Nearly 80% of enterprise applications are now a Software as a Service (SaaS)*. Even the most critical applications are now migrating to SaaS in cloud computing environments. Application migration to cloud hosted SaaS applications generates massive savings, easier global access for work from home and distributed employees, faster app deployment and simpler management .
A form of cloud migration, shifting to SaaS cloud applications requires a careful cloud migration strategy to secure digital experience, employee and business productivity. Migrating users from applications running on premises to cloud SaaS applications can have significant business implications. Cloud, SaaS: both share potentially disruptive cloud migration challenges when it comes to monitoring, troubleshooting, and securing the employee digital experience.
Popular SaaS cloud migrations include business critical applications like ERPs (SAP4 / HANA), cloud CRMs (Force.com / Salesforce or Hubspot), SaaS office suites like Office 365, cloud-hosted unified communication platforms such as Avaya cloud office, Webex, and Teams, collaborative information platforms (like SharePoint and Microsoft’s Power Platform) and document management systems like NetDocuments.
What Are the Risks to Digital Experience in SaaS and Cloud Application Migration?
Performance issues during migration to SaaS cloud applications
The main performance risks that arise when shifting to SaaS applications in the cloud include:
Degraded SaaS application response times caused by user-to-SaaS / cloud connectivity
Users had a clear network path between application sites and data centres. Your IT team controlled all infrastructure between. Adding public networks (internet or cloud access network) to a private network (WAN/MPLS, VPN, LAN) can increase network latency from users to the application (see ‘SaaS Performance Drivers’ for more). This can increase end-user response times, negatively affecting their experience.
IT and Service Desk Teams’ Lack Tools to Handle Performance Issues
Moving to the cloud is a paradigm shift: you must manage services delivered by others (connectivity service providers, CASB/ZTNA, cloud and SaaS providers) as well as your own (endpoints, local networks, WiFi, etc.) Network and application monitoring tools designed to monitor private and packaged applications lack visibility into SaaS in cloud computing environments, which cannot be monitored using traffic analysis.
Potential Business Consequences
IT can struggle to maintain end-to-end delivery chain control when moving to cloud-based SaaS. Without a comprehensive monitoring and management tool, IT teams may struggle to identify service outages. Lack of SaaS visibility can cause costly downtime and missed SLAs. IT departments may struggle with SaaS availability and performance.
The consequences of inadequate tools to quickly resolve SaaS and cloud-based performance issues directly impact the business in multiple ways:
IT Helpdesk Efficiency
Your IT helpdesk may be unprepared to handle end-user tickets about SaaS response times and remote work. Inability to qualify and troubleshoot these cases will lower closure rates, increase resolution times, and frustrate users. This can lead to a lack of IT team credibility, lower productivity, and longer resolution times.
IT and Network Operations Productivity
Without proper monitoring, it is impossible to quickly identify the source of any bottlenecks, meaning that businesses are unable to maximize the productivity of their IT resources. Additionally, without the ability to identify these cases, the IT team is unable to take proactive measures to prevent future issues.
Inefficient Vendor Management
IT teams must measure each layer’s contribution and hold internal teams and external vendors accountable for issues. IT can’t optimize performance or have productive conversations with providers without an end-to-end view and a clear breakdown of response times. Knowing who’s responsible speeds up issue resolution.
Impact on Business Productivity
Poor SaaS performance affects business processes and user productivity. Slow app performance frustrates and reduces user productivity. This can reduce their productivity and a company’s efficiency.
How to Mitigate Risks During SaaS and Cloud Migration
Your company is considering migrating to the cloud and replacing locally hosted apps with SaaS. How can you minimize the impact of performance issues on your employees and customers?
1. Start With a Clear Baseline
Securing digital experience requires comparisons between time periods, versions, user groups, etc. to determine response times. First, you must have visibility on the current start situation and application and infrastructure experience.
What are the most important pieces of information you’ll need to understand how application use affects your infrastructure and how users’ digital experiences will change after the migration? Here is a short list:
- Usage profile
- Distribution of users by location (remote, country, site, …)
- Most common transactions
- Response times and error rates
- By transaction
- By location
- User path to the application
- Corresponding network performance (latency, packet loss)
- Infrastructure services impacting latency (firewall, proxy, etc…)
This will serve as a reference for current application digital experience and will assist you in realizing:
- Current gaps and weaknesses in the original application
- Whether app response has been improved or degraded
2. “Sanity Checks” During SaaS Migration Preparation
You can perform some tests even before the application is made available to the first group of pilot users. Performance tests are used to ensure that response times and throughputs remain within acceptable limits.
Users will most likely access the SaaS application differently than they do legacy applications, for example:
- To access legacy applications, most users will connect to your datacenter via SD-WAN/MPLS or a VPN.
- They will most likely use different paths in a SaaS cloud-centric architecture, such as:
- Direct internet access
- A home Internet access through a CASB
- A local internet breakout from a SD-WAN connected site
- A VPN/MPLS/SD-WAN connection, followed by the internet or secure cloud gateway
Having this in mind, you need to:
- Define precisely what changes are involved in application access architecture
- Record a connectivity “Birth certificate”
- Latency on private networks
- Latency induced by security devices and gateways
- Latency to the app components
- Measure end-to-end latency from the user to app components
- Compare these to the latency users are exposed when accessing the legacy application
- Determine the potential for optimizing this connectivity
SaaS vendor infrastructure audit
Every SaaS application platform combines data center-, CDN-, cloud-, and third-party services. SaaS platforms often have 20+ hostnames in different hosting environments, regions, and cloud providers.
Each of these hostnames is resolved into an IP address, which should connect your users to the nearest available node for each service. Having said that, you should acquire detailed knowledge of the SaaS application platform:
- App structure and reachability
- Which hostnames are used?
- How much do they impact the application’s ability to run properly? (Some hostnames will be used extensively, while others will be occasionally used)
- “Cloud Capillarity”
- Which services are offered from highly dispersed cloud locations versus a small number of sites around the world?
- How close are each critical hostnames to current user locations in terms of network latency and packet loss rate?
- Platform connection to 3P (on premise, SaaS, PaaS hosted)
- Many SaaS platforms will use APIs to access data stored in other platforms.
- Monitoring network latency between on-premise data centers/clouds and the SaaS platform is also important. You must know the reliability and quality of connectivity.
By validating these points, you’ll likely find business-impacting issues you can fix before you scale up SaaS app users. When launching a SaaS platform, network reliability is crucial. Regularly monitoring network latency, routing and path performance can ensure a smooth, reliable digital experience.
3. Performance Monitoring During Beta Testing
Beta testing gives you real-world performance data on the SaaS application and digital experience. It’s a chance to optimize the end-to-end application delivery chain before production.
- Real life performance check
- Locations, devices, browsers, transactions with poor response times
- Infrastructure layer corresponding to performance bottlenecks
- Measure performance rates for all beta testers
- Closed loop optimization
- At this stage you can easily try alternative ways to access the SaaS application and measure the actual impact on end user response times
- Utilize the knowledge gained to improve future performance for regular users.
- Proactively engage vendors to deliver best-possible performance before scaling up
- Collect objective data on performance bottlenecks
- Get telecom providers, CASB and SaaS vendors to maximize performance
Before introducing real users, use this stage to fine-tune the app. Before scaling up, proactively engage vendors to deliver the best performance.
4. Scale Up SaaS Application Use and Performance
Your infrastructure and your vendors’ infrastructure may respond differently as application usage increases. Each of these components handles volumes differently, and any one of them could be a bottleneck or weak link. If you want a high-performing SaaS app, optimize each component.
During ramp up, monitoring end-user response times and infrastructure behavior will shorten MTTR and allow fast optimization. In addition to monitoring application performance, it’s important to analyze each layer’s usage during peak and low user traffic:
- Real life check
- Manage vendors along the way
Run Stage: Achieve and Maintain Peak Performance
Once in production, the performance of SaaS applications will keep evolving:
- Through software releases
- Upgrades and changes in the SaaS platforms
- Changes and production incidents on telecom networks, internet exchanges and peering points, cloud access gateways
- Internal network upgrades and incidents
Application architecture must be constantly monitored and optimized to ensure peak performance and the best digital experience for users.
Kadiska’s digital experience monitoring platform monitors SaaS application performance and network connectivity. With Kadiska, businesses can be proactively alerted to potential infrastructure issues and maintain peak performance by closely monitoring networks, SaaS applications, cloud hosting environments, and cloud access gateways.
Secure the Digital Experience of Your SaaS Migration
Learn more about SaaS migration, performance, and digital experience with these videos, case studies, and articles: