The Challenge

Without further information and definitions of short and unproductive calls, it is difficult to suggest a way to reduce them. Benchmark figures predict a range of about 15 – 25% of calls falling within that category. All program owners have a significant chunk of short calls, and usually, those calls are ignored and labeled as either “transferred” or “disconnected” calls. Short calls are usually considered to be of no usage from an analytics standpoint because they are considered to be of no value. But what business owners need to understand is that there is still a significant cost attached to these calls. They need to look for patterns or anomalies – high levels of transferred calls, individual agents with unusually high numbers of short calls, particular inbound numbers with high numbers of short calls – and drill down from there.

Take a look at the monthly trend for a technical LOB of a telecom company – Around 1/4th of overall volume falling into the “short calls” bucket.

 

Mk Approach

This was a significant amount of volume being overlooked for a long time and the MK team decided to do a further drill down using Speech Analytics capabilities. One of the widely used applications of Speech Analytics is content categorization or call driver analysis. After identifying few key phrases, queries/ extractors can be written to automatically segregate calls in specific categories which then can be analyzed

 

Findings

Below you see the first level drill down after a few basic extractors build. Just by looking at these basic details, few outliers would be fairly apparent and point towards the need for further drill-down (increased number of transfers, for example). Root cause analysis can be done through a quick targeted listening activity and then the queries/extractors also serve as a control mechanism giving almost a real-time trending of the drivers.

 

Results

The team did a further drill down for transferred calls. A high percentage of calls should have ideally been directed towards service queues like desktop assistance, printers, cloud services, etc. However, calls landing up in the wrong queue led to high transfers, in turn adding to costs.

 

The identified problem directs towards the potential solution of a more robust IVR process that directs the calls intended to the correct department instead of needing to spend unnecessary time on transfers and increase handle time. Similar other drill-down findings will help business owners to assess impacts to cost and find solutions within agent or process level.

MK’s SmartSpeech is a blazingly fast, highly accurate, and cost-efficient Speech Analytics solution. This solution assures key business intelligence from a much larger sampling of conversations and actionable insights needed to improve customer experience, sales performance and ensure compliance.

 

About MK Analytics:

MK Analytics (Part of MattsenKumar (MK) founded in 2010, a full suite BPO services company) offers in-depth consulting, social media analytics, and speech. They have over a decade of experience in delivering next-generation analytical solutions. The MK Analytics team comprises Six Sigma, Lean, and PMP professionals who deliver best-in-class assessments, findings, and recommendations. MK Analytics’ consulting solution enables clients to engage us across a wide array of initiatives with a definitive ROI for all projects undertaken. With SmartSpeech, MK Analytics provides a customized solution to clients across industries and verticals. Their SmartSpeech Technology is powered by Voci’s proprietary HyperVox™ hardware-accelerated speech recognition technology, with highly accurate output that can then be further analyzed to customized client needs. Services range from voice data solutions for customer experience, call center operations and transcriptions are guaranteed accurate out of the box, to give companies an all-in-one research platform