Create evaluations by agents

Genesys Cloud allows you to create policies and create evaluations based on your selected matching criteria. In a policy, you can use the Create Evaluations per Agents action to create a specific number of evaluations for an agent during a selected time period, such as a week or month. You can also specify which evaluators you want to distribute the evaluations to. For more information about creating a policy with this action, see Create a policy.

Note: When a policy creates an evaluation for this action, the policy selects the last agent that participated in the interaction that meets the specified users, queues, or teams matching the criteria. The policy does not select a user who was a monitor or coach.

Set the Create evaluations per agents action

In this option, you can create a specific number of evaluations for an agent during a selected time period, such as a week or month.

  • You can set Agent connected duration, which defines the period from when the agent joins the interaction until they disconnect. This setting excludes after call work and dialing, but includes hold time.
  • Set the time zone and the evaluation form that you want the action to use.
  • You can specify which evaluators you want to distribute the evaluations to.
  • You can set a Sampling criteria as a fixed number of evaluations per period, or as a percentage of interactions with a minimum number of evaluations.

Details about the Create evaluations per agents action

  • The policy must include users, teams, or queues as part of the matching criteria.
  • Genesys Cloud distributes evaluations evenly among the selected evaluators. If the number of evaluations does not divide evenly among the evaluators, then Genesys Cloud randomly distributes any remaining evaluation to the evaluators.
  • You can choose to create up to the maximum number of evaluations per agent per time period.
    The maximum number of evaluations that you can create per agent per time interval:
    • Day – 50
    • Week – 175
    • Month – 700
  • Genesys Cloud tries to create evaluations evenly throughout the time period. If there are too few evaluations at internal checkpoints during the time period, Genesys Cloud creates evaluations for older interactions at that checkpoint instead of creating more evaluations at the end of the time period to meet the specified number of interactions. 
  • Genesys Cloud uses time zones to determine when to create evaluations, usually at the start of the next day.
  • If the customer was the last participant on the interaction, Genesys Cloud does not create evaluations for email and messaging interactions.
  • The message threading timeline determines how long message conversations remain open.
  • The time period length refers to a calendar unit, such as Sunday through Saturday for a week or all of January for a month. This setting does not refer to a number of days, such as seven days for a week or 30 days for a month, from the time of policy creation. 
  • When you create a policy with this action, for the first time period, Genesys Cloud bases the number of evaluations it creates on the amount of time remaining in the time period.
    Example: If I create a policy on Wednesday at 8:00 AM that specifies 10 evaluations per agent per week, then the first time period, a week, is almost half way complete. For that first week, Genesys Cloud then creates only half the number of evaluations specified, five evaluations per agent. Every week after that is a full week, so Genesys Cloud creates 10 evaluations per week per agent for each following week.
  • If you change a policy’s parameters can have an impact on ongoing actions, based on what you change:
    •  If you add and remove agents that does not affect the entire policy. Genesys Cloud creates evaluations for new agents based on the amount of time remaining in the time period.
    • If you change other parameters, the effect similar to turning the policy off and on. Genesys Cloud does not include any previously created evaluations in the total count and instead bases the number of evaluations on the amount of time remaining in the time period. 

Pro-rating

If an agent is included as part of a team or a queue, but not explicitly included as a user in the policy’s matching criteria, pro-rating rules apply.

Pro-rating rules mean that when an agent matches the policy for the first time, Genesys Cloud pro-rates the number of evaluations created for the agent in this first time interval proportionally from the date and time when the first interaction happened.

Example scenario
  1. A policy is created with a Create evaluations by agents action with a Monthly time interval, and the number of interactions to evaluate per agent is set to 10.
  2. The policy uses a team for the matching criteria.
  3. A new agent, Anna is added to that team on June 1, but she does not get any interactions that match the policy until June 15. Since Anna is not explicitly listed in the policy as a specific user, her evaluations for June are pro-rated. In this way, Anna receives only 5 evaluations for June because of the pro-rating, according to the calculation:

    (15 days left in June / 30 days in June) * 10 evaluations for the time interval = 5 evaluations.
    The number of evaluations are rounded down.

Pro-rating and agent inactivity

Pro-rating applies to Monthly, Weekly, and Daily time intervals.

Pro-rating also applies if an agent has an extended period of inactivity in the policy; for example, in case of a leave of absence or long vacation. If pro-rating rules apply to an agent and they have not had any interactions for a certain amount of time that match the policy, then their next interaction that does match the policy are subject to pro-rating rules.

The amount of inactivity that causes pro-rating depends on the time interval of the policy:

* If the time interval is Monthly, pro-rating is applied after 2+ months of inactivity.
* If the time interval is Weekly, pro-rating is applied after 5+ weeks of inactivity.
* If the time interval is Daily, pro-rating is applied after 30+ days of inactivity.

Once pro-rating has been applied to an agent for a time interval, it is not applied to future time intervals for that agent as long as they do not have inactivity in the policy as described above.

Example scenario

Let’s look at the above example where agent Anna only received 5 evaluations for June.

For the month of July, agent Anna took an extended vacation from July 1 – July 19, and the first interaction she has that matches the policy is on July 20; but Anna also has more than 10 interactions that match the policy from July 20 through July 31. In this case, Anna receives her full quota of 10 evaluations for the month of July, because she had interactions in the previous month and is not pro-rated. Anna is never pro-rated in the future, unless she has no interactions that match the policy for 2 or more consecutive months.