Using AI/ML for Non-English Language Smart Indicators

Image
Case Study: AI ML Solutions / Advanced Analytics

Using AI/ML for Non-English Language Smart Indicators

SUMMARY

One of the largest telecommunications provider partnered with Spearhead to use AI/ML solutions for behavior-content dependent predictive modeling to recommend the best language (among the bilingual set) to communicate with a customer regardless of the consumer’s initial indicated preference.

Challenges

  • Client offers its customers the option of a preferred language of communication during the ordering time. The customer indicated preference is then used to route calls to appropriate bilingual agents when the customer calls in the future.
  • The client’s objective is to reduce the call average handling time and improve customer loyalty and satisfaction.
  • However, despite the cost incurred by the Client in providing such a bilingual language option, none of the intended business KPIs were improving.

Solutions Offered

  • Spearhead’s analysis of the Client’s data showed that reliance on the customer indicated preferred language is not a driver to improve the Client’s intended KPI because the effectiveness of language preference changes based on the nature of the issue and the content interaction itself. In most cases, the indicated preference can result in non-efficient call routings and un-needed use of bilingual agents, driving costs up.
  • Instead, Spearhead’s AI/ML solution uses behavior-content dependent predictive modeling to recommend the best language (among the bilingual set) to communicate with a customer regardless of the consumer’s initial indicated preference.
  • The solution delivered better mapping between the customer to agent bilingual language to effectively optimize the interaction. The solution considers the complexity of the content, the nature of the exchanges, and the need of the consumers to drive the recommendation in addition to demographics and segmentation data.