When you create a queue, you can set the routing method and, for those that support it, skills evaluation methods. Genesys Cloud offers the following routing methods: standard, predictive routing, bullseye, and preferred agent routing.

• Standard routing sends interactions to agents based on the skills evaluation method you select.
• Bullseye routing also sends interactions to agents according to the selected skills evaluation method, but includes expansion parameters that adjust the routing target pool over time. Predictive routing and preferred agent routing use different methods to identify the agents used for routing.
• Predictive routing uses machine learning to analyze your historical contact center data and predict the best matchup between available agents and interactions to optimize your selected KPI.
• Preferred agent routing routes to preferred agents first, and also according to the bullseye routing rules that you configure. 
• Conditional group routing allows you to share groups of agents between different queues based on the rules you set. A rule is combination of KPI and agent availability.
• Direct routing routes interactions to dedicated personnel in your organization directly. The customer has direct access to the personnel and does not have to wait in the queue to be assisted by the next available agent. 

Predictive routing, preferred agent routing and bullseye routing are optional settings.

Predictive routing

For a full discussion of what predictive routing does and how to start using it, see Predictive routing overview.

Preferred agent routing

Preferred agents are agents that you select to handle interactions based on characteristics you determine, such as skill level, responsibility, or customer relationships. For example, route interactions to agents with higher levels of product knowledge, with historical information about a case, or who are responsible for a particular account.

Preferred agent routing enables you to first route interactions to a designated pool of these preferred agents. If no preferred agents are available based on your configuration rules, Genesys Cloud routes the interaction to a wider pool of agents. 

In Genesys Cloud, you can specify preferred agent routing behavior as you set up queue configuration. You can create up to six rings to expand the selection pool gradually when Genesys Cloud does not find a match on a previous ring. For each ring, you can also route interactions to all preferred agents for a specific length of time. For more information, see Create and configure queues.  In Architect, you can create collections that support up to 20 agent score pairs. For more information, see Transfer to ACD action.

Agent scores allow you to specify the priority of preferred agents. If for example, you have agents that you prefer most, set their scores higher, and set backup agents lower. If you don’t have a preference, set the score to 100. For more information, see Set up preferred agents and scores in Architect.

All interactions associated with preferred agent routing are also associated with a queue, so metrics and details appear in supervisor views, analytics data, and reports.

Set up preferred agents and scores in Architect

In Architect, to influence routing you can create an agent score pair collection that supports up to 20 agent/score pairs.

To create an individual agent score pair value, use the MakeAgentScorePair function. For example, the following expression in a Transfer to ACD action’s Preferred Agent setting creates an agent score pair collection with two agent score pairs. The first user has a score of 100 and the second has a score of 90:

MakeList(
 MakeAgentScorePair(
  FindUserById("<put_user_guid_string_here>"),
  100
 ),
 MakeAgentScorePair(
  FindUserById("<put_user2_guid_string_here>"),
  90
 )
)

You can use a MakeListAgentScorePair function that takes a collection of users along with an integer collection of scores to associate with the users. Following that method, this example generates the same value as the previous expression:

MakeListAgentScorePair(
 MakeList(
  FindUserById("<put_user_guid_string_here>"),
  FindUserById("<put_user2_guid_string_here>")
 ),
 MakeList(
  100,
  90
 )
)

To know the user ID of a user, see Edit a user’s profile.

You can also use an Update Data action to assign variables. Then, use those variables to supply individual user values or user collections, an individual integer score or integer collections, and agent score pairs or collections. For more information about the MakeAgentScorePair, MakeListAgentScorePair, MakeList, FindUserById, and other functions, see Architect’s Expression Help feature. Each function provides more details and examples.

Bullseye routing

With bullseye routing, you can route interactions to agents based on routing rules that you create during queue configuration. With the skills routing method, an agent must have the required skills to be eligible for an interaction. The bullseye routing method is similar. However, it can expand the agent selection pool when no agent with the required skills is available within the configurable amount of time. To ensure compatibility with WEM forecasting and scheduling, you can manage expansion by creating skill expression groups defined with the same skills but with varying levels of proficiency. 

Think of the expansion as a set of concentric rings, similar to the rings on a target, with each ring being a fallback for the previous ring.

Create up to six rings that expand the selection pool gradually and optionally remove specific skills. The selection pool automatically includes preferred agents, regardless of assigned skills. For more information, see Bullseye routing and Create and configure queues.

Conditional group routing

For an overview about how conditional group routing selects agents, see Conditional group routing overview.

Direct routing 

For an overview about how interactions are routed to agents directly, see Direct routing overview.