Think it’s scary this Halloween? Imagine being in charge of your company’s contact center and being held responsible for providing a terrible customer experience. Now there’s true terror for you!
Last week I highlighted few examples of issues discovered through predeployment testing. Today I’d like to highlight the goblins that await you if you don’t implement a rigorous day-two monitoring strategy.
Below are a few examples of scenarios we at Empirix have seen our customers experience over the past 30 days. These can be pretty frightful, but all can be pinpointed and corrected through an active monitoring strategy.
Six Things That Can Go “Bump” In the Network
Like end-to-end testing, monitoring can be grouped into key categories. This will enable you to better focus on issues to get them resolved quickly, with minimal impact to the customer experience.
Here are some issues with IVRs we’ve helped correct recently:
1. Large insurance company: This organization performed an IVR test that started to alert. A prompt that used to play in 8 seconds was now taking more than 10.5 seconds. Following investigation, we discovered that there was a problem with the back end system that was affecting IVR performance. The prompts were still playing, but it was taking the system longer to retrieve and play the appropriate message. The insurance company was able to run reports and see that over time, there had been a gradual decline in the IVR response time, eventually pushing it to the point that it was frequently exceeding the threshold originally chosen. The question to think about: What are your thresholds; are they consistent with business goals, and how are other systems impacting your metrics?
2. Major Bank: At this bank, calls would proceed normally until the point where the IVR would ask for a PIN. After collecting the PIN, the system would play a “One Moment please” message and then disconnect the call. The question to think about: How would you be alerted to this? Is a customer complaint call (if it gets through) your first line of alerting?
3. Passenger Rail: This company was alerted to a major telco outage in their area by their Empirix “IVR Live” tests. The questions to think about: How would you know if something like this happened? How long would it take before you’d realize it was going on? Remember, customers can’t call you to tell you your phone isn’t working!
4. State Government: This organization’s monitoring platform alerted them to an IVR processing error: the system wasn’t properly collecting the nine-digit SSN entered by the callers. Instead, the system was collecting eight of the nine digits and then playing an error message that said it couldn’t find that SSN. The questions to think about: How would you handle this? What might your csat scores show on a post call survey?
5. Financial Services Company: A voice quality (VQ) test failed on the main greeting. Why? Because the IVR wasn’t saying anything! The question to think about: Maybe you checked your IVR once when you went live, but how do you know it’s working today?
6. Chemical Company: This company discovered that it had an intermittent issue: the first call a customer made was ring no answer and the second call would get disconnected at the main greeting. Because it was alerted to this issue, the company was able to begin investigating a possible trunking problem. The questions to think about: How do changes to one part of your communications system affect another part? Specifically, how do you monitor trunk usage and provisioning based on the continuous changes to your customer experience management system?
These are just a few of the examples we’ve seen recently of challenges our customers are facing every day. I hope they provided you with insight that can help you think in a clearer way about your own environment and how it affects the customer experience.
Managing a complex contact center can be pretty scary, but it doesn’t have to be!