Historically, it has been very difficult to optimize voice applications—to analyze the granular success of each component and use that data to improve the overall customer experience. It’s been impossible to do it without logging sensitive data.
Our vision of a hosted IVR infrastructure has always been: scalable, flexible and extremely fault tolerant. With our new private logging feature, we’ve now added easily, securely optimized.
Plum’s VoiceTrends analytics tool automatically captures and reports on numerous data points within our IVR systems. It makes light work of performance analysis, showing developers the portions of their applications that are in need of attention.
When we released VoiceTrends, it had a simple logging control feature (enable/disable) built directly into the VoiceXML platform. With this feature, developers were able to disable platform logging to keep sensitive sections of the call flow hidden completely from Plum’s logging.
However, while this approach solved our customers’ security needs, it restricted what our analytics platform was able to capture.
Now, we’ve completely redesigned how our VoiceXML platform considers data logging to automatically filter out sensitive information. The new feature enables users to analyze their application while protecting sensitive customer data and remaining PCI-compliant.
As VoiceXML developers, ourselves, we are acutely aware of how difficult it is to use a new feature. That’s why we built the Voice Trends tool from the ground up to be completely automated and click-to-enable.
Rather than requiring developers to flag each and every sensitive element in their application, our platform enables them to control sensitivity with just one line of code at any level: application, page, form or field.
For security, this is especially important. If securing your application data becomes too complex, developers may end up cutting corners or making mistakes that leave data exposed.
When in “private” mode, the platform takes into consideration the source of the data it is about to log and automatically masks any data originating from a sensitive source.
This provides the best of both worlds—masking sensitive data in our logs while still allowing the analytics platform to track all of the details required for performance analysis.
As an added bonus, this feature also makes debugging applications easier by allowing the developer to trace the complete call flow while masking only the sensitive data.