Regarding customer service, 45% of end users prefer virtual assistants as the primary mode for answering questions.
There is an increasing need to arrange things online. The trick is to help customers in an efficient, effective and fast way through online channels that are available 24 hours a day, 7 days a week. In this way, an actual omni-channel customer service is created, where customers have the choice between digital and traditional channels. The expectation is that in 2025 a virtual assistant will even take over 80% of all customer interaction.
'Over 50% of medium to large enterprises will have deployed virtual assistants by 2020.' Gartner Inc.
Business case examples
- Significant cost reduction of customer care
- Improved customer satisfaction as result of 24/7 instant service and resolution
- Increase of sales by providing information, recommendation and instant check-out
The value of virtual assistants
Achieve a higher customer satisfaction through self service, 24/7 customer support and immediate response. More efficient, and effective customer care, allowing your operators to spend more time for complex questions that require more attention. Value: good user experience
Closed loop learning
User confirmation and feedback on conversational agent replies are monitored by multiple quality metrics, providing you with insights of its performance and which features to implement next. Value: continuous improvement
Natural-Language Processing techniques give the interactive agent access to the user's intent as well as relevant entities (e.g., locations). The interactive agent can then use domain-specific knowledge models to determine which info is needed to answer the user's query in as few steps as possible. Value: custom domain dialogue, no repetition of facts
Connect & handover
Your conversational agent is able to connect to back-office systems for retrieving relevant data. In case that the conversational agent is unable to assist your client, e.g. no solution found or negative sentiment, a handover to a human agent can take place. Value: integration in existing processes
The difference between old and new technology translated into an example
VIQTOR DAVIS implements intelligent and context-aware virtual assistants, capable of having a non-linear conversation mimicking an actual dialogue with the customer. Employees no longer have to deal with mundane Q&A’s and are available for clients with more complicated questions. As for clients, interacting with a virtual assistant means no waiting in line, 24/7 support and a self-service approach to customer care.
Experience shows that clients react very positively to these new developments and are increasingly making use of this type of communication.
A solution up and running within four months
VIQTOR DAVIS provides an open bot framework that combines in-house build software with proven vendor or open source technology. We act in four clear phases;
- Phase 1
We put together a team, define the use- and business case as well as determining the architecture.
- Phase 2
In phase two we work on a basic dialogue, live handover, single channel and platform foundation.
- Phase 3
Maturity starts already in phase three where we work on medium to complex intents, knowledge modeling, multichannel and voice, user authentication and basic transactions.
- Phase 4
In phase four we scale-up with omni-channel, pro-active resolution, advice intents, commercial transactions and computer vision.
A reference project is the best proof of success
Banking business case
The bank’s 30.000 employees depend for their daily work on a multitude of products and (internal) services. An employee has to deal with many channels, processes and a huge volume of scattered information making it tedious to get served on its daily needs.
VIQTOR DAVIS has created a Virtual Assistant that acts as the single point of contact for any question. The foundation is a bot that provides intuitive humanlike interaction that leverages the existing platforms for information retrieval, incident resolution and service fulfilment, provides for handover to a live-agent and uses AI to act intelligent.
Insurance company business case
With the growth of artificial intelligence and its proven value, one of the largest insurance companies in the Netherlands needs to build up knowledge and expertise to keep up with the pace of development and the changing market. Using techniques like virtual agents, the company can serve its customers 24/7.
VIQTOR DAVIS has set up a center of excellence and trained a team of insurance company employees to develop virtual agents, using knowledge graphs, natural language understanding and machine learning. VIQTOR DAVIS also delivered a containerized non-linear virtual agent solution that can be trained for domain specific use cases. Together with the client learning on the job, VIQTOR DAVIS delivered customer-facing virtual assistants.
Pension company business case
One of the largest Pension providers in the Netherlands servicing over 4 million participants have a customer care department that resolves over 3.000 questions per day. The pension fund participants demand an increased service level that is available outside office hours and the customer care department is required to ensure a high customer satisfaction while being cost-efficient.
We initially focused on the intents with high frequency that are relevant and suitable for automated resolution, including FAQ, Administrative updates (ie. address and partner) and information requests. In parallel we worked on the complete solution that is capable to provide full service by the virtual assistant by means of DigID authentication, back-office integration and call-center handover to a live-agent when needed. The first virtual agent was live within 4 months and continuous improved since.
The virtual assistant is now the primary channel on the Pension Providers website. Within 6 months of initial go-live, the virtual assistant already handled over 300 chats a day and covered 20% of all customer interaction. This resulted in a yearly cost-saving of over 100.000 EUR.
Government business case
A large government agency needed a solution to service their agricultural customers outside of their busy office hours. The solution needed to be able to handle sensitive data in the future, hence no information should be sent to (foreign) vendors; the components of the virtual assistant that handle sensitive information should be able to run completely on premise.
VIQTOR DAVIS developed a chatbot that combines custom and open-source components. The virtual assistant is able to detect and correct spelling mistakes specific to the complex agricultural context. The parts of the virtual assistant that handle sensitive information can be run completely on premise should this be required. Customer care employees with no previous experience were trained to monitor user conversations and improve the language model.
This virtual assistant stands as one of the early examples of a successful AI project within the Dutch government.
Contact VIQTOR DAVIS
Are you interested in what VIQTOR DAVIS can do for your organisation?
Contact Sebastiaan Brandsma directly;
+31 6 11 25 63 99
For this solution we cooperate with respected partners.