replaced.
Cloud-based systems have the advantage of alleviating the expensive, and sometimes difficult, need to employ IT people to make sure it stays up and running 24/7. But even including all those personnel expenses, a SaaS/PaaS is still probably a little more expensive… but you have a whole lot less to worry about.
Once you get past the expense, SaaS/PaaS advantages start to pile up. For one, it’s scalable. Most labs experience an ebb and flow to business. So, let’s say you have a sudden increase in business that surpasses your hardware’s capability. You might find yourself restricted with your current system and need to buy additional hardware. And what if that surge is temporary? In the cloud, there are provisions for additional bandwidth that allow you to easily increase or decrease your usage depending on the volume of work coming into your lab.
Yet there are limits to SaaS. For one, you’re generally not going to get to customize the software per se beyond the basic software package (our fulCRM is an exception). But that also means that everything is generally stable, because a cloud-based software company won’t stay in business long with an unstable product, so you won’t be confronted with the inevitable bugs that typically come with individual releases (especially those that are then customized). SaaS makes sure everything is tested, secure, and stable before shipping, so it’s as issue-proof as software can be.
Those Security Issues
The biggest drawback to going to a SaaS model from a healthcare perspective is that people do not feel it is as secure as a data center on the premises. If you’re in a cloud with AWS, Microsoft, or Google, the perception is there is a big target on your company. But we’re seeing many companies getting attacked – especially with the financial industry. So, it becomes about how you manage your system and whether you can recover quickly (and don’t forget attacks can and do happen on in-house data centers).
But there is actually another concern with SaaS: latency. It’s possible you would experience longer latencies than an in-house solution, if there are delays in communication between server and instrument because the server is in the cloud. If your organization can’t tolerate that, you may decide to move your server inside your business or consider a hybrid solution.
The Bottom Line
We’ve suggested that some companies use a hybrid model, entering in a situation where components are in the cloud along with elements installed in-house, all configured to allow for updates and typical maintenance. This is appealing because you can stay current without having to purchase licenses. And as always, we’re happy to analyze any company’s processes and business requirements, then recommend the best solutions for their unique situation.
I am myself a convert. I used to be ‘old school’ with regard to the lab industry, feeling that building a data center in-house made the most sense. But increasingly I believe that the advantages of the cloud make it more attractive. It’s just easier to manage – and that’s not a small thing in our business.