The contact center industry manages customer support enquiries around the world – and is facing huge challenges caused by the COVID-19 pandemic.
It has previously been standard practice for workers within the industry to perform their duties in bespoke office spaces where they have immediate access to:
- appropriate hardware and software tools
- confidential company and customer data
- knowledge base and subject matter expertise to resolve especially complex cases
The “on-hand” expertise of peers, supervisors and managers is important in creating the elevated customer experience that contact centers strive for. The accrued wisdom of the center is often the difference between a successful outcome to a customer query and a “case” which drags on.
As with many other industries, COVID-19 triggered a rapid and massive shift to Work From Home (WFH). The WFH conditions have frequently been far from ideal for the contact center workforce to work as effectively and productively as they do from their fully equipped office facility. With schools closed and general caring responsibilities, some workers have not been able to work full time - and others not at all. COVID-19 has therefore led to a shortage of talent to resolve customer service issues.
We have heard of situations where the available contact center capacity is being bolstered by branch staff because branches are closed. With an already stretched workforce multi-tasking, incoming calls are not landing with specialists as often as they should, and CSAT scores are under threat (even while customers are a little more forgiving during a crisis**).
Furthermore, finding time to assign in-house experts to upskill existing workers (or train any new hires) to perform the missing functions is proving difficult, when managers are scrambling to find workers to deploy.
The upshot of all this is the essential need for technology to resolve many of the issues. Technology that can rapidly scale-up to handle as many of the easy (but high-volume) issues as possible is in high-demand. Machine learning (ML) had already established itself in this industry and continues to grow its offering, coverage and reputation; but the last four months have made the case unarguable that ML needs to play an ever more significant role in covering this “COVID-gap”.
At Ness we are helping customers overcomes barriers in the customer journey using ML. Such solutions can be used to manage the load, through:
- circumventing the shortage of agents using call deflection to intelligent virtual agents that leverage customer journey mapping and analytics.
- suggesting real-time next best actions for the agents.
- automating the most frequent service flows to resolution.
- elevating the quality and immediacy of real-time information available to agents to deliver faster issue resolution.
- improving the productivity and efficiency of existing contact center operations.
ML based solutions may exist pre-embedded in some platforms that are being used within the industry, but several legacy contact center platforms (with significant market share) were not designed with AI/ML capabilities in mind.
Whether the notional capability already exists within the platform - or not - we see common reasons why ML is not being put to work more frequently across the industry. The most obvious hindrance is a human one: long-serving customer service managers who do not have access to the knowledge, talent or technology experience to promote the case for it - or the AI/ML engineering capabilities to implement such solutions once the case is made.
This is where Ness is helping customers.
We are helping these managers by creating ML solutions for commonly encountered contact center problems. We are able to rapidly deliver value by resolving queries faster and lightening the load on the over-stretched human call center workers.
Recent examples include
- Intelligent virtual assistant for faster customer self-care troubleshooting. Our solution modernizes the usual customer question flow which is effectively a menu of questions mostly based on a basic decision tree and forces you to start all over again if you abandon the call for any reason... If you are using the traditional virtual assistant to fix your slow broadband speed, you will have to go through the same process from the very beginning every single time. We have added a Machine Learning layer to increase the intelligence of the virtual assistant by removing the repetition and frustration from the process to get to the solution faster – often without the need for a hand-off to a live agent.
- ML supported Next Best Action for sales and contact center agents based on real time events. Rather than the agent relying on a traditional fixed list of pre-approved options, Ness has a solution that offers options personalized to the customer and mapped to a range of dynamic offers based on the user journey. The intention is to de-stress the customer and the agent, elevate the Customer Experience, point towards a satisfactory resolution early in the interactions and, potentially, steer the engagement towards an upsell.
- Automation of workforce coverage forecasting based on real-time events (traditionally a time-consuming statistical analysis of historic patterns). We have created Deep Learning models that update the predicted number of staff that will be required every 30 minutes. This gives managers extremely granular detail for planning purposes and delivers huge cost efficiencies based on real-time data.
- Fast ticket resolution with Natural Language search. Our Amazon Kendra-based solution uses an intuitive NLP intent based search (as opposed to keyword search) to help the call center agent find the ideal answer faster.
- Duplicate ticket identification using Deep Learning partner solutions to rationalize the workload and interpret real customer intent to reach resolution faster.
The areas above showcase the vast range of common business problems that Ness’s AI engineers are working towards solving. Reach out to TMT@ness.com to discuss and demonstrate how one of them can start offering value to your business tomorrow.