Salesforce Application Performance Monitoring

Can you imagine trying to keep track of all your prospect- and customer-related activities on a spreadsheet? What about ye olde days of rolodexes (do people still remember what those are?!)? Thank goodness for Salesforce, the Customer Relationship Management (CRM) solution that revolutionized sales, marketing, and customer care - and how we interact with customers in general.

Salesforce is a critical component for many businesses. Any downtime could affect customer acquisition, lead generation, and customer service and retention. That’s why it’s critical to monitor the cloud application and ensure that any issues are immediately detected and handled.

Businesses that implement a comprehensive monitoring program receive a variety of benefits. In addition to mitigating downtime, IT can determine the origin of performance issues, whether they stem from local resources or problems at Salesforce. With this information, IT staff can upgrade internal infrastructure or mitigate downtime by using failover services or notifying employees.


‍

Now stating, “Monitor Salesforce!” is one thing - but there are lots of ways that can be handled. Let’s take a look at the different monitoring strategies businesses can take and what methods work best for keeping Salesforce up and running.

Active and Passive Monitoring

There are a lot of different ways to perform monitoring, but the two main categories are passive and active techniques. The one you choose depends on your business use case, but enterprise organizations generally use both for critical applications such as Salesforce. Why? Combining active and passive monitoring techniques gives you valuable insights on network performance and availability of the SaaS application.

Active Monitoring

Active monitoring uses scripts, plugins, and third-party software to generate behavior similar to that of a real-world user. This enables you to understand the performance errors a user could experience. In fact, scripting user behavior as tests, you can identify issues before users experience them.

Active monitoring is usually done in intervals and “behind the scenes,” so tests don’t interfere with everyday productivity. For active Salesforce monitoring, the monitoring software might make user-like calls to Salesforce to determine if it’s responding to requests. That way you can identify errors in your code or issues with the Salesforce application.

Salesforce also has an API used by enterprise organizations to automate many functions and synchronize data between local storage and the cloud application. Using active monitoring, you can create your own GET and POST requests to constantly ensure that the API functions as you expect it to and returns results that indicate Salesforce isn’t experiencing any downtime or internal failures.

Passive Monitoring

Active monitoring is great for many things, but keep in mind that you need more data to perform complete analytics and understand the ways your users work with an application. Passive monitoring runs in the background collecting user data that can be used to later run reports, analytics, and determine performance.

Passive monitoring is generally used to determine performance issues so that administrators can deploy additional bandwidth, appliances, or user devices to handle high-traffic loads. For example, you could use passive monitoring to gather network traffic data on the sales team network segment to identify performance problems or resource exhaustion on the local network.

Which Monitoring Method is Necessary?

Many enterprise organizations implement both passive and active monitoring strategies, but active monitoring is critical for Salesforce monitoring. With active monitoring, a plugin installed on the user’s browser can collect data and perform actions on behalf of the user to determine if performance or access issues are present.

Active monitoring also emulates real-world situations so you can identify issues the same way a user would encounter them. It also alerts you to issues before users experience them, which gives you time to mitigate and remediate bugs before they affect user activity. In other words, active monitoring gives you predictive analysis of the application so administrators and developers can be proactive in remediating issues.

That said, it’s not an “either/or” situation - passive monitoring also has its place. It’s good for collecting data and doesn’t require as much network and local device resources. If you want to identify the resources used to connect to Salesforce or determine if malware is maliciously accessing network resources, passive monitoring will give you this information.

“Many parts of Autodesk’s business – from customer support to operations – rely upon SaaS applications like Salesforce.com for important functions. Catchpoint’s SaaS monitoring and alerting help the IT team keep our lines of business up and running." Maira Zarate, Application Monitoring Engineer, Autodesk

Strategies to Monitor Salesforce

There are a number of steps you can take that will strengthen your Salesforce monitoring strategy. You can monitor Salesforce in a series of tests, depending on the issues that you want to identify.

The most effective and thorough way to monitor the application is to use scripts that perform an HTTP action. For example, the script could query the Salesforce application for a record. If the record data is returned, you know that the application is active. For simpler HTTP requests, your script could query the website server for a response. If a 200 OK response is returned with an HTML document, you can assume that the application server is functioning.

If you run custom code that works with the Salesforce API, another strategy is to send GET and POST requests to API endpoints. Endpoints return data that indicates if the request was successful. This data can tell you if endpoints are working properly. If data sent to API endpoints uses the wrong syntax or is invalid, Salesforce returns an error. You can use this information to identify bugs in your own applications.

HTTP and API endpoint requests are useful for uptime and system errors, but you can choose other methods for detecting performance or simple uptime status. Here are a few additional ways you can monitor uptime on Salesforce:

DNS Checks

If DNS fails, any HTTP/S requests will fail. DNS failures could be from the host or it could be that your local network is unable to reach the ISP’s DNS servers. If DNS servers or the Salesforce record can’t be retrieved, browsers and other applications using the friendly web host name rather than an IP address will fail.

Ping Tests

The ping tool is used by network administrators to identify whether a remote server or computer is up and responding to network requests. It can also be used to determine if a website server is up and responding to requests. You can use a ping tool to identify if the Salesforce web server hosting the main application or the API is responding to requests.

Ping tests are beneficial for basic uptime monitoring, especially if you have offices in several locations. They help you determine if Salesforce access is limited in a specific geolocation or if the issue is more generalized. For instance, ISP failure in one geographic area could cause serious downtime for your office in one location, but this may not be an issue for offices in other countries.

Employee Device Checks

The network or Salesforce servers aren’t the only possible issues. Employee devices could be the reason for slow application performance. Devices could have malware hijacking CPU and memory resources, or they could need additional hardware to run browser-based software.

Malware hijacking resources could be a more serious problem than Salesforce downtime and should be identified and contained. Devices could be used in cryptojacking or a vector to distribute ransomware on the network. Monitoring devices help administrators identify whether downtime from a specific device is from the local machine and not network resources.

Conclusion

Salesforce is often a critical component in sales, customer care, and marketing. These are business areas that you don’t want to lose control of.

Monitoring Salesforce is a beneficial way to identify issues early on, mitigate them, and reduce the number of call center tickets when downtime happens. It can also help you stay proactive with performance issues and help IT identify opportunities for improvement across the network.

In fact, monitoring your technology environment in general for anomalies that would otherwise go unnoticed is critical - who wants to be caught unaware when something negative happens? Catching issues as quickly as possible is a good business practice that benefits businesses and customers alike.

Download the Performance Monitoring for Salesforce white paper here.