Skip to content

Contact Center AI Routing

Contact Center AI offers a robust and flexible approach to routing conversations within contact centers. Designed to handle a wide range of use cases, Contact Center AI empowers organizations to seamlessly connect customers with the most suitable agents, ensuring efficient service delivery.

Routing Scenarios

Assigning Conversations to Queues

Contact Center AI simplifies the assignment of conversations to queues using the no-code Experience Builder for straightforward use cases. For advanced requirements, queues and skills can be dynamically configured programmatically throughout the conversation lifecycle, leading to an agent transfer.

Routing Methods

Contact Center AI provides a variety of routing methods to suit different operational needs:

  1. Named Routing: Routes conversation directly to specific agents for the flow.
  2. Standard Routing(V2): Matches conversations with agents based on all required skills and language match.
    1. Routing with Proficiency: Incorporates agent skill and language proficiency into the decision-making process.
  3. Advanced Routing(V2)
    1. Routing to Preferred Agent: Overrides skill-based rules to prioritize agent-specific requests.
    2. Time-based Skill Relaxation: When no agents possess all required skills, conversations are routed using a partial skill match based on skill expiry.
  4. Load Balanced Routing(V3): Used in simpler environments, this method focuses on skill eligibility within a queue while ignoring proficiency matching. Conversations are assigned based on the agent's oldest last assignment time, ensuring fair distribution. When multiple agents match the required skills, the conversation is routed to the agent with the longest last acceptance timeline, preventing recently assigned agents from being prioritized. This ensures a balanced workload among agents. Learn more.
  5. Reprioritization: Allows reprioritizing conversations during queue transfers to adapt to changing customer or business demands.
  6. Routing Fallback: Invokes the 'No Agents Available' flow if no agents are logged into the system or invokes the ‘Out of Operational Hours Flow’ if the conversations are received outside configured working hours.

Pre-Transfer Workflow

Before transferring a conversation:

  • Conversations are queued, and required skills can be attached to the conversation.
  • Queues: Act as holding areas for conversations and are independent of skills.
  • Skills: Represent agent attributes such as expertise in specific topics or languages.

Transfer Workflow

Once the transfer is initiated, Contact Center AI employs two primary routing strategies:

  1. Standard Routing (V2): Matches conversations with agents based on skill and language eligibility.
  2. Advanced Routing (V2): Adds layers of complexity, such as:
    • Matching with a preferred agent.
    • Expiring specific skills over time to widen the eligibility pool.
  3. Load Balanced Routing (V3): Used in simpler environments, this method focuses on skill eligibility within a queue while ignoring proficiency matching. Conversations are assigned based on the agent's oldest last assignment time, ensuring fair distribution. When multiple agents match the required skills, the conversation is routed to the agent with the longest last acceptance timeline, preventing recently assigned agents from being prioritized. This ensures a balanced workload among agents.

Multi-Queue Assignment

Agents assigned to multiple queues are managed using the following logic:

  • Conversations are assigned a priority level (1 to 10).
  • Higher-priority conversations are matched with agents first.
  • For conversations with the same priority, preference is given to:
    • Conversations where the agent is a preferred agent.
    • If no preferred agent exists, skill match determines the assignment.

Language-Based Routing

Language plays a crucial role in matching conversations with agents. Both conversations and agents are assigned language attributes, ensuring that only agents proficient in a given language handle those conversations.

Winning Agent Selection

Contact Center AI determines the best-suited agent for a conversation based on:

  • Routing method (preferred agent, skill-based, or skill expiry over time).
  • Agent availability and skill and language matching.
  • Additional factors, if enabled, such as:
    • Skill proficiency.
    • Current load (for digital interactions).

Agents are ranked based on these parameters. The best agent is selected from this ranking. Alternatively, a Load balanced routing method can be employed for simpler environments, which focuses solely on skill eligibility within a queue.

Key Benefits

  • Efficient Handling: Reduces wait times by intelligently matching conversations with suitable agents.
  • Flexibility: Supports simple and advanced routing needs, catering to diverse business scenarios.
  • Customer Satisfaction: Enhances customer experience by prioritizing preferred agents and language compatibility.