Genesys Cloud - Main

The IVR inbound call flow will route calls to about 10 ACD queues, and other 10 queues are for transfer purpose (not from IVR). Total 150 agents across these 20 queues. Some agents are in two different queues.
All queues have equal priority. But individual agents have different priority requirement. Agent1 need anwser Queue1 calls before Queue2. Agent2 need answer Queue2 first. Agent3 has higher priority on Queue3 calls than Queue1, and so on. Can Genesys routing achieve this goal? How to config?

The IVR inbound call flow will route calls to about 10 ACD queues, and other 10 queues are for transfer purpose (not from IVR). Total 150 agents across these 20 queues. Some agents are in two different queues.
All queues have equal priority. But individual agents have different priority requirement. Agent1 need anwser Queue1 calls before Queue2. Agent2 need answer Queue2 first. Agent3 has higher priority on Queue3 calls than Queue1, and so on. Can Genesys routing achieve this goal? How to config?

The IVR inbound call flow will route calls to about 10 ACD queues, and other 10 queues are for transfer purpose (not from IVR). Total 150 agents across these 20 queues. Some agents are in two different queues.
All queues have equal priority. But individual agents have different priority requirement. Agent1 need anwser Queue1 calls before Queue2. Agent2 need answer Queue2 first. Agent3 has higher priority on Queue3 calls than Queue1, and so on. Can Genesys routing achieve this goal? How to config?

OK, so first up, Priorities are set on interactions, not Queues or Agents. What they actually do is to place an offset on the arrival time of the interaction that is used in the selection calculation (so it appears that the interaction has been waiting for a longer, or shorter, time than it actually has.)

Now, to the case in point. As far as I can see, Genesys Cloud provides plenty of scope for tuning which agent gets selected for an interaction, but much less when it comes to selecting an interaction for an agent. Bullseye won't work for this reason.

I initially thought Skills, but I'm not certain they factor in to the interaction selection calculation (they used to in PureConnect). Can anyone from Genesys explain how Genesys Cloud chooses between valid interactions for an agent? Is it purely done on time (with Priority potentially modifying that) or are Skill Proficiencies somehow incorporated?

Also, you didn't indicate what you want to happen to interactions that have been waiting a long time? Say, for example, There is an Interaction in Queue 1 that has been waiting 5 seconds, but one in Queue 2 that has been waiting for 2 minutes. Should Agent 1 still be given the Queue 1 interaction? Usually, in systems like this, you have to have the calculation factor in wait time to account for this.

The IVR inbound call flow will route calls to about 10 ACD queues, and other 10 queues are for transfer purpose (not from IVR). Total 150 agents across these 20 queues. Some agents are in two different queues.
All queues have equal priority. But individual agents have different priority requirement. Agent1 need anwser Queue1 calls before Queue2. Agent2 need answer Queue2 first. Agent3 has higher priority on Queue3 calls than Queue1, and so on. Can Genesys routing achieve this goal? How to config?

Thanks for all the replies. In this example, There is an Interaction in Queue 1 that has been waiting 5 seconds, but one in Queue 2 that has been waiting for 2 minutes. Should Agent 1 still be given the Queue 1 interaction? Yes. Agent1 always handles Q1 if there is call in Q1.

We try to use skills, but in Blind or Consult Transfer other queues, skills will cause issue to route calls to another queue (different queue has different skill). Also in transfer calls, agent skill proficiency seems not factor into agent selection, even we use Best available routing method.

OK, so first up, Priorities are set on interactions, not Queues or Agents. What they actually do is to place an offset on the arrival time of the interaction that is used in the selection calculation (so it appears that the interaction has been waiting for a longer, or shorter, time than it actually has.)

Now, to the case in point. As far as I can see, Genesys Cloud provides plenty of scope for tuning which agent gets selected for an interaction, but much less when it comes to selecting an interaction for an agent. Bullseye won't work for this reason.

I initially thought Skills, but I'm not certain they factor in to the interaction selection calculation (they used to in PureConnect). Can anyone from Genesys explain how Genesys Cloud chooses between valid interactions for an agent? Is it purely done on time (with Priority potentially modifying that) or are Skill Proficiencies somehow incorporated?

Also, you didn't indicate what you want to happen to interactions that have been waiting a long time? Say, for example, There is an Interaction in Queue 1 that has been waiting 5 seconds, but one in Queue 2 that has been waiting for 2 minutes. Should Agent 1 still be given the Queue 1 interaction? Usually, in systems like this, you have to have the calculation factor in wait time to account for this.

The IVR inbound call flow will route calls to about 10 ACD queues, and other 10 queues are for transfer purpose (not from IVR). Total 150 agents across these 20 queues. Some agents are in two different queues.
All queues have equal priority. But individual agents have different priority requirement. Agent1 need anwser Queue1 calls before Queue2. Agent2 need answer Queue2 first. Agent3 has higher priority on Queue3 calls than Queue1, and so on. Can Genesys routing achieve this goal? How to config?

OK, that is unusual, but OK!

Firstly, the transfer. The easiest way to deal with that, where skills are concerned, is to set up a dummy DNIS, which routes to a flow, which specifies the skills and so on and then have the agents transfer to that dummy DNIS rather than directly to the Queue. Many sites then use External Contacts as a way to provide a meaningful name for those DNIS's. There are a couple of ideas open (here and here) to allow transfers directly to Flows, which would make it much more straightforward. Please go and up-vote them!

Secondly, and this is where we need Genesys to weigh in, I am not sure if Skill Proficiency is used when selecting an interaction to give to a user. It certainly is when selecting a user for the interaction. If it is, then the easiest way would be to have a skill for each queue, require the skill when the interaction goes to the queue and then set the proficiencies as you describe. That is certainly how you could've done it in PureConnect! @Melissa Bailey, can you provide any insights here? You would also need to somehow make sure the in-queue wait time does not get considered when selecting an interaction (or at least, is only used as a tie-break if it's selecting two interactions that are otherwise the same.) Again, relatively easy in PureConnect, but I don't know if this kind of functionality has been ported to Genesys Cloud yet.

As I said before, Genesys seem to have focused primarily on making sure the most appropriate agent for an interaction is selected, not so much on making sure the most appropriate interaction for an agent is delivered!

Thanks for all the replies. In this example, There is an Interaction in Queue 1 that has been waiting 5 seconds, but one in Queue 2 that has been waiting for 2 minutes. Should Agent 1 still be given the Queue 1 interaction? Yes. Agent1 always handles Q1 if there is call in Q1.

We try to use skills, but in Blind or Consult Transfer other queues, skills will cause issue to route calls to another queue (different queue has different skill). Also in transfer calls, agent skill proficiency seems not factor into agent selection, even we use Best available routing method.

OK, so first up, Priorities are set on interactions, not Queues or Agents. What they actually do is to place an offset on the arrival time of the interaction that is used in the selection calculation (so it appears that the interaction has been waiting for a longer, or shorter, time than it actually has.)

Now, to the case in point. As far as I can see, Genesys Cloud provides plenty of scope for tuning which agent gets selected for an interaction, but much less when it comes to selecting an interaction for an agent. Bullseye won't work for this reason.

I initially thought Skills, but I'm not certain they factor in to the interaction selection calculation (they used to in PureConnect). Can anyone from Genesys explain how Genesys Cloud chooses between valid interactions for an agent? Is it purely done on time (with Priority potentially modifying that) or are Skill Proficiencies somehow incorporated?

Also, you didn't indicate what you want to happen to interactions that have been waiting a long time? Say, for example, There is an Interaction in Queue 1 that has been waiting 5 seconds, but one in Queue 2 that has been waiting for 2 minutes. Should Agent 1 still be given the Queue 1 interaction? Usually, in systems like this, you have to have the calculation factor in wait time to account for this.

The IVR inbound call flow will route calls to about 10 ACD queues, and other 10 queues are for transfer purpose (not from IVR). Total 150 agents across these 20 queues. Some agents are in two different queues.
All queues have equal priority. But individual agents have different priority requirement. Agent1 need anwser Queue1 calls before Queue2. Agent2 need answer Queue2 first. Agent3 has higher priority on Queue3 calls than Queue1, and so on. Can Genesys routing achieve this goal? How to config?

>>I am not sure if Skill Proficiency is used when selecting an interaction to give to a user. It certainly is when selecting a user for the interaction.
Not when in conversation surplus.  See cheat sheet below.

OK, that is unusual, but OK!

Firstly, the transfer. The easiest way to deal with that, where skills are concerned, is to set up a dummy DNIS, which routes to a flow, which specifies the skills and so on and then have the agents transfer to that dummy DNIS rather than directly to the Queue. Many sites then use External Contacts as a way to provide a meaningful name for those DNIS's. There are a couple of ideas open (here and here) to allow transfers directly to Flows, which would make it much more straightforward. Please go and up-vote them!

Secondly, and this is where we need Genesys to weigh in, I am not sure if Skill Proficiency is used when selecting an interaction to give to a user. It certainly is when selecting a user for the interaction. If it is, then the easiest way would be to have a skill for each queue, require the skill when the interaction goes to the queue and then set the proficiencies as you describe. That is certainly how you could've done it in PureConnect! @Melissa Bailey, can you provide any insights here? You would also need to somehow make sure the in-queue wait time does not get considered when selecting an interaction (or at least, is only used as a tie-break if it's selecting two interactions that are otherwise the same.) Again, relatively easy in PureConnect, but I don't know if this kind of functionality has been ported to Genesys Cloud yet.

As I said before, Genesys seem to have focused primarily on making sure the most appropriate agent for an interaction is selected, not so much on making sure the most appropriate interaction for an agent is delivered!

Thanks for all the replies. In this example, There is an Interaction in Queue 1 that has been waiting 5 seconds, but one in Queue 2 that has been waiting for 2 minutes. Should Agent 1 still be given the Queue 1 interaction? Yes. Agent1 always handles Q1 if there is call in Q1.

We try to use skills, but in Blind or Consult Transfer other queues, skills will cause issue to route calls to another queue (different queue has different skill). Also in transfer calls, agent skill proficiency seems not factor into agent selection, even we use Best available routing method.

OK, so first up, Priorities are set on interactions, not Queues or Agents. What they actually do is to place an offset on the arrival time of the interaction that is used in the selection calculation (so it appears that the interaction has been waiting for a longer, or shorter, time than it actually has.)

Now, to the case in point. As far as I can see, Genesys Cloud provides plenty of scope for tuning which agent gets selected for an interaction, but much less when it comes to selecting an interaction for an agent. Bullseye won't work for this reason.

I initially thought Skills, but I'm not certain they factor in to the interaction selection calculation (they used to in PureConnect). Can anyone from Genesys explain how Genesys Cloud chooses between valid interactions for an agent? Is it purely done on time (with Priority potentially modifying that) or are Skill Proficiencies somehow incorporated?

Also, you didn't indicate what you want to happen to interactions that have been waiting a long time? Say, for example, There is an Interaction in Queue 1 that has been waiting 5 seconds, but one in Queue 2 that has been waiting for 2 minutes. Should Agent 1 still be given the Queue 1 interaction? Usually, in systems like this, you have to have the calculation factor in wait time to account for this.

The IVR inbound call flow will route calls to about 10 ACD queues, and other 10 queues are for transfer purpose (not from IVR). Total 150 agents across these 20 queues. Some agents are in two different queues.
All queues have equal priority. But individual agents have different priority requirement. Agent1 need anwser Queue1 calls before Queue2. Agent2 need answer Queue2 first. Agent3 has higher priority on Queue3 calls than Queue1, and so on. Can Genesys routing achieve this goal? How to config?

>>I am not sure if Skill Proficiency is used when selecting an interaction to give to a user. It certainly is when selecting a user for the interaction.
Not when in conversation surplus.  See cheat sheet below.

OK, that is unusual, but OK!

Firstly, the transfer. The easiest way to deal with that, where skills are concerned, is to set up a dummy DNIS, which routes to a flow, which specifies the skills and so on and then have the agents transfer to that dummy DNIS rather than directly to the Queue. Many sites then use External Contacts as a way to provide a meaningful name for those DNIS's. There are a couple of ideas open (here and here) to allow transfers directly to Flows, which would make it much more straightforward. Please go and up-vote them!

Secondly, and this is where we need Genesys to weigh in, I am not sure if Skill Proficiency is used when selecting an interaction to give to a user. It certainly is when selecting a user for the interaction. If it is, then the easiest way would be to have a skill for each queue, require the skill when the interaction goes to the queue and then set the proficiencies as you describe. That is certainly how you could've done it in PureConnect! @Melissa Bailey, can you provide any insights here? You would also need to somehow make sure the in-queue wait time does not get considered when selecting an interaction (or at least, is only used as a tie-break if it's selecting two interactions that are otherwise the same.) Again, relatively easy in PureConnect, but I don't know if this kind of functionality has been ported to Genesys Cloud yet.

As I said before, Genesys seem to have focused primarily on making sure the most appropriate agent for an interaction is selected, not so much on making sure the most appropriate interaction for an agent is delivered!

Thanks for all the replies. In this example, There is an Interaction in Queue 1 that has been waiting 5 seconds, but one in Queue 2 that has been waiting for 2 minutes. Should Agent 1 still be given the Queue 1 interaction? Yes. Agent1 always handles Q1 if there is call in Q1.

We try to use skills, but in Blind or Consult Transfer other queues, skills will cause issue to route calls to another queue (different queue has different skill). Also in transfer calls, agent skill proficiency seems not factor into agent selection, even we use Best available routing method.

OK, so first up, Priorities are set on interactions, not Queues or Agents. What they actually do is to place an offset on the arrival time of the interaction that is used in the selection calculation (so it appears that the interaction has been waiting for a longer, or shorter, time than it actually has.)

Now, to the case in point. As far as I can see, Genesys Cloud provides plenty of scope for tuning which agent gets selected for an interaction, but much less when it comes to selecting an interaction for an agent. Bullseye won't work for this reason.

I initially thought Skills, but I'm not certain they factor in to the interaction selection calculation (they used to in PureConnect). Can anyone from Genesys explain how Genesys Cloud chooses between valid interactions for an agent? Is it purely done on time (with Priority potentially modifying that) or are Skill Proficiencies somehow incorporated?

Also, you didn't indicate what you want to happen to interactions that have been waiting a long time? Say, for example, There is an Interaction in Queue 1 that has been waiting 5 seconds, but one in Queue 2 that has been waiting for 2 minutes. Should Agent 1 still be given the Queue 1 interaction? Usually, in systems like this, you have to have the calculation factor in wait time to account for this.

The IVR inbound call flow will route calls to about 10 ACD queues, and other 10 queues are for transfer purpose (not from IVR). Total 150 agents across these 20 queues. Some agents are in two different queues.
All queues have equal priority. But individual agents have different priority requirement. Agent1 need anwser Queue1 calls before Queue2. Agent2 need answer Queue2 first. Agent3 has higher priority on Queue3 calls than Queue1, and so on. Can Genesys routing achieve this goal? How to config?

Thanks for all the replies. In this example, There is an Interaction in Queue 1 that has been waiting 5 seconds, but one in Queue 2 that has been waiting for 2 minutes. Should Agent 1 still be given the Queue 1 interaction? Yes. Agent1 always handles Q1 if there is call in Q1.

We try to use skills, but in Blind or Consult Transfer other queues, skills will cause issue to route calls to another queue (different queue has different skill). Also in transfer calls, agent skill proficiency seems not factor into agent selection, even we use Best available routing method.

OK, so first up, Priorities are set on interactions, not Queues or Agents. What they actually do is to place an offset on the arrival time of the interaction that is used in the selection calculation (so it appears that the interaction has been waiting for a longer, or shorter, time than it actually has.)

Now, to the case in point. As far as I can see, Genesys Cloud provides plenty of scope for tuning which agent gets selected for an interaction, but much less when it comes to selecting an interaction for an agent. Bullseye won't work for this reason.

I initially thought Skills, but I'm not certain they factor in to the interaction selection calculation (they used to in PureConnect). Can anyone from Genesys explain how Genesys Cloud chooses between valid interactions for an agent? Is it purely done on time (with Priority potentially modifying that) or are Skill Proficiencies somehow incorporated?

Also, you didn't indicate what you want to happen to interactions that have been waiting a long time? Say, for example, There is an Interaction in Queue 1 that has been waiting 5 seconds, but one in Queue 2 that has been waiting for 2 minutes. Should Agent 1 still be given the Queue 1 interaction? Usually, in systems like this, you have to have the calculation factor in wait time to account for this.

The IVR inbound call flow will route calls to about 10 ACD queues, and other 10 queues are for transfer purpose (not from IVR). Total 150 agents across these 20 queues. Some agents are in two different queues.
All queues have equal priority. But individual agents have different priority requirement. Agent1 need anwser Queue1 calls before Queue2. Agent2 need answer Queue2 first. Agent3 has higher priority on Queue3 calls than Queue1, and so on. Can Genesys routing achieve this goal? How to config?

@Lei Cao 
Blind transfers: can you please check the setting on the Org level: Stripping Skills on Blind Transfer? 

Thanks for all the replies. In this example, There is an Interaction in Queue 1 that has been waiting 5 seconds, but one in Queue 2 that has been waiting for 2 minutes. Should Agent 1 still be given the Queue 1 interaction? Yes. Agent1 always handles Q1 if there is call in Q1.

We try to use skills, but in Blind or Consult Transfer other queues, skills will cause issue to route calls to another queue (different queue has different skill). Also in transfer calls, agent skill proficiency seems not factor into agent selection, even we use Best available routing method.

OK, so first up, Priorities are set on interactions, not Queues or Agents. What they actually do is to place an offset on the arrival time of the interaction that is used in the selection calculation (so it appears that the interaction has been waiting for a longer, or shorter, time than it actually has.)

Now, to the case in point. As far as I can see, Genesys Cloud provides plenty of scope for tuning which agent gets selected for an interaction, but much less when it comes to selecting an interaction for an agent. Bullseye won't work for this reason.

I initially thought Skills, but I'm not certain they factor in to the interaction selection calculation (they used to in PureConnect). Can anyone from Genesys explain how Genesys Cloud chooses between valid interactions for an agent? Is it purely done on time (with Priority potentially modifying that) or are Skill Proficiencies somehow incorporated?

Also, you didn't indicate what you want to happen to interactions that have been waiting a long time? Say, for example, There is an Interaction in Queue 1 that has been waiting 5 seconds, but one in Queue 2 that has been waiting for 2 minutes. Should Agent 1 still be given the Queue 1 interaction? Usually, in systems like this, you have to have the calculation factor in wait time to account for this.

The IVR inbound call flow will route calls to about 10 ACD queues, and other 10 queues are for transfer purpose (not from IVR). Total 150 agents across these 20 queues. Some agents are in two different queues.
All queues have equal priority. But individual agents have different priority requirement. Agent1 need anwser Queue1 calls before Queue2. Agent2 need answer Queue2 first. Agent3 has higher priority on Queue3 calls than Queue1, and so on. Can Genesys routing achieve this goal? How to config?

@Lei Cao 
Blind transfers: can you please check the setting on the Org level: Stripping Skills on Blind Transfer? 

Thanks for all the replies. In this example, There is an Interaction in Queue 1 that has been waiting 5 seconds, but one in Queue 2 that has been waiting for 2 minutes. Should Agent 1 still be given the Queue 1 interaction? Yes. Agent1 always handles Q1 if there is call in Q1.

We try to use skills, but in Blind or Consult Transfer other queues, skills will cause issue to route calls to another queue (different queue has different skill). Also in transfer calls, agent skill proficiency seems not factor into agent selection, even we use Best available routing method.

OK, so first up, Priorities are set on interactions, not Queues or Agents. What they actually do is to place an offset on the arrival time of the interaction that is used in the selection calculation (so it appears that the interaction has been waiting for a longer, or shorter, time than it actually has.)

Now, to the case in point. As far as I can see, Genesys Cloud provides plenty of scope for tuning which agent gets selected for an interaction, but much less when it comes to selecting an interaction for an agent. Bullseye won't work for this reason.

I initially thought Skills, but I'm not certain they factor in to the interaction selection calculation (they used to in PureConnect). Can anyone from Genesys explain how Genesys Cloud chooses between valid interactions for an agent? Is it purely done on time (with Priority potentially modifying that) or are Skill Proficiencies somehow incorporated?

Also, you didn't indicate what you want to happen to interactions that have been waiting a long time? Say, for example, There is an Interaction in Queue 1 that has been waiting 5 seconds, but one in Queue 2 that has been waiting for 2 minutes. Should Agent 1 still be given the Queue 1 interaction? Usually, in systems like this, you have to have the calculation factor in wait time to account for this.

The IVR inbound call flow will route calls to about 10 ACD queues, and other 10 queues are for transfer purpose (not from IVR). Total 150 agents across these 20 queues. Some agents are in two different queues.
All queues have equal priority. But individual agents have different priority requirement. Agent1 need anwser Queue1 calls before Queue2. Agent2 need answer Queue2 first. Agent3 has higher priority on Queue3 calls than Queue1, and so on. Can Genesys routing achieve this goal? How to config?

@Tatjana Knezevic , unfortunately, I don't think that would help. For this to work, they would need to remove one set of skills and add a different one. That would require a Flow....

@Lei Cao 
Blind transfers: can you please check the setting on the Org level: Stripping Skills on Blind Transfer? 

Thanks for all the replies. In this example, There is an Interaction in Queue 1 that has been waiting 5 seconds, but one in Queue 2 that has been waiting for 2 minutes. Should Agent 1 still be given the Queue 1 interaction? Yes. Agent1 always handles Q1 if there is call in Q1.

We try to use skills, but in Blind or Consult Transfer other queues, skills will cause issue to route calls to another queue (different queue has different skill). Also in transfer calls, agent skill proficiency seems not factor into agent selection, even we use Best available routing method.

OK, so first up, Priorities are set on interactions, not Queues or Agents. What they actually do is to place an offset on the arrival time of the interaction that is used in the selection calculation (so it appears that the interaction has been waiting for a longer, or shorter, time than it actually has.)

Now, to the case in point. As far as I can see, Genesys Cloud provides plenty of scope for tuning which agent gets selected for an interaction, but much less when it comes to selecting an interaction for an agent. Bullseye won't work for this reason.

I initially thought Skills, but I'm not certain they factor in to the interaction selection calculation (they used to in PureConnect). Can anyone from Genesys explain how Genesys Cloud chooses between valid interactions for an agent? Is it purely done on time (with Priority potentially modifying that) or are Skill Proficiencies somehow incorporated?

Also, you didn't indicate what you want to happen to interactions that have been waiting a long time? Say, for example, There is an Interaction in Queue 1 that has been waiting 5 seconds, but one in Queue 2 that has been waiting for 2 minutes. Should Agent 1 still be given the Queue 1 interaction? Usually, in systems like this, you have to have the calculation factor in wait time to account for this.

The IVR inbound call flow will route calls to about 10 ACD queues, and other 10 queues are for transfer purpose (not from IVR). Total 150 agents across these 20 queues. Some agents are in two different queues.
All queues have equal priority. But individual agents have different priority requirement. Agent1 need anwser Queue1 calls before Queue2. Agent2 need answer Queue2 first. Agent3 has higher priority on Queue3 calls than Queue1, and so on. Can Genesys routing achieve this goal? How to config?

Hi @Paul Simpson and @Lei Cao. 

Thank you. 

Please note, the comment regarding Stripping Skills on Blind Transfer was a question, not a suggestion :-) Lei posted "Also in transfer calls, agent skill proficiency seems not factor into agent selection, even we use Best available routing method." and I suggested checking the setting, not to change it. 

@Paul Simpson

@Tatjana Knezevic , unfortunately, I don't think that would help. For this to work, they would need to remove one set of skills and add a different one. That would require a Flow....

@Lei Cao 
Blind transfers: can you please check the setting on the Org level: Stripping Skills on Blind Transfer? 

Thanks for all the replies. In this example, There is an Interaction in Queue 1 that has been waiting 5 seconds, but one in Queue 2 that has been waiting for 2 minutes. Should Agent 1 still be given the Queue 1 interaction? Yes. Agent1 always handles Q1 if there is call in Q1.

We try to use skills, but in Blind or Consult Transfer other queues, skills will cause issue to route calls to another queue (different queue has different skill). Also in transfer calls, agent skill proficiency seems not factor into agent selection, even we use Best available routing method.

OK, so first up, Priorities are set on interactions, not Queues or Agents. What they actually do is to place an offset on the arrival time of the interaction that is used in the selection calculation (so it appears that the interaction has been waiting for a longer, or shorter, time than it actually has.)

Now, to the case in point. As far as I can see, Genesys Cloud provides plenty of scope for tuning which agent gets selected for an interaction, but much less when it comes to selecting an interaction for an agent. Bullseye won't work for this reason.

I initially thought Skills, but I'm not certain they factor in to the interaction selection calculation (they used to in PureConnect). Can anyone from Genesys explain how Genesys Cloud chooses between valid interactions for an agent? Is it purely done on time (with Priority potentially modifying that) or are Skill Proficiencies somehow incorporated?

Also, you didn't indicate what you want to happen to interactions that have been waiting a long time? Say, for example, There is an Interaction in Queue 1 that has been waiting 5 seconds, but one in Queue 2 that has been waiting for 2 minutes. Should Agent 1 still be given the Queue 1 interaction? Usually, in systems like this, you have to have the calculation factor in wait time to account for this.

The IVR inbound call flow will route calls to about 10 ACD queues, and other 10 queues are for transfer purpose (not from IVR). Total 150 agents across these 20 queues. Some agents are in two different queues.
All queues have equal priority. But individual agents have different priority requirement. Agent1 need anwser Queue1 calls before Queue2. Agent2 need answer Queue2 first. Agent3 has higher priority on Queue3 calls than Queue1, and so on. Can Genesys routing achieve this goal? How to config?

Hi everyone - a lot going on this this post so I'm going to try to break down a few different comments/thoughts/options you have here.  Thanks to everyone for adding comments so far - all helping to get Lei pushed in the right direction.

• To start with, as David points out - we will use the agent's proficiency score to make a routing decision assuming the evaluation method set on the queue is set to 'Best Available Skills' (All Skills Matching and Ignore Skills do NOT use the proficiency score.)  Lei - it looks like you are set up correctly there so we can check that box.
• If all agents can work all queues, but you want to prioritize some work over others, this is really where skills and proficiencies should come into play.  If every agent can, at some point, answer every call, then its important that you give each agent all skills and then 'rank' them using proficiency values.  (Note:  this is the most effective way to work with our WFM tools for forecasting and scheduling).
• One you have all the agent's skills and rated appropriately - now you can start to ask questions around routing.  If your goal is to ensure that there are 'primary' agents for each queue, I'd recommend you create a skill for each queue and rate those agents who are always primary to 5 stars and those who should only be added when other queues are struggling to 0 or 1 stars (literally the lowest ranking).
• Lei, it sounds like you want to make sure the best agents for each queue/skill have the best shot at getting the calls for their primary queue and really only want others to help out when needed so, what you can do is use Bullseye routing...and here's how!  (Paul - this is a new one so let me know if this makes sense - happy to discuss further!)
• Step 1:  Define skill expression groups:  You can now 'group' agents together by defining a skill expression group…for example:  Skill A = 5 or Skill B = 5.  (This means find all agents will skill A OR skill B who have a proficiency equal to 5).  This will allow you to get all your TOP agents into a group and use them as a routing target.  (I'll elaborate on how to add them to queues in a moment). 
• From here – create additional groups where the skills stay the same, but the proficiency value goes down (e.g. Skill A = 4 or Skill B = 4).
• Step 2:  Add these groups as members of the queue.  In the Queue administration screen, click on the 'members' tab and then look for the 'Groups' tab.   On the right-hand side select the 'Skill Group' option in the dropdown menu and select all of the skill expression groups you want to route to.
• Step 3:  Go to the Routing tab and select bullseye routing and add the number of rings (up to 5) you want to define and add the duration of time to wait between each ring.  (NOTE:  The maximum amount of time you want to use really should be UNDER your patience threshold for that queue (meaning how long customers stay on hold before they abandon) for synchronous channels like Calls and Chats, but you can extend the timer quite a way out for digital media like emails as they can sit much longer but make sure whatever values you use allow you to meet your service goals.
• Step 4:  Go back to the members tab, click on groups again and assign the appropriate bullseye ring number to each group.   Assuming you put all the agents with proficiency of 5 in the first ring, and proficiency 4 in the next ring and so on, you will effectively give the best agents a shot at answering their preferred interactions ahead of any other agent.   NOTE:  This does assume you have enough availability to be able to address this – if the 5-star agents are not available – they may not answer the interaction but that is likely still preferable than the customer abandoning.

I will admit that this isn't a perfect scenario for what you describe, Lei – it does not guarantee, with 100% certainty that the primary agent will always answer interactions for which they are primary – but it should give you a much better chance to meet that need and still service customers promptly and/or meet your SLAs.

Here are a couple screenshots to outline what I'm talking about.

Skill Expression Groups – (screenshot)

Assign Skill Groups to a Queue:

Bullseye Routing Expansion via Groups:

A lot to digest - but hopefully this helps.  I'll be back with a follow-up post around transfers and skills b/c, unfortunately saving/stripping skills works differently with different media types so its not as clear as one might hope!

Feel free to post back with comments and questions.

Hi @Paul Simpson and @Lei Cao. 

Thank you. 

Please note, the comment regarding Stripping Skills on Blind Transfer was a question, not a suggestion :-) Lei posted "Also in transfer calls, agent skill proficiency seems not factor into agent selection, even we use Best available routing method." and I suggested checking the setting, not to change it. 

@Paul Simpson

@Tatjana Knezevic , unfortunately, I don't think that would help. For this to work, they would need to remove one set of skills and add a different one. That would require a Flow....

@Lei Cao 
Blind transfers: can you please check the setting on the Org level: Stripping Skills on Blind Transfer? 

Thanks for all the replies. In this example, There is an Interaction in Queue 1 that has been waiting 5 seconds, but one in Queue 2 that has been waiting for 2 minutes. Should Agent 1 still be given the Queue 1 interaction? Yes. Agent1 always handles Q1 if there is call in Q1.

We try to use skills, but in Blind or Consult Transfer other queues, skills will cause issue to route calls to another queue (different queue has different skill). Also in transfer calls, agent skill proficiency seems not factor into agent selection, even we use Best available routing method.

OK, so first up, Priorities are set on interactions, not Queues or Agents. What they actually do is to place an offset on the arrival time of the interaction that is used in the selection calculation (so it appears that the interaction has been waiting for a longer, or shorter, time than it actually has.)

Now, to the case in point. As far as I can see, Genesys Cloud provides plenty of scope for tuning which agent gets selected for an interaction, but much less when it comes to selecting an interaction for an agent. Bullseye won't work for this reason.

I initially thought Skills, but I'm not certain they factor in to the interaction selection calculation (they used to in PureConnect). Can anyone from Genesys explain how Genesys Cloud chooses between valid interactions for an agent? Is it purely done on time (with Priority potentially modifying that) or are Skill Proficiencies somehow incorporated?

Also, you didn't indicate what you want to happen to interactions that have been waiting a long time? Say, for example, There is an Interaction in Queue 1 that has been waiting 5 seconds, but one in Queue 2 that has been waiting for 2 minutes. Should Agent 1 still be given the Queue 1 interaction? Usually, in systems like this, you have to have the calculation factor in wait time to account for this.

The IVR inbound call flow will route calls to about 10 ACD queues, and other 10 queues are for transfer purpose (not from IVR). Total 150 agents across these 20 queues. Some agents are in two different queues.
All queues have equal priority. But individual agents have different priority requirement. Agent1 need anwser Queue1 calls before Queue2. Agent2 need answer Queue2 first. Agent3 has higher priority on Queue3 calls than Queue1, and so on. Can Genesys routing achieve this goal? How to config?

Hey Chris,

Thank you for the very detailed and informative response. A lot to take in there! 😲

Anyway, I did want to follow up as I see a contradiction between your response and David's (with the table.)

At issue here is the situation when you have more interactions waiting than agents to process them (so some waiting in queue.) When an agent becomes available, the system has to select an interaction to give to them.

Now assuming we have a Routing methodology that includes skills, only interactions that the agent has the necessary skills to service are considered, however how the system decides between those interactions is the issue here. According to David, only the interaction's effective time in queue is considered (actual time +/- priority.) That is what the graphic he posted says. According to you (I think) proficiencies come into play here as well.

Let's consider a situation where two interactions are waiting (let's ignore Priority for simplicity!). The first interaction has been waiting for 30 seconds and requires Skill A. The second for 2 minutes and requires Skill B. An agent becomes available, who has both skills. Skill A Proficiency 5 and Skill B Proficiency 1.

So, do they get the first interaction (the one they have the greater skill for) or the second (the one that has been waiting the longest)? If I'm understanding the original requirement correctly, we need it to be the first, however David's explanation suggests it would be the second.

In PureConnect, the exact formula was clear (somewhat convoluted, but certainly well defined and explained!) and, to a certain extent, tunable. This situation in Genesys Cloud is much less well defined and (if David's explanation is correct) less flexible. If David is, indeed, correct then I will be creating an Idea...

Sorry to beat a dead horse here, but I want to thoroughly understand what is going on!

Hi everyone - a lot going on this this post so I'm going to try to break down a few different comments/thoughts/options you have here.  Thanks to everyone for adding comments so far - all helping to get Lei pushed in the right direction.

• To start with, as David points out - we will use the agent's proficiency score to make a routing decision assuming the evaluation method set on the queue is set to 'Best Available Skills' (All Skills Matching and Ignore Skills do NOT use the proficiency score.)  Lei - it looks like you are set up correctly there so we can check that box.
• If all agents can work all queues, but you want to prioritize some work over others, this is really where skills and proficiencies should come into play.  If every agent can, at some point, answer every call, then its important that you give each agent all skills and then 'rank' them using proficiency values.  (Note:  this is the most effective way to work with our WFM tools for forecasting and scheduling).
• One you have all the agent's skills and rated appropriately - now you can start to ask questions around routing.  If your goal is to ensure that there are 'primary' agents for each queue, I'd recommend you create a skill for each queue and rate those agents who are always primary to 5 stars and those who should only be added when other queues are struggling to 0 or 1 stars (literally the lowest ranking).
• Lei, it sounds like you want to make sure the best agents for each queue/skill have the best shot at getting the calls for their primary queue and really only want others to help out when needed so, what you can do is use Bullseye routing...and here's how!  (Paul - this is a new one so let me know if this makes sense - happy to discuss further!)
• Step 1:  Define skill expression groups:  You can now 'group' agents together by defining a skill expression group…for example:  Skill A = 5 or Skill B = 5.  (This means find all agents will skill A OR skill B who have a proficiency equal to 5).  This will allow you to get all your TOP agents into a group and use them as a routing target.  (I'll elaborate on how to add them to queues in a moment). 
• From here – create additional groups where the skills stay the same, but the proficiency value goes down (e.g. Skill A = 4 or Skill B = 4).
• Step 2:  Add these groups as members of the queue.  In the Queue administration screen, click on the 'members' tab and then look for the 'Groups' tab.   On the right-hand side select the 'Skill Group' option in the dropdown menu and select all of the skill expression groups you want to route to.
• Step 3:  Go to the Routing tab and select bullseye routing and add the number of rings (up to 5) you want to define and add the duration of time to wait between each ring.  (NOTE:  The maximum amount of time you want to use really should be UNDER your patience threshold for that queue (meaning how long customers stay on hold before they abandon) for synchronous channels like Calls and Chats, but you can extend the timer quite a way out for digital media like emails as they can sit much longer but make sure whatever values you use allow you to meet your service goals.
• Step 4:  Go back to the members tab, click on groups again and assign the appropriate bullseye ring number to each group.   Assuming you put all the agents with proficiency of 5 in the first ring, and proficiency 4 in the next ring and so on, you will effectively give the best agents a shot at answering their preferred interactions ahead of any other agent.   NOTE:  This does assume you have enough availability to be able to address this – if the 5-star agents are not available – they may not answer the interaction but that is likely still preferable than the customer abandoning.

I will admit that this isn't a perfect scenario for what you describe, Lei – it does not guarantee, with 100% certainty that the primary agent will always answer interactions for which they are primary – but it should give you a much better chance to meet that need and still service customers promptly and/or meet your SLAs.

Here are a couple screenshots to outline what I'm talking about.

Skill Expression Groups – (screenshot)

Assign Skill Groups to a Queue:

Bullseye Routing Expansion via Groups:

A lot to digest - but hopefully this helps.  I'll be back with a follow-up post around transfers and skills b/c, unfortunately saving/stripping skills works differently with different media types so its not as clear as one might hope!

Feel free to post back with comments and questions.

Hi @Paul Simpson and @Lei Cao. 

Thank you. 

Please note, the comment regarding Stripping Skills on Blind Transfer was a question, not a suggestion :-) Lei posted "Also in transfer calls, agent skill proficiency seems not factor into agent selection, even we use Best available routing method." and I suggested checking the setting, not to change it. 

@Paul Simpson

@Tatjana Knezevic , unfortunately, I don't think that would help. For this to work, they would need to remove one set of skills and add a different one. That would require a Flow....

@Lei Cao 
Blind transfers: can you please check the setting on the Org level: Stripping Skills on Blind Transfer? 

Thanks for all the replies. In this example, There is an Interaction in Queue 1 that has been waiting 5 seconds, but one in Queue 2 that has been waiting for 2 minutes. Should Agent 1 still be given the Queue 1 interaction? Yes. Agent1 always handles Q1 if there is call in Q1.

We try to use skills, but in Blind or Consult Transfer other queues, skills will cause issue to route calls to another queue (different queue has different skill). Also in transfer calls, agent skill proficiency seems not factor into agent selection, even we use Best available routing method.

OK, so first up, Priorities are set on interactions, not Queues or Agents. What they actually do is to place an offset on the arrival time of the interaction that is used in the selection calculation (so it appears that the interaction has been waiting for a longer, or shorter, time than it actually has.)

Now, to the case in point. As far as I can see, Genesys Cloud provides plenty of scope for tuning which agent gets selected for an interaction, but much less when it comes to selecting an interaction for an agent. Bullseye won't work for this reason.

I initially thought Skills, but I'm not certain they factor in to the interaction selection calculation (they used to in PureConnect). Can anyone from Genesys explain how Genesys Cloud chooses between valid interactions for an agent? Is it purely done on time (with Priority potentially modifying that) or are Skill Proficiencies somehow incorporated?

Also, you didn't indicate what you want to happen to interactions that have been waiting a long time? Say, for example, There is an Interaction in Queue 1 that has been waiting 5 seconds, but one in Queue 2 that has been waiting for 2 minutes. Should Agent 1 still be given the Queue 1 interaction? Usually, in systems like this, you have to have the calculation factor in wait time to account for this.

The IVR inbound call flow will route calls to about 10 ACD queues, and other 10 queues are for transfer purpose (not from IVR). Total 150 agents across these 20 queues. Some agents are in two different queues.
All queues have equal priority. But individual agents have different priority requirement. Agent1 need anwser Queue1 calls before Queue2. Agent2 need answer Queue2 first. Agent3 has higher priority on Queue3 calls than Queue1, and so on. Can Genesys routing achieve this goal? How to config?

I believe Chris and I are both right :)

In conversation surplus, the selection of the interaction from those which are available to the agent will always be by conversation score.

What Chris is suggesting is controlling which interactions are available to an agent using Bullseye with group-based target expansion.  The group is defined by the proficiency level, so Chris is suggesting you could configure it so that an interaction is only made available to the lower proficiency group after say 3 minutes. So in your example, with 30 second and 2 minute wait times, only the first interaction would be available to the agent (because the one for which they're lower proficiency hasn't waited long enough to be put in their list) and so would be assigned.  If the second interaction had been waiting longer than the 3 minute threshold, then the second interaction would also be in their list, and then it would be assigned as it's longest waiting.

Hopefully that helps!

Hey Chris,

Thank you for the very detailed and informative response. A lot to take in there! 😲

Anyway, I did want to follow up as I see a contradiction between your response and David's (with the table.)

At issue here is the situation when you have more interactions waiting than agents to process them (so some waiting in queue.) When an agent becomes available, the system has to select an interaction to give to them.

Now assuming we have a Routing methodology that includes skills, only interactions that the agent has the necessary skills to service are considered, however how the system decides between those interactions is the issue here. According to David, only the interaction's effective time in queue is considered (actual time +/- priority.) That is what the graphic he posted says. According to you (I think) proficiencies come into play here as well.

Let's consider a situation where two interactions are waiting (let's ignore Priority for simplicity!). The first interaction has been waiting for 30 seconds and requires Skill A. The second for 2 minutes and requires Skill B. An agent becomes available, who has both skills. Skill A Proficiency 5 and Skill B Proficiency 1.

So, do they get the first interaction (the one they have the greater skill for) or the second (the one that has been waiting the longest)? If I'm understanding the original requirement correctly, we need it to be the first, however David's explanation suggests it would be the second.

In PureConnect, the exact formula was clear (somewhat convoluted, but certainly well defined and explained!) and, to a certain extent, tunable. This situation in Genesys Cloud is much less well defined and (if David's explanation is correct) less flexible. If David is, indeed, correct then I will be creating an Idea...

Sorry to beat a dead horse here, but I want to thoroughly understand what is going on!

Hi everyone - a lot going on this this post so I'm going to try to break down a few different comments/thoughts/options you have here.  Thanks to everyone for adding comments so far - all helping to get Lei pushed in the right direction.

• To start with, as David points out - we will use the agent's proficiency score to make a routing decision assuming the evaluation method set on the queue is set to 'Best Available Skills' (All Skills Matching and Ignore Skills do NOT use the proficiency score.)  Lei - it looks like you are set up correctly there so we can check that box.
• If all agents can work all queues, but you want to prioritize some work over others, this is really where skills and proficiencies should come into play.  If every agent can, at some point, answer every call, then its important that you give each agent all skills and then 'rank' them using proficiency values.  (Note:  this is the most effective way to work with our WFM tools for forecasting and scheduling).
• One you have all the agent's skills and rated appropriately - now you can start to ask questions around routing.  If your goal is to ensure that there are 'primary' agents for each queue, I'd recommend you create a skill for each queue and rate those agents who are always primary to 5 stars and those who should only be added when other queues are struggling to 0 or 1 stars (literally the lowest ranking).
• Lei, it sounds like you want to make sure the best agents for each queue/skill have the best shot at getting the calls for their primary queue and really only want others to help out when needed so, what you can do is use Bullseye routing...and here's how!  (Paul - this is a new one so let me know if this makes sense - happy to discuss further!)
• Step 1:  Define skill expression groups:  You can now 'group' agents together by defining a skill expression group…for example:  Skill A = 5 or Skill B = 5.  (This means find all agents will skill A OR skill B who have a proficiency equal to 5).  This will allow you to get all your TOP agents into a group and use them as a routing target.  (I'll elaborate on how to add them to queues in a moment). 
• From here – create additional groups where the skills stay the same, but the proficiency value goes down (e.g. Skill A = 4 or Skill B = 4).
• Step 2:  Add these groups as members of the queue.  In the Queue administration screen, click on the 'members' tab and then look for the 'Groups' tab.   On the right-hand side select the 'Skill Group' option in the dropdown menu and select all of the skill expression groups you want to route to.
• Step 3:  Go to the Routing tab and select bullseye routing and add the number of rings (up to 5) you want to define and add the duration of time to wait between each ring.  (NOTE:  The maximum amount of time you want to use really should be UNDER your patience threshold for that queue (meaning how long customers stay on hold before they abandon) for synchronous channels like Calls and Chats, but you can extend the timer quite a way out for digital media like emails as they can sit much longer but make sure whatever values you use allow you to meet your service goals.
• Step 4:  Go back to the members tab, click on groups again and assign the appropriate bullseye ring number to each group.   Assuming you put all the agents with proficiency of 5 in the first ring, and proficiency 4 in the next ring and so on, you will effectively give the best agents a shot at answering their preferred interactions ahead of any other agent.   NOTE:  This does assume you have enough availability to be able to address this – if the 5-star agents are not available – they may not answer the interaction but that is likely still preferable than the customer abandoning.

I will admit that this isn't a perfect scenario for what you describe, Lei – it does not guarantee, with 100% certainty that the primary agent will always answer interactions for which they are primary – but it should give you a much better chance to meet that need and still service customers promptly and/or meet your SLAs.

Here are a couple screenshots to outline what I'm talking about.

Skill Expression Groups – (screenshot)

Assign Skill Groups to a Queue:

Bullseye Routing Expansion via Groups:

A lot to digest - but hopefully this helps.  I'll be back with a follow-up post around transfers and skills b/c, unfortunately saving/stripping skills works differently with different media types so its not as clear as one might hope!

Feel free to post back with comments and questions.

Hi @Paul Simpson and @Lei Cao. 

Thank you. 

Please note, the comment regarding Stripping Skills on Blind Transfer was a question, not a suggestion :-) Lei posted "Also in transfer calls, agent skill proficiency seems not factor into agent selection, even we use Best available routing method." and I suggested checking the setting, not to change it. 

@Paul Simpson

@Tatjana Knezevic , unfortunately, I don't think that would help. For this to work, they would need to remove one set of skills and add a different one. That would require a Flow....

@Lei Cao 
Blind transfers: can you please check the setting on the Org level: Stripping Skills on Blind Transfer? 

Thanks for all the replies. In this example, There is an Interaction in Queue 1 that has been waiting 5 seconds, but one in Queue 2 that has been waiting for 2 minutes. Should Agent 1 still be given the Queue 1 interaction? Yes. Agent1 always handles Q1 if there is call in Q1.

We try to use skills, but in Blind or Consult Transfer other queues, skills will cause issue to route calls to another queue (different queue has different skill). Also in transfer calls, agent skill proficiency seems not factor into agent selection, even we use Best available routing method.

OK, so first up, Priorities are set on interactions, not Queues or Agents. What they actually do is to place an offset on the arrival time of the interaction that is used in the selection calculation (so it appears that the interaction has been waiting for a longer, or shorter, time than it actually has.)

Now, to the case in point. As far as I can see, Genesys Cloud provides plenty of scope for tuning which agent gets selected for an interaction, but much less when it comes to selecting an interaction for an agent. Bullseye won't work for this reason.

I initially thought Skills, but I'm not certain they factor in to the interaction selection calculation (they used to in PureConnect). Can anyone from Genesys explain how Genesys Cloud chooses between valid interactions for an agent? Is it purely done on time (with Priority potentially modifying that) or are Skill Proficiencies somehow incorporated?

Also, you didn't indicate what you want to happen to interactions that have been waiting a long time? Say, for example, There is an Interaction in Queue 1 that has been waiting 5 seconds, but one in Queue 2 that has been waiting for 2 minutes. Should Agent 1 still be given the Queue 1 interaction? Usually, in systems like this, you have to have the calculation factor in wait time to account for this.

The IVR inbound call flow will route calls to about 10 ACD queues, and other 10 queues are for transfer purpose (not from IVR). Total 150 agents across these 20 queues. Some agents are in two different queues.
All queues have equal priority. But individual agents have different priority requirement. Agent1 need anwser Queue1 calls before Queue2. Agent2 need answer Queue2 first. Agent3 has higher priority on Queue3 calls than Queue1, and so on. Can Genesys routing achieve this goal? How to config?

Ahhh, gotcha!

I can see this improving things, but also leading to complexity that would not exist if the Agent's proficiency with respect to the skills required by the interaction were evaluated directly. I have worked with scenarios where there might be multiple skills and a "best overall fit" is needed, without hard cut-offs like this provided.

I need to look at it all much more closely, and play around, but what happens if the agent has a lower proficiency, but has no other interactions? Do they immediately get the interaction, or will it have to wait for the bullseye to do it's thing? (So, in the case that sparked off this thread, If the agent is targeting Queue A and "backup" for Queue B, but there is nothing waiting in Queue A, would they have to wait before being offered something from Queue B? (I hope that makes sense!)

Sorry, not trying to be argumentative here, just trying to fully understand how this new functionality works.

I believe Chris and I are both right :)

In conversation surplus, the selection of the interaction from those which are available to the agent will always be by conversation score.

What Chris is suggesting is controlling which interactions are available to an agent using Bullseye with group-based target expansion.  The group is defined by the proficiency level, so Chris is suggesting you could configure it so that an interaction is only made available to the lower proficiency group after say 3 minutes. So in your example, with 30 second and 2 minute wait times, only the first interaction would be available to the agent (because the one for which they're lower proficiency hasn't waited long enough to be put in their list) and so would be assigned.  If the second interaction had been waiting longer than the 3 minute threshold, then the second interaction would also be in their list, and then it would be assigned as it's longest waiting.

Hopefully that helps!

Hey Chris,

Thank you for the very detailed and informative response. A lot to take in there! 😲

Anyway, I did want to follow up as I see a contradiction between your response and David's (with the table.)

At issue here is the situation when you have more interactions waiting than agents to process them (so some waiting in queue.) When an agent becomes available, the system has to select an interaction to give to them.

Now assuming we have a Routing methodology that includes skills, only interactions that the agent has the necessary skills to service are considered, however how the system decides between those interactions is the issue here. According to David, only the interaction's effective time in queue is considered (actual time +/- priority.) That is what the graphic he posted says. According to you (I think) proficiencies come into play here as well.

Let's consider a situation where two interactions are waiting (let's ignore Priority for simplicity!). The first interaction has been waiting for 30 seconds and requires Skill A. The second for 2 minutes and requires Skill B. An agent becomes available, who has both skills. Skill A Proficiency 5 and Skill B Proficiency 1.

So, do they get the first interaction (the one they have the greater skill for) or the second (the one that has been waiting the longest)? If I'm understanding the original requirement correctly, we need it to be the first, however David's explanation suggests it would be the second.

In PureConnect, the exact formula was clear (somewhat convoluted, but certainly well defined and explained!) and, to a certain extent, tunable. This situation in Genesys Cloud is much less well defined and (if David's explanation is correct) less flexible. If David is, indeed, correct then I will be creating an Idea...

Sorry to beat a dead horse here, but I want to thoroughly understand what is going on!

Hi everyone - a lot going on this this post so I'm going to try to break down a few different comments/thoughts/options you have here.  Thanks to everyone for adding comments so far - all helping to get Lei pushed in the right direction.

• To start with, as David points out - we will use the agent's proficiency score to make a routing decision assuming the evaluation method set on the queue is set to 'Best Available Skills' (All Skills Matching and Ignore Skills do NOT use the proficiency score.)  Lei - it looks like you are set up correctly there so we can check that box.
• If all agents can work all queues, but you want to prioritize some work over others, this is really where skills and proficiencies should come into play.  If every agent can, at some point, answer every call, then its important that you give each agent all skills and then 'rank' them using proficiency values.  (Note:  this is the most effective way to work with our WFM tools for forecasting and scheduling).
• One you have all the agent's skills and rated appropriately - now you can start to ask questions around routing.  If your goal is to ensure that there are 'primary' agents for each queue, I'd recommend you create a skill for each queue and rate those agents who are always primary to 5 stars and those who should only be added when other queues are struggling to 0 or 1 stars (literally the lowest ranking).
• Lei, it sounds like you want to make sure the best agents for each queue/skill have the best shot at getting the calls for their primary queue and really only want others to help out when needed so, what you can do is use Bullseye routing...and here's how!  (Paul - this is a new one so let me know if this makes sense - happy to discuss further!)
• Step 1:  Define skill expression groups:  You can now 'group' agents together by defining a skill expression group…for example:  Skill A = 5 or Skill B = 5.  (This means find all agents will skill A OR skill B who have a proficiency equal to 5).  This will allow you to get all your TOP agents into a group and use them as a routing target.  (I'll elaborate on how to add them to queues in a moment). 
• From here – create additional groups where the skills stay the same, but the proficiency value goes down (e.g. Skill A = 4 or Skill B = 4).
• Step 2:  Add these groups as members of the queue.  In the Queue administration screen, click on the 'members' tab and then look for the 'Groups' tab.   On the right-hand side select the 'Skill Group' option in the dropdown menu and select all of the skill expression groups you want to route to.
• Step 3:  Go to the Routing tab and select bullseye routing and add the number of rings (up to 5) you want to define and add the duration of time to wait between each ring.  (NOTE:  The maximum amount of time you want to use really should be UNDER your patience threshold for that queue (meaning how long customers stay on hold before they abandon) for synchronous channels like Calls and Chats, but you can extend the timer quite a way out for digital media like emails as they can sit much longer but make sure whatever values you use allow you to meet your service goals.
• Step 4:  Go back to the members tab, click on groups again and assign the appropriate bullseye ring number to each group.   Assuming you put all the agents with proficiency of 5 in the first ring, and proficiency 4 in the next ring and so on, you will effectively give the best agents a shot at answering their preferred interactions ahead of any other agent.   NOTE:  This does assume you have enough availability to be able to address this – if the 5-star agents are not available – they may not answer the interaction but that is likely still preferable than the customer abandoning.

I will admit that this isn't a perfect scenario for what you describe, Lei – it does not guarantee, with 100% certainty that the primary agent will always answer interactions for which they are primary – but it should give you a much better chance to meet that need and still service customers promptly and/or meet your SLAs.

Here are a couple screenshots to outline what I'm talking about.

Skill Expression Groups – (screenshot)

Assign Skill Groups to a Queue:

Bullseye Routing Expansion via Groups:

A lot to digest - but hopefully this helps.  I'll be back with a follow-up post around transfers and skills b/c, unfortunately saving/stripping skills works differently with different media types so its not as clear as one might hope!

Feel free to post back with comments and questions.

Hi @Paul Simpson and @Lei Cao. 

Thank you. 

Please note, the comment regarding Stripping Skills on Blind Transfer was a question, not a suggestion :-) Lei posted "Also in transfer calls, agent skill proficiency seems not factor into agent selection, even we use Best available routing method." and I suggested checking the setting, not to change it. 

@Paul Simpson

@Tatjana Knezevic , unfortunately, I don't think that would help. For this to work, they would need to remove one set of skills and add a different one. That would require a Flow....

@Lei Cao 
Blind transfers: can you please check the setting on the Org level: Stripping Skills on Blind Transfer? 

Thanks for all the replies. In this example, There is an Interaction in Queue 1 that has been waiting 5 seconds, but one in Queue 2 that has been waiting for 2 minutes. Should Agent 1 still be given the Queue 1 interaction? Yes. Agent1 always handles Q1 if there is call in Q1.

We try to use skills, but in Blind or Consult Transfer other queues, skills will cause issue to route calls to another queue (different queue has different skill). Also in transfer calls, agent skill proficiency seems not factor into agent selection, even we use Best available routing method.

OK, so first up, Priorities are set on interactions, not Queues or Agents. What they actually do is to place an offset on the arrival time of the interaction that is used in the selection calculation (so it appears that the interaction has been waiting for a longer, or shorter, time than it actually has.)

Now, to the case in point. As far as I can see, Genesys Cloud provides plenty of scope for tuning which agent gets selected for an interaction, but much less when it comes to selecting an interaction for an agent. Bullseye won't work for this reason.

I initially thought Skills, but I'm not certain they factor in to the interaction selection calculation (they used to in PureConnect). Can anyone from Genesys explain how Genesys Cloud chooses between valid interactions for an agent? Is it purely done on time (with Priority potentially modifying that) or are Skill Proficiencies somehow incorporated?

Also, you didn't indicate what you want to happen to interactions that have been waiting a long time? Say, for example, There is an Interaction in Queue 1 that has been waiting 5 seconds, but one in Queue 2 that has been waiting for 2 minutes. Should Agent 1 still be given the Queue 1 interaction? Usually, in systems like this, you have to have the calculation factor in wait time to account for this.

The IVR inbound call flow will route calls to about 10 ACD queues, and other 10 queues are for transfer purpose (not from IVR). Total 150 agents across these 20 queues. Some agents are in two different queues.
All queues have equal priority. But individual agents have different priority requirement. Agent1 need anwser Queue1 calls before Queue2. Agent2 need answer Queue2 first. Agent3 has higher priority on Queue3 calls than Queue1, and so on. Can Genesys routing achieve this goal? How to config?

Hi everyone - a lot going on this this post so I'm going to try to break down a few different comments/thoughts/options you have here.  Thanks to everyone for adding comments so far - all helping to get Lei pushed in the right direction.

• To start with, as David points out - we will use the agent's proficiency score to make a routing decision assuming the evaluation method set on the queue is set to 'Best Available Skills' (All Skills Matching and Ignore Skills do NOT use the proficiency score.)  Lei - it looks like you are set up correctly there so we can check that box.
• If all agents can work all queues, but you want to prioritize some work over others, this is really where skills and proficiencies should come into play.  If every agent can, at some point, answer every call, then its important that you give each agent all skills and then 'rank' them using proficiency values.  (Note:  this is the most effective way to work with our WFM tools for forecasting and scheduling).
• One you have all the agent's skills and rated appropriately - now you can start to ask questions around routing.  If your goal is to ensure that there are 'primary' agents for each queue, I'd recommend you create a skill for each queue and rate those agents who are always primary to 5 stars and those who should only be added when other queues are struggling to 0 or 1 stars (literally the lowest ranking).
• Lei, it sounds like you want to make sure the best agents for each queue/skill have the best shot at getting the calls for their primary queue and really only want others to help out when needed so, what you can do is use Bullseye routing...and here's how!  (Paul - this is a new one so let me know if this makes sense - happy to discuss further!)
• Step 1:  Define skill expression groups:  You can now 'group' agents together by defining a skill expression group…for example:  Skill A = 5 or Skill B = 5.  (This means find all agents will skill A OR skill B who have a proficiency equal to 5).  This will allow you to get all your TOP agents into a group and use them as a routing target.  (I'll elaborate on how to add them to queues in a moment). 
• From here – create additional groups where the skills stay the same, but the proficiency value goes down (e.g. Skill A = 4 or Skill B = 4).
• Step 2:  Add these groups as members of the queue.  In the Queue administration screen, click on the 'members' tab and then look for the 'Groups' tab.   On the right-hand side select the 'Skill Group' option in the dropdown menu and select all of the skill expression groups you want to route to.
• Step 3:  Go to the Routing tab and select bullseye routing and add the number of rings (up to 5) you want to define and add the duration of time to wait between each ring.  (NOTE:  The maximum amount of time you want to use really should be UNDER your patience threshold for that queue (meaning how long customers stay on hold before they abandon) for synchronous channels like Calls and Chats, but you can extend the timer quite a way out for digital media like emails as they can sit much longer but make sure whatever values you use allow you to meet your service goals.
• Step 4:  Go back to the members tab, click on groups again and assign the appropriate bullseye ring number to each group.   Assuming you put all the agents with proficiency of 5 in the first ring, and proficiency 4 in the next ring and so on, you will effectively give the best agents a shot at answering their preferred interactions ahead of any other agent.   NOTE:  This does assume you have enough availability to be able to address this – if the 5-star agents are not available – they may not answer the interaction but that is likely still preferable than the customer abandoning.

I will admit that this isn't a perfect scenario for what you describe, Lei – it does not guarantee, with 100% certainty that the primary agent will always answer interactions for which they are primary – but it should give you a much better chance to meet that need and still service customers promptly and/or meet your SLAs.

Here are a couple screenshots to outline what I'm talking about.

Skill Expression Groups – (screenshot)

Assign Skill Groups to a Queue:

Bullseye Routing Expansion via Groups:

A lot to digest - but hopefully this helps.  I'll be back with a follow-up post around transfers and skills b/c, unfortunately saving/stripping skills works differently with different media types so its not as clear as one might hope!

Feel free to post back with comments and questions.

Hi @Paul Simpson and @Lei Cao. 

Thank you. 

Please note, the comment regarding Stripping Skills on Blind Transfer was a question, not a suggestion :-) Lei posted "Also in transfer calls, agent skill proficiency seems not factor into agent selection, even we use Best available routing method." and I suggested checking the setting, not to change it. 

@Paul Simpson

@Tatjana Knezevic , unfortunately, I don't think that would help. For this to work, they would need to remove one set of skills and add a different one. That would require a Flow....

@Lei Cao 
Blind transfers: can you please check the setting on the Org level: Stripping Skills on Blind Transfer? 

Thanks for all the replies. In this example, There is an Interaction in Queue 1 that has been waiting 5 seconds, but one in Queue 2 that has been waiting for 2 minutes. Should Agent 1 still be given the Queue 1 interaction? Yes. Agent1 always handles Q1 if there is call in Q1.

We try to use skills, but in Blind or Consult Transfer other queues, skills will cause issue to route calls to another queue (different queue has different skill). Also in transfer calls, agent skill proficiency seems not factor into agent selection, even we use Best available routing method.

OK, so first up, Priorities are set on interactions, not Queues or Agents. What they actually do is to place an offset on the arrival time of the interaction that is used in the selection calculation (so it appears that the interaction has been waiting for a longer, or shorter, time than it actually has.)

Now, to the case in point. As far as I can see, Genesys Cloud provides plenty of scope for tuning which agent gets selected for an interaction, but much less when it comes to selecting an interaction for an agent. Bullseye won't work for this reason.

I initially thought Skills, but I'm not certain they factor in to the interaction selection calculation (they used to in PureConnect). Can anyone from Genesys explain how Genesys Cloud chooses between valid interactions for an agent? Is it purely done on time (with Priority potentially modifying that) or are Skill Proficiencies somehow incorporated?

Also, you didn't indicate what you want to happen to interactions that have been waiting a long time? Say, for example, There is an Interaction in Queue 1 that has been waiting 5 seconds, but one in Queue 2 that has been waiting for 2 minutes. Should Agent 1 still be given the Queue 1 interaction? Usually, in systems like this, you have to have the calculation factor in wait time to account for this.

The IVR inbound call flow will route calls to about 10 ACD queues, and other 10 queues are for transfer purpose (not from IVR). Total 150 agents across these 20 queues. Some agents are in two different queues.
All queues have equal priority. But individual agents have different priority requirement. Agent1 need anwser Queue1 calls before Queue2. Agent2 need answer Queue2 first. Agent3 has higher priority on Queue3 calls than Queue1, and so on. Can Genesys routing achieve this goal? How to config?