Task
In 2020, many businesses started implementing new digital improvements into their communication and analytical processes.
Acoustic Group, specializing in the manufacture and trade of materials for sound insulation, vibration insulation, and acoustic comfort, approached us. The company needed a chatbot for their web portal https://www.acoustic.ru, which was supposed to take the routine off the employees so that they could spend more time with customers.
The chatbot was supposed to solve the following tasks:
- Online consultant and online tech. support (selection of soundproofing solutions based on the client’s problem).
- Online application and order acceptance.
As a chat center for correspondence with customers, the client used LiveTex.
Results of chatbot implementation
In the case of Acoustic Group, the main necessity for a chatbot was that it was necessary to relieve operators at the stage of collecting information about the sound insulation problem in the potential client’s room, based on which design engineers could provide a solution and sound insulation assessment. However, the client did not understand how to compile a selling script.
After our team conducted an analysis of dialogues over a six-month period and interviewed company employees, target customer segments were identified, a list of key questions was compiled, and ultimately, a marketing funnel reflecting the specifics of the company was formed.
The chatbot was integrated with LiveTex and launched in November 2020. After the chatbot was launched, its operation was analyzed in the first month, and changes were immediately made to improve clients’ communication with the virtual operator.
As a result, already in the second month of operation, the chatbot was able to independently successfully handle 60% of all incoming dialogs. And two months later, after gradual adjustments and adding missing scripts, 85% could be classified as successfully handled dialogs, the results of which became full-fledged applications to the marketing and engineering departments.
In 2021, as part of project support, 12 monthly and 2 quarterly reports on the bot’s work were compiled to improve communication with people and add new more relevant information.
Also, API integration with Bitrix24 was refined for transferring the application directly to the department’s employees in CRM. The set task of reducing the time for clarifying details with clients by operators was fully achieved, and now the bot continues to improve in accordance with the company’s actual information.
You can contact the Acoustic Group chatbot right now by visiting the company’s web portal and clicking on the chat window in the lower right corner of the screen: https://www.acoustic.ru
Tools
The chatbot used a standard support chatbot configuration: a self-service menu with natural recognition, as well as a ticketing system.
In addition, API integration was added: sending applications to the engineering and installation departments, as well as triggers for resetting abandoned dialogs a week after their creation.
Dialogflow from Google, an NLP technology for natural language recognition, has been used with the application of context to limit the recognition of information for a specific product.
More than 100 intents have been created to switch between scenarios and provide answers to frequently asked questions, based on real user requests.
All consultation request scenarios at the endpoint can be transferred to a call center operator or to another scenario. Before transferring to an operator, the chatbot automatically displays all the collected information and requests the user’s contact details, if they are not available: a phone for communication.
The call center works in conjunction with Metabot + LiveTex, using the Metabot platform’s chat widget window for dialogues with customers, customized specifically for the company’s website design, and LiveTex as the operator’s workplace where the chatbot transfers dialogues for people. The chatbot also has a Telegram account, dialogues from which are also transferred to the LiveTex omnichannel panel.
For more detailed information about the functionality, you can study the article by Alyona Rvacheva, a designer-architect, co-author of the chatbot.
