Contact Center

Automatic Call Distribution: How it works in Noda

Inbound call distribution (call routing) is а vital function for any contact center platform. The way call routing is implemented determines the level of customer satisfaction, agent workload, and a number of other important call center business metrics.

In Noda Contact Center, there is a special component, Queue Process Manager (QPM), which is responsible for distributing inbound calls (both voice calls and text messages from digital channels) among the agents.

Calls are assigned to available agents with the most applicable skill sets (skills-based routing). By skills we mean the knowledge and qualification levels that allow an agent to handle a call in compliance with the project or company rules. These skills can be: foreign language proficiency, technical knowledge, etc.

In Noda Contact Center, each agent can possess several skills, so that they can handle several queues or projects. The level of each agent’s skills is registered in the system. At the same time, every project has preset skill requirements, including their acceptable range and relevance for the project.

The level of skills applicable to a particular project, and their relevance are used to determine the agent’s qualification level in the project (project rating). This rating is further used to select the most and the least qualified agents for handling calls from a particular queue.

Routing calls to available agents

When a new call is placed in the project queue, the system selects an available agent (if any) with skills enabling them to handle calls in the project. After that, the system selects the agent most suited to the call distribution policy adopted in the project.

You can apply the following call distribution policies:

  • A random agent from the group of agents serving the queue;
  • The most qualified agent;
  • The least qualified agent;
  • The least busy agent with the smallest number of calls, or with the least total call handling time;
  • The agent with the smallest share of labor hours, i.e. the lowest ratio of the total call handling time to the total labor hours;
  • The agent with the longest idle time (the time that passes since handling the last call);
  • The least busy out of the most qualified: one of the agents with the highest qualification level and the smallest total call handling time;
  • The most available among the most qualified: one of the agents with the highest qualification level with the longest idle time since the previous call.

An example of call distribution based on “the most available” policy

As can be seen from the graph, a call from the “Corporate Clients” project is handled by one of the agents possessing applicable skills. From this group of agents, the one with the longest idle time is chosen, as the policy for the project is “the most available”.

An agent becomes available; there is a queue of calls waiting

When there are several projects with queues of more than one call, an available agent is assigned to the queue matching their skills. After that, the system chooses a call from this queue that corresponds to the selected distribution policy to the fullest.

In this case the following distribution policies can be applied:

  • Top-priority calls, longest-in-the-queue calls: the agent first handles the top-priority calls, followed by the calls that have been longest in the queue;
  • Top priority calls, longest-on-hold calls: the agent first handles the top-priority calls, followed by calls with the longest on-hold time (in case of a phone call – after the caller has listened to all the prerecorded messages, IVR menus, etc.);
  • Longest in the queue: the agent gets the call that has been in the queue for the longest time.
  • Longest in the queue * priority: the agent gets the call that has been in the queue for the longest time and with the highest level of priority.
  • Longest waiting: the agent gets the call that has been on hold longest, waiting for an answer from a live agent (in case of a phone call – after the caller has listened to all the prerecorded messages, IVR menus, etc.).
  • Longest waiting * priority: the agent gets the call that has been on hold longest, waiting for an answer from a live agent (in case of a phone call – after the caller has listened to all the prerecorded messages, IVR menus, etc.), and with the highest level of priority.

An example of call distribution based on ” longest in the queue * priority” policy

As can be seen from the graph, the available agent has skills applicable to serving both “Individual customers” and “Corporate clients” groups, therefore a call is chosen out of these two categories of calls. Since these queues operate within the “longest in the queue * priority” policy, the wait time for each call in the queue is multiplied by its priority, and the call with the biggest value is routed to the available agent.

Leave a Comment