Skip to end of metadata
Go to start of metadata

You are viewing an old version of this content. View the current version.

Compare with Current View Version History

Version 1 Next »

Whether configuring a Q call flow, or troubleshooting a customer incident related to a Q call flow, it is important that we understand how Sharpen connects interactions and agents at the right time.

Sharpen Q is designed to pair the right interaction with the right agent. This is accomplished by utilizing several parameters which determine how valuable an interaction is, and how available an agent is. Before getting started, it is important to be familiar with some key terms.

Terms:

Available Agent - Sharpen Q user who is...

  • Assigned an Empower (CustomerCare) license

  • Logged into and configured for a Q associated with the interaction path

  • Un-paused

  • Not occupied on a call

  • Not in wrap-up

  • Does not have an interaction on hold

  • Has appropriate required skills (if configured on Q)

Q - Entry point, holding place, or parking lot for interactions which are to be automatically assigned to available agents.

Queue Weight - Priority level assigned to a Q. This dictates which Q gets priority when evaluating multiple interactions waiting across multiple Qs.  

Q Call Point - Priority level assigned to an Interaction which is done so at some point during the inbound call flow (Inbound Route, Advanced Workflow, Logic & Actions). This value prioritizes which interaction is chosen among other waiting interactions.

Q Skill - Configuration associated to interactions and assigned to agents allowing for preferred or gated routing.

Required Skill - Skill which requires an agent, configured for such skill, is available before routing for an interaction which is tagged tagged with the skill.

Skill Weight - Priority level assigned to a Skill. The skill weight is then applied to available agents, qualifying the best available agent for calls with skills.

  • It should be noted that Sharpen does not apply unique skill proficiencies by user. Skill weights are applied on the skill and relate to users with the assigned skill, universally.

How are interactions assigned?

The natural state of interaction handling suggests there are really two primary scenarios of queueing involved with connecting the right caller to the right agent.

  • An Agent becomes available and needs to be assigned a waiting interaction

  • A new interaction enters Q and needs to be assigned to an available agent

Sharpen exists in a perpetual state of evaluation of the best interactions and agents to solve both situations. This perpetual state involves the periodic execution of a hierarchy of selection jobs to identify waiting interaction and available agents in a prioritized list. We then select the best item from the list and send it to the right place. The sequence goes like this...

Queue Selection

During this selection process, the system is prioritizing which Q to pick an interaction from. It does so by evaluating these criteria

  • Are there available agents?

  • Are there waiting interactions?

  • What is the assigned Queue Weight?

Based on the returns, the list is ordered by weight in a descending fashion, and the top Q is selected to then identify which interaction to choose.

Interaction Selection

The interaction selection process fires off immediately after we’ve chosen which Q to evaluate. The job of this selection is to evaluate all interactions in Q to choose the highest priority. The following criteria are evaluated to complete this job.

  • Is the interaction awaiting agent assignment?

  • How many queue points are assigned?

  • In what order did the interaction enter Q?

Based on the returns, the job filters by all which are awaiting agent assignment, and then sorts first by the order of Q entry (QCMID) in an ascending fashion, and then by the highest value of queue points in a descending fashion. This allows for the highest point value to always outweigh interaction age. If the competing interactions match on points, the oldest interaction is selected. If there are no points, the oldest interaction is selected.

Agent Selection

Finally, once the interaction has been chosen, the Q system looks to find the most available agent. For interactions which do not involve skills, the mechanism is simple. We lookup the available agent who has the longest duration since their last interaction. The previously noted “Available Agent” criteria must be met for the agent to be considered “available”. 

When skills are involved, the same criteria are involved, but skills influence the sorting of available agents to make the best choice based on the count of skills and total value of skill weight which applies to the interaction selected. The system will sort the available agents by the number of skills (descending), then by the sum of the skill weights relating to those skills (descending), then finally by the last time the agent had a call (ascending). In practical terms, this means the agent who has not taken an interaction in a while and possesses the most skills at the highest weight, will get the interaction first. If the skills are set to “required,” then only agents with those skills will be evaluated.

  • No labels