Software-as-a-service (SaaS) offerings provide centrally-hosted programs over a Web client. SaaS has become one of the most popular deployment methods for cloud-based applications. Because SaaS programs are so common, the need to monitor the performance of your SaaS apps has become greater and greater. End-to-end SaaS monitoring allows users to examine the programs they use from an end-user perspective.
We spoke with Vandan Desai, Senior Product Manager of Digital Experience Monitor (DEM) provider Catchpoint, for his thoughts on SaaS monitoring. He goes into what makes SaaS monitoring unique and why it’s important for users of an SaaS solution.
Even though SaaS providers present service guarantees and SLAs, why should SaaS customers still invest in a SaaS monitoring solution?
VD: SaaS apps are cloud-based, and cloud infrastructures – in spite of their rapid scalability, resources on demand benefits – are not immune to outages and slowdowns. Freak events can always occur, like in 2015 when a Google data center in Belgium was struck by lightning several consecutive times, leading to permanent data loss for several unlucky customers. It’s an extreme example, but it shows how using a cloud-based SaaS service does not guarantee you impeccable performance; you have to take a proactive approach yourself in monitoring and ensuring this performance.
The impact of not being in control of SaaS apps is lost employee productivity and satisfaction (which becomes even more frustrating when you don’t know how to turn back on service) and lost revenues – both of which can drag down a business. Maybe you have a great SLA for availability. But are you measuring speed specifically? Outages are bad, but sometimes slowdowns are worse – people keep trying to use the application and failing, leading to significant frustration. A SaaS provider might have upheld its 99.9% uptime promise, and perhaps you only lost an hour in the month. But what about the 10 hours the service was running at a 50 percent speed when your end-users wanted to put their fist through their screen? We’ve all been there, working on a web interface when every step, every click, every screen load, results in that little wheel turning and turning and the seconds tick by. Do your SLAs reflect that lost productivity?
Even if a SaaS provider’s servers are showing 100% availability, that does not mean your end-users across the globe are having a universally excellent experience. There are many geography-specific elements between the SaaS provider’s data center and your end-users that influence the ultimate experience outcome. If you don’t monitor end-user SaaS performance yourself, you don’t have a true view on that outcome. Micro-outages are also becoming more prevalent. Micro-outages are silent SLA killers that can frequently go undetected by SaaS providers.
There are usually many opportunities on a SaaS user’s own side (behind one’s own firewall) to optimize SaaS performance, including localized configurations – and SaaS monitoring can help identify these. To take back control over SaaS performance, it all starts with visibility; what gets measured, what gets improved. To get visibility, enterprises want to understand their employee experience, end-to-end. They need to understand it from wherever these employees are, on whatever they’re using, throughout the delivery stack to the origin from where the application is being served. To be in control, enterprises must be able to anticipate their end-users’ problems and hold their vendors accountable to their SLAs.
What makes SaaS monitoring different from other types of monitoring – in particular, IaaS and PaaS monitoring?
VD: SaaS is becoming the standard for enterprise applications. According to research by financesonline.com, 38 percent of businesses began moving to SaaS-only workplaces in 2017.
Comprehensive SaaS monitoring is now a critical requirement:
- 73% of business will run almost exclusively on SaaS by 2020.
- 80% of end users prefer to use SaaS for collaboration, organization, and communication.
While monitoring performance for all of these types of services is critical, SaaS monitoring has a particular sense of urgency because organizations typically have more SaaS end-users (everyday workers across multiple departments) than IaaS end-users (IT workers) or PaaS end-users (developers). Think of popular SaaS CRM applications that essentially power sales, marketing, finance customer service operations and more. If an application slows down to the point that end-users give up – or it goes offline totally – the ripple effect across the organization tends to be much larger. Organizations have become so dependent on these SaaS solutions that when the applications falter, so does the entire operation’s productivity and revenue-generating engine.
Choosing a SaaS solution means relinquishing management of several responsibilities. How does a SaaS monitoring tool help ensure good performance when so much is out of your control?
VD: A SaaS monitoring tool gives you a constant, realistic view into how end-users in various geographies are experiencing the application. If performance is strong, you have peace of mind. If there’s a problem, advanced diagnostics can precisely identify the source in order to take action or avoid wasted time finger-pointing. You can take back control by monitoring via a neutral third party and access to actionable, shareable reporting data to help vendors. In the event of a problem detected on the SaaS provider’s side, you can flag it to them and work together to fix it and enforce your SLA. If the problem is on your side (behind your firewall), you can identify it quickly and dramatically reduce your Mean Time To Repair (MTTR) in fixing it.
Or perhaps the problem is an element in between that neither you nor the SaaS provider has control over – a poorly performing regional ISP, for example. Then you have the opportunity to proactively communicate with your end-users as opposed to wasting time needlessly war-rooming. Imagine being able to put a status update on your portal, or an email to your employees with an update letting them know you are aware of the problem and working with the SaaS provider to resolve it. That’s going to save you hundreds of support tickets and improve morale due to more transparency.
Thanks to Vandan for sharing his perspective!
Check us out on Twitter for the latest in NetMon news and developments!