The phrase Active Monitoring can mean many different things, ranging from a periodic ping to verify that a server is reachable, to an HTTP transaction to measure responsiveness of a website, to a multi-party voice transaction to determine the quality levels that users are experiencing. Today, I am going to focus on active voice transactions.
A voice call is still one of the most important methods for communication – yet it’s still way too common that users have a bad experience.
One of the main reasons why quality is so important with voice calls is because a voice call is almost always a real time and interactive. Whether you are interacting with an application in a contact center, such as an Interactive Voice Response system (IVR), or talking with one or more people through a unified communications system, there is a certain dynamic that is taking place: one side says or does something, the other side responds, and so on. If something goes bad, it tends to sour the entire experience. So what could go wrong?
We’ve all gone through an IVR maze trying to finally reach a live person. Then when we do reach a human being, that person re-asks all of the questions we’ve already answered or transfers us to the “right” department because the IVR wasn’t doing its job properly. In either case, wasting agent’s time and frustrating customers certainly aren’t the results the company was hoping for when they deployed the IVR.
Another common problem is poor voice quality. If you can’t hear what the other person is saying, it doesn’t matter whether they are across the building or across the world – it’s extremely frustrating and you’re likely to hang up. And when you hang up, someone is losing money.
The Benefits of Active Monitoring
Active Monitoring using voice transactions can also be thought of as continuous end-to-end testing. This isn’t stress testing, whereby you increase the call load to find out how much the system can handle. Instead, it involves periodic, low volume voice calls that emulate real users trying to communicate.
Since you are not talking about humans who describe problems using words, you can collect all sorts of metrics that tell you definitively what went wrong. For example, you can discover MOS values, jitter, packet loss, unexpected disconnects, long wait times for the IVR, wrong path through the IVR, and so on.
Another benefit of using Active Monitoring is that transactions can be recorded and played back later. This is much more reliable than having a person describe what they experienced, as there is no bias involved and you don’t have to depend on the accuracy of someone’s memory.
End-to-end Active Monitoring can be deployed in different configurations – with software-based endpoints in your data centers, on the PSTN, or in the cloud. You can continuously test your network infrastructure and applications, the same network infrastructure and applications that customers are using and that your business is relying on. And although Active Monitoring won’t necessarily prevent problems from happening, it’s certainly one of the best ways to detect them.
How do you monitor your communications systems? Let me know in the comments below.