Description:
automation of testing the effectiveness of advertising campaigns and assessing the activities of advertising agencies and assessing the share of conversion in the customer acquisition funnel at the stage of processing a telephone call from a client to a contact center operator requires, from a recording of a telephone conversation, presented in the form of text, an assessment of the probability of the presence of one of the designated events in conversation between a call center operator and a client. These events are targeted customer actions and call intents.
Context:
a company is launching an advertising campaign, the goal of which is to create a flow of calls to the contact center on a specific topic (for example, to sign up for a test drive of a car). The effectiveness of such campaigns is difficult to measure, since the content of each dialogue must be understood. Manually marking up dialogue was an inefficient process.
Solution:
it was proposed to build a dialogue tagging model based on the created markup.
Results:
The solution is built and adapted for:
Integration into the BI cloud platform Automation of CRM maintenance for customer clients:
To build the model we used:
Simulation results:
Customer: Contact Center, Telecom
Technology stack: text-preprocessing models, classification models, language representation models, Flask, Python, PyTorch.
automation of testing the effectiveness of advertising campaigns and assessing the activities of advertising agencies and assessing the share of conversion in the customer acquisition funnel at the stage of processing a telephone call from a client to a contact center operator requires, from a recording of a telephone conversation, presented in the form of text, an assessment of the probability of the presence of one of the designated events in conversation between a call center operator and a client. These events are targeted customer actions and call intents.
Context:
a company is launching an advertising campaign, the goal of which is to create a flow of calls to the contact center on a specific topic (for example, to sign up for a test drive of a car). The effectiveness of such campaigns is difficult to measure, since the content of each dialogue must be understood. Manually marking up dialogue was an inefficient process.
Solution:
it was proposed to build a dialogue tagging model based on the created markup.
Results:
The solution is built and adapted for:
- Car dealer companies;
- Real estate selling companies;
- Medical center companies.
Integration into the BI cloud platform Automation of CRM maintenance for customer clients:
- CRM completion is 40% more and 20% more accurate;
- Automation of conversion level measurement in acquisition channels;
- Reduced labor costs by 60%;
To build the model we used:
- Dialogues between call center operators and clients, broken down into replicas;
- Information about the presence of events in dialogues;
- A priori expert knowledge about the essence of events.
Simulation results:
- Model for predicting the probability of an event;
- Conversation Dataset Analytics;
- Advertising campaign conversion analytics.
Customer: Contact Center, Telecom
Technology stack: text-preprocessing models, classification models, language representation models, Flask, Python, PyTorch.
