Create a human-like conversation between a Bot and a customer. Evaluate sentiment and take real time decisions using artificial intelligence.
OneContact QM (Quality Monitoring) is a tool to monitor the contact centre quality level, ideal to evaluate and improve agent performance while meeting company goals. Supervisors can access voice and screen recordings, emails, chat sessions and social media interactions in real-time.
Integrated environment that allows both the creation of flows within a traditional IVR (Interactive Voice Response) system and flows of intents and entities with chat or vocal Bot assistance. These AI multilingual Bots are self-learning software systems that can be trained by the company’s staff.
Customers have come to expect automated responses in certain stages of their support interactions. However, it’s important to offer the same engaging experience and clear information throughout the different channels.
That’s the goal of ONEFLOW, which unifies the conversation scripts to be used both by the Interactive Voice Response system and the conversational bot. This solution is key to further develop the chatbot “behavior.”
The chatbot will be trained to correctly identify “INTENTIONS” within 100 to 200 possible scenarios, as well as the “ENTITIES” involved, and what kind of immediate help can be provided.
The more you train the chatbot, the better it will be able to detect the “EMOTIONS” displayed by the customer and rank them. Based on the valuation, the bot will determine if the conversation needs to be transferred to a human agent.
1Stream has natively integrated market leading speech recognition capabilities into OneContact Quality Monitoring. This technology examines calls and assesses both the agent and customer’s tone; it transcribes voice into text to extract relevant information such as intents expressed through keyword trends or areas that need improvement. The results are indexed, searchable and can be used to improve customer experience and identify selling opportunities. This tool is split by 3 different modules:
For a bigger picture of the contact center. Analyze global insights like sentiment distribution, talkover and silence periods, overall sentiment of the interaction, word cloud with the most used words during the interaction, improper language and others.
Watch interactions in real time with a transcription of everything being said. In case there is an urgent issue or significant drop in sentiment, an alert is triggered. A supervisor can then act on the interaction by redirecting it to a human agent. This is particularly useful to avoid that a problem escalates to something bigger like a social media crisis.
You can access all the information that was extracted from the recording and see the evolution of sentiment levels (both for agents and customers)
and act on urgent issues in real-time, based on customer sentiment & keywords (verbal or written)
automatically the most relevant topics and outcomes, within millions of stored conversations
with machine learning long term Contact Center productivity – a continuous improvement process