Directory bot that prompts contact center operators
Website widget
Main information
01/
Clients
Business IT services developer
Tasks
Make a widget for a energy company's website that unloads a call center by automatically answering calls
Audience
Business and ordinary people who use electricity
The largest power grid company in Tatarstan
A «smart» bot that assist users on the client's site
Development object
Bot development process
02/
preparatory stage
main stage
Step 01
Study materials from the client
Step 02
Step 03
Step 04
Transfer to development
Develop a bot on our zDialog platform
Step 05
Step 06
Integrate the solution with the customer's program
Step 07
Conduct joint testing
Launch, debug
Design of the program architecture
Develop scripts and knowledge base
Tools
03/
To create a bot, we used our own services:
Simple and convenient platform for developing chatbots, voice assistants and contact center automation systems.
zDialog
AI service for chatbots analytics. Allows you to conduct a deeper analysis of human-robot dialogues and get a more accurate assessment of the effectiveness of the bot.
OneDash
Scheme of work
04/
The bot is looking for an answer
in the knowledge base
in the knowledge base
The client asks
a question on the site
a question on the site
Gives help
Transfer
to operator
to operator
bot work
finds the answer
can't find an answer
The client doesn't have to wait and is therefore satisfied
What the widget looks like
05/
The bot is launched in a dialog box on the client's site
Introduces and offers tooltip buttons
The bot recognizes the user's question and sends the answer
Bot database
06/
01
The bot looks for answers to questions in the knowledge base and redirects the user to the operator if it cannot answer itself.
We've added them to make it easier for operators.
02
03
Key skills
Extra skills
When a person communicates with a user in a chat, the bot works in the prompter mode and offers the operator options for answers.
The bot «learns»: operators can add new questions and answers to the knowledge base.
We've added them to make it easier for operators.
We've added them to make it easier for operators.
We showed the demo version of the project to the President of Tatarstan
Interesting fact
Project team
07/
Project Manager
Ivan Starkov
Developer
Anton Kapridov
Technical Director
Sergey Verentsov
Analyst and QA-tester
Anna Michnik
Developer
Alexey Sherchenkov
The project took nine months - we can say that we gave birth to it, like a child
Quote
Ivan Starkov
Project manager
Difficulties of the project
08/
01
There is a lot of information in the knowledge base. To make the bot «smart», we processed a lot of texts.
02
Since the the base must be constantly be updated, we added the option to "train" a bot
Project results
09/
This is the first time we created a program that included both bot and human interaction with the user
We have greatly improved the widget while working on the project. It learned to send messages to the user instead of the operator.
We have successfully integrated with the internal systems of the ICL customer, so we easily made changes to the project during work.
Contact us
PRODUCTS
SECTIONS
STUDIOS
Russia, Innopolis, st. Universitetskaya, 7
2023 © EORA. All right reserved