I am currently exploring the possibilities of
developing a digital sales assistant. The digital assistant would help the b2b customer in his buying process. The main benefits of a digital assistant versus a real person are that a digital assistant is available 7/24, it is patient and it knows all the little and boring details.
The idea of developing a digital sales assistant has come from the need to enhance the buyer’s user experience on diverse web sites and of course from the need to increase sales. The main target group of the digital assistant are small companies with typically only 1-3 employees and the entrepreneur. The main sales item aimed at is an item that is necessary, but somehow routinely or technical and not really related to the main business of the company. Examples of such purchases could be an Internet connection, an upgrade to a security system, one more chair to an existing furniture collection and accounting services. In such buying tasks a digital sales assistant might be the right tool for helping the customer in finding the information that he needs in order to make the buying decision.
In addition to providing information, the digital sales assistant of course asks the right questions in order to find out the needs of the customer. In the best case, it reveals even the hidden needs of the customer. After having made a diagnosis on the customer needs, the digital sales assistant recommends one or two products. In the ideal case, the recommendations are directly links to a store and lead to a purchase.
While surveying the needs of the customer, the digital sales assistant is able to collect data on the needs of customers. In some cases there are no products matching the customer’s needs or the existing product satisfies only partially the customer’s needs. In such cases the data collected by the digital sales assistant may be used in product development and in deepening customer insight.
The envisaged techniques to be used in developing the digital sales assistant are manifold. According to a preliminary plan, the software development will start in the fall term at Haaga-Helia University of Applied Sciences. Since the software development model used in Haaga-Helia is iterative, each iteration will provide a functioning prototype that may be evaluated by the stakeholders. In addition, the requirements of the digital sales assistant may be refined at each iteration.
According to the preliminary plan, in the very first iteration, the digital sales assistant will be based on a file containing frequently asked questions and answers. The questions are typical questions of a salesman and the answers are typical answers given by the customer. We may need to incorporate also a traditional FAQ in order to handle the situations where the customer is seeking factual information and asks questions from the digital sales assistant. In the second iteration, a rule-based model on the interaction between a salesman and the b2b customer might be useful. In the third phase, the project may exploit interaction data to enhance the rule-based model. In order to construct this data-driven interaction model, pattern matching techniques may be used to create prototypes of both salesman and customer inputs. A first version of the digital sales assistant software is planned to be finished by the end of the fall term. I am looking forward to this exciting project.