Genesys Cloud - Main

Hi Genesys Community,

Keen to hear from the community on your feedback or experience with Genesys CoPilot for Knowledge Surfacing especially around responsiveness and usability during live voice interactions. Has anyone experienced latency issues with CoPilot's real-time knowledge surfacing during voice interactions? It would be good to understand if there is any noticeable lag between the conversation and when knowledge articles are surfaced, or the way it searches to determine the right moment to trigger a search. This could make it challenging for agents to navigate the surfaced content while actively on a call. Thanks in advance :)

Hello Deepa,

That's a great question. I'm genuinely curious to see what the Community's experience has been with the responsiveness of the platform.

Hi Genesys Community,

Keen to hear from the community on your feedback or experience with Genesys CoPilot for Knowledge Surfacing especially around responsiveness and usability during live voice interactions. Has anyone experienced latency issues with CoPilot's real-time knowledge surfacing during voice interactions? It would be good to understand if there is any noticeable lag between the conversation and when knowledge articles are surfaced, or the way it searches to determine the right moment to trigger a search. This could make it challenging for agents to navigate the surfaced content while actively on a call. Thanks in advance :)

Thanks Jason - Do you know if many Genesys customers have implemented Genesys CoPilot Knowledge surfacing for the voice channel? 

------------------------------
Deepa Galaiya
Product Owner, Customer Interactions

Original Message:
Sent: 08-26-2025 11:52
From: Jason Kleitz
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Hello Deepa,

That's a great question. I'm genuinely curious to see what the Community's experience has been with the responsiveness of the platform.

------------------------------
Jason Kleitz
Online Community Manager/Moderator

Original Message:
Sent: 08-26-2025 04:13
From: Deepa Galaiya
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Hi Genesys Community,

Keen to hear from the community on your feedback or experience with Genesys CoPilot for Knowledge Surfacing especially around responsiveness and usability during live voice interactions. Has anyone experienced latency issues with CoPilot's real-time knowledge surfacing during voice interactions? It would be good to understand if there is any noticeable lag between the conversation and when knowledge articles are surfaced, or the way it searches to determine the right moment to trigger a search. This could make it challenging for agents to navigate the surfaced content while actively on a call. Thanks in advance :)

#ConversationalAI(Bots,AgentAssist,etc.)

------------------------------
Deepa Galaiya
Product Owner, Customer Interactions
------------------------------

Greetings @Deepa Galaiya - Our experience has been great thus far with the [Agent Copilot] knowledge surfacing during voice interactions with minimal to no lag time. The only latency we noticed was based off the fact that Copilot seems to wait to do its KB article search once the Caller pauses their speech or stops talking altogether. The surfacing has been so good that it often continues to refine/requery the knowledge results as the conversation continues, which means the agent may have more than one KB article result to consider over time.

That said, it's important to note that the answers to your questions depend heavily on:

• Network Performance | Any latency in the home/office connectivity impacts overall Genesys Cloud performance.
• Well-Written KB Article Content | If the articles are vague, broad, or have any overlap, the system will have a harder time locating the appropriate KB article. Leveraging KB Article Phrases (where necessary) has helped us dial in KB surfacing.
• Clear Communication | I know...this seems like common sense, but unlike with a digital channel, if Copilot cannot understand what the caller is saying (e.g. mumbled words, slang, no relevant keywords, etc), it either will not surface anything or will not surface the right thing.
• Picking a Side Copilot Should Monitor | In our case we programmed our overall Copilot knowledge surfacing rule to only listen to the Caller side so KB articles were only surfaced based off what the Caller said, and didn't query every time the agent said something as well. Prior to that change we noticed the system was pinging & ponging KB article results, especially if the search terms it chose to use varied based on who was speaking.

------------------------------
Brian T. Jones | Ascension | Senior Specialist - Technology

Original Message:
Sent: 08-27-2025 20:38
From: Deepa Galaiya
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Thanks Jason - Do you know if many Genesys customers have implemented Genesys CoPilot Knowledge surfacing for the voice channel? 

------------------------------
Deepa Galaiya
Product Owner, Customer Interactions

Original Message:
Sent: 08-26-2025 11:52
From: Jason Kleitz
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Hello Deepa,

That's a great question. I'm genuinely curious to see what the Community's experience has been with the responsiveness of the platform.

------------------------------
Jason Kleitz
Online Community Manager/Moderator

Original Message:
Sent: 08-26-2025 04:13
From: Deepa Galaiya
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Hi Genesys Community,

Keen to hear from the community on your feedback or experience with Genesys CoPilot for Knowledge Surfacing especially around responsiveness and usability during live voice interactions. Has anyone experienced latency issues with CoPilot's real-time knowledge surfacing during voice interactions? It would be good to understand if there is any noticeable lag between the conversation and when knowledge articles are surfaced, or the way it searches to determine the right moment to trigger a search. This could make it challenging for agents to navigate the surfaced content while actively on a call. Thanks in advance :)

#ConversationalAI(Bots,AgentAssist,etc.)

------------------------------
Deepa Galaiya
Product Owner, Customer Interactions
------------------------------

Hi Brian,

Thanks so much for sharing your experience and insights with Knowledge Surfacing. That is incredibly helpful and encouraging to hear! Glad to hear that KS has been responsive is continuing to learn! I hope you don't mind a couple of additional questions:

• What was your experience with the initial effort required to get started with trigger rules and phrases? 
• Is there a specific KB article format/structure that performs better?
• Have you found that there is ongoing manual effort required to finetune responses and phrases?
• Are there any defects you have noticed with this feature? 

Thanks again for taking time to respond. Really appreciate it :)

Deepa

Greetings @Deepa Galaiya - Our experience has been great thus far with the [Agent Copilot] knowledge surfacing during voice interactions with minimal to no lag time. The only latency we noticed was based off the fact that Copilot seems to wait to do its KB article search once the Caller pauses their speech or stops talking altogether. The surfacing has been so good that it often continues to refine/requery the knowledge results as the conversation continues, which means the agent may have more than one KB article result to consider over time.

That said, it's important to note that the answers to your questions depend heavily on:

• Network Performance | Any latency in the home/office connectivity impacts overall Genesys Cloud performance.
• Well-Written KB Article Content | If the articles are vague, broad, or have any overlap, the system will have a harder time locating the appropriate KB article. Leveraging KB Article Phrases (where necessary) has helped us dial in KB surfacing.
• Clear Communication | I know...this seems like common sense, but unlike with a digital channel, if Copilot cannot understand what the caller is saying (e.g. mumbled words, slang, no relevant keywords, etc), it either will not surface anything or will not surface the right thing.
• Picking a Side Copilot Should Monitor | In our case we programmed our overall Copilot knowledge surfacing rule to only listen to the Caller side so KB articles were only surfaced based off what the Caller said, and didn't query every time the agent said something as well. Prior to that change we noticed the system was pinging & ponging KB article results, especially if the search terms it chose to use varied based on who was speaking.

------------------------------
Brian T. Jones | Ascension | Senior Specialist - Technology

Original Message:
Sent: 08-27-2025 20:38
From: Deepa Galaiya
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Thanks Jason - Do you know if many Genesys customers have implemented Genesys CoPilot Knowledge surfacing for the voice channel? 

------------------------------
Deepa Galaiya
Product Owner, Customer Interactions

Original Message:
Sent: 08-26-2025 11:52
From: Jason Kleitz
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Hello Deepa,

That's a great question. I'm genuinely curious to see what the Community's experience has been with the responsiveness of the platform.

------------------------------
Jason Kleitz
Online Community Manager/Moderator

Original Message:
Sent: 08-26-2025 04:13
From: Deepa Galaiya
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Hi Genesys Community,

Keen to hear from the community on your feedback or experience with Genesys CoPilot for Knowledge Surfacing especially around responsiveness and usability during live voice interactions. Has anyone experienced latency issues with CoPilot's real-time knowledge surfacing during voice interactions? It would be good to understand if there is any noticeable lag between the conversation and when knowledge articles are surfaced, or the way it searches to determine the right moment to trigger a search. This could make it challenging for agents to navigate the surfaced content while actively on a call. Thanks in advance :)

#ConversationalAI(Bots,AgentAssist,etc.)

------------------------------
Deepa Galaiya
Product Owner, Customer Interactions
------------------------------

No problem, Deepa:

• Q: What was your experience with the initial effort required to get started with trigger rules and phrases? 
• A: For us it was minimal given we found sufficient success with the default "No other rule applies > search knowledge base" and adding a "Greeting" intent that automatically surfaced our "Opening Greeting" Canned response for the agent. Here's an example:

• Q: Is there a specific KB article format/structure that performs better?
• A: Great question that I'm sure other community members have various opinions on. We have tried both PORTRAIT and LANDSCAPE layouts. As clean as the table format is, it can be challenging to copy/paste/format those tables from our outside source documents, answer-highlighting doesn't work as well (if at all with tables), and the new multi-panel columnar format for the Agent Workspace scrunches up the table to the point where it's hard for our agents to scroll & read both horizontally and vertically. As such, we stuck with the vertical format given it plays better with the multi-panel config, is easier to format, & answer highlighting works fairly consistently.

• Q: Have you found that there is ongoing manual effort required to finetune responses and phrases?
• A: Anyone that says KB/responses/phrase/copilot management is not ongoing is lying to you, haha. That said, I will say if the KBs are well-written and tuned upfront, any ongoing effort should be minimal & tolerable, if not welcomed.

• Q: Are there any defects you have noticed with this feature? 
• A: The biggest issues for us are the KB formatting as well as the display challenges within the new multi-panel config. Also as mentioned above, if you do have tables in your KBs, the answer highlighting doesn't work as well, if at all, and you cannot see the article content in the surfaced article (only the title). Although not a "defect", if you do choose to use Copilot's Answer Generation feature, have your staff give feedback on its accuracy. We turned Answer Generation off given it proved too risky for our lines of business to rely on [at this time] versus reviewing the highlighted answer or full KB. That's truly a "results & adoption varies by customer" feature.

Hi Brian,

Thanks so much for sharing your experience and insights with Knowledge Surfacing. That is incredibly helpful and encouraging to hear! Glad to hear that KS has been responsive is continuing to learn! I hope you don't mind a couple of additional questions:

• What was your experience with the initial effort required to get started with trigger rules and phrases? 
• Is there a specific KB article format/structure that performs better?
• Have you found that there is ongoing manual effort required to finetune responses and phrases?
• Are there any defects you have noticed with this feature? 

Thanks again for taking time to respond. Really appreciate it :)

Deepa

Greetings @Deepa Galaiya - Our experience has been great thus far with the [Agent Copilot] knowledge surfacing during voice interactions with minimal to no lag time. The only latency we noticed was based off the fact that Copilot seems to wait to do its KB article search once the Caller pauses their speech or stops talking altogether. The surfacing has been so good that it often continues to refine/requery the knowledge results as the conversation continues, which means the agent may have more than one KB article result to consider over time.

That said, it's important to note that the answers to your questions depend heavily on:

• Network Performance | Any latency in the home/office connectivity impacts overall Genesys Cloud performance.
• Well-Written KB Article Content | If the articles are vague, broad, or have any overlap, the system will have a harder time locating the appropriate KB article. Leveraging KB Article Phrases (where necessary) has helped us dial in KB surfacing.
• Clear Communication | I know...this seems like common sense, but unlike with a digital channel, if Copilot cannot understand what the caller is saying (e.g. mumbled words, slang, no relevant keywords, etc), it either will not surface anything or will not surface the right thing.
• Picking a Side Copilot Should Monitor | In our case we programmed our overall Copilot knowledge surfacing rule to only listen to the Caller side so KB articles were only surfaced based off what the Caller said, and didn't query every time the agent said something as well. Prior to that change we noticed the system was pinging & ponging KB article results, especially if the search terms it chose to use varied based on who was speaking.

------------------------------
Brian T. Jones | Ascension | Senior Specialist - Technology

Original Message:
Sent: 08-27-2025 20:38
From: Deepa Galaiya
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Thanks Jason - Do you know if many Genesys customers have implemented Genesys CoPilot Knowledge surfacing for the voice channel? 

------------------------------
Deepa Galaiya
Product Owner, Customer Interactions

Original Message:
Sent: 08-26-2025 11:52
From: Jason Kleitz
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Hello Deepa,

That's a great question. I'm genuinely curious to see what the Community's experience has been with the responsiveness of the platform.

------------------------------
Jason Kleitz
Online Community Manager/Moderator

Original Message:
Sent: 08-26-2025 04:13
From: Deepa Galaiya
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Hi Genesys Community,

Keen to hear from the community on your feedback or experience with Genesys CoPilot for Knowledge Surfacing especially around responsiveness and usability during live voice interactions. Has anyone experienced latency issues with CoPilot's real-time knowledge surfacing during voice interactions? It would be good to understand if there is any noticeable lag between the conversation and when knowledge articles are surfaced, or the way it searches to determine the right moment to trigger a search. This could make it challenging for agents to navigate the surfaced content while actively on a call. Thanks in advance :)

#ConversationalAI(Bots,AgentAssist,etc.)

------------------------------
Deepa Galaiya
Product Owner, Customer Interactions
------------------------------

Hi Brian, Apologies for the delayed response. Thanks for sharing your knowledge and experience!

The point about articles that are in table formats is really interesting. We use Live Pro as our Knowledge Management system, and all of our articles are in tabular format. I wonder if this could be a potential showstopper...

Does KS display an AI generated summary of the knowledge article in the Genesys UI?

Thanks again and have a great day!

Deepa

No problem, Deepa:

• Q: What was your experience with the initial effort required to get started with trigger rules and phrases? 
• A: For us it was minimal given we found sufficient success with the default "No other rule applies > search knowledge base" and adding a "Greeting" intent that automatically surfaced our "Opening Greeting" Canned response for the agent. Here's an example:

• Q: Is there a specific KB article format/structure that performs better?
• A: Great question that I'm sure other community members have various opinions on. We have tried both PORTRAIT and LANDSCAPE layouts. As clean as the table format is, it can be challenging to copy/paste/format those tables from our outside source documents, answer-highlighting doesn't work as well (if at all with tables), and the new multi-panel columnar format for the Agent Workspace scrunches up the table to the point where it's hard for our agents to scroll & read both horizontally and vertically. As such, we stuck with the vertical format given it plays better with the multi-panel config, is easier to format, & answer highlighting works fairly consistently.

• Q: Have you found that there is ongoing manual effort required to finetune responses and phrases?
• A: Anyone that says KB/responses/phrase/copilot management is not ongoing is lying to you, haha. That said, I will say if the KBs are well-written and tuned upfront, any ongoing effort should be minimal & tolerable, if not welcomed.

• Q: Are there any defects you have noticed with this feature? 
• A: The biggest issues for us are the KB formatting as well as the display challenges within the new multi-panel config. Also as mentioned above, if you do have tables in your KBs, the answer highlighting doesn't work as well, if at all, and you cannot see the article content in the surfaced article (only the title). Although not a "defect", if you do choose to use Copilot's Answer Generation feature, have your staff give feedback on its accuracy. We turned Answer Generation off given it proved too risky for our lines of business to rely on [at this time] versus reviewing the highlighted answer or full KB. That's truly a "results & adoption varies by customer" feature.

Hi Brian,

Thanks so much for sharing your experience and insights with Knowledge Surfacing. That is incredibly helpful and encouraging to hear! Glad to hear that KS has been responsive is continuing to learn! I hope you don't mind a couple of additional questions:

• What was your experience with the initial effort required to get started with trigger rules and phrases? 
• Is there a specific KB article format/structure that performs better?
• Have you found that there is ongoing manual effort required to finetune responses and phrases?
• Are there any defects you have noticed with this feature? 

Thanks again for taking time to respond. Really appreciate it :)

Deepa

Greetings @Deepa Galaiya - Our experience has been great thus far with the [Agent Copilot] knowledge surfacing during voice interactions with minimal to no lag time. The only latency we noticed was based off the fact that Copilot seems to wait to do its KB article search once the Caller pauses their speech or stops talking altogether. The surfacing has been so good that it often continues to refine/requery the knowledge results as the conversation continues, which means the agent may have more than one KB article result to consider over time.

That said, it's important to note that the answers to your questions depend heavily on:

• Network Performance | Any latency in the home/office connectivity impacts overall Genesys Cloud performance.
• Well-Written KB Article Content | If the articles are vague, broad, or have any overlap, the system will have a harder time locating the appropriate KB article. Leveraging KB Article Phrases (where necessary) has helped us dial in KB surfacing.
• Clear Communication | I know...this seems like common sense, but unlike with a digital channel, if Copilot cannot understand what the caller is saying (e.g. mumbled words, slang, no relevant keywords, etc), it either will not surface anything or will not surface the right thing.
• Picking a Side Copilot Should Monitor | In our case we programmed our overall Copilot knowledge surfacing rule to only listen to the Caller side so KB articles were only surfaced based off what the Caller said, and didn't query every time the agent said something as well. Prior to that change we noticed the system was pinging & ponging KB article results, especially if the search terms it chose to use varied based on who was speaking.

------------------------------
Brian T. Jones | Ascension | Senior Specialist - Technology

Original Message:
Sent: 08-27-2025 20:38
From: Deepa Galaiya
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Thanks Jason - Do you know if many Genesys customers have implemented Genesys CoPilot Knowledge surfacing for the voice channel? 

------------------------------
Deepa Galaiya
Product Owner, Customer Interactions

Original Message:
Sent: 08-26-2025 11:52
From: Jason Kleitz
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Hello Deepa,

That's a great question. I'm genuinely curious to see what the Community's experience has been with the responsiveness of the platform.

------------------------------
Jason Kleitz
Online Community Manager/Moderator

Original Message:
Sent: 08-26-2025 04:13
From: Deepa Galaiya
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Hi Genesys Community,

Keen to hear from the community on your feedback or experience with Genesys CoPilot for Knowledge Surfacing especially around responsiveness and usability during live voice interactions. Has anyone experienced latency issues with CoPilot's real-time knowledge surfacing during voice interactions? It would be good to understand if there is any noticeable lag between the conversation and when knowledge articles are surfaced, or the way it searches to determine the right moment to trigger a search. This could make it challenging for agents to navigate the surfaced content while actively on a call. Thanks in advance :)

#ConversationalAI(Bots,AgentAssist,etc.)

------------------------------
Deepa Galaiya
Product Owner, Customer Interactions
------------------------------

Yes, if you choose to use Copilot's answer generation feature, AI will generate a summary based on the KB it would've otherwise surfaced. You'll want to test this with the table heavy KBs though to see if it works for you.

Hi Brian, Apologies for the delayed response. Thanks for sharing your knowledge and experience!

The point about articles that are in table formats is really interesting. We use Live Pro as our Knowledge Management system, and all of our articles are in tabular format. I wonder if this could be a potential showstopper...

Does KS display an AI generated summary of the knowledge article in the Genesys UI?

Thanks again and have a great day!

Deepa

No problem, Deepa:

• Q: What was your experience with the initial effort required to get started with trigger rules and phrases? 
• A: For us it was minimal given we found sufficient success with the default "No other rule applies > search knowledge base" and adding a "Greeting" intent that automatically surfaced our "Opening Greeting" Canned response for the agent. Here's an example:

• Q: Is there a specific KB article format/structure that performs better?
• A: Great question that I'm sure other community members have various opinions on. We have tried both PORTRAIT and LANDSCAPE layouts. As clean as the table format is, it can be challenging to copy/paste/format those tables from our outside source documents, answer-highlighting doesn't work as well (if at all with tables), and the new multi-panel columnar format for the Agent Workspace scrunches up the table to the point where it's hard for our agents to scroll & read both horizontally and vertically. As such, we stuck with the vertical format given it plays better with the multi-panel config, is easier to format, & answer highlighting works fairly consistently.

• Q: Have you found that there is ongoing manual effort required to finetune responses and phrases?
• A: Anyone that says KB/responses/phrase/copilot management is not ongoing is lying to you, haha. That said, I will say if the KBs are well-written and tuned upfront, any ongoing effort should be minimal & tolerable, if not welcomed.

• Q: Are there any defects you have noticed with this feature? 
• A: The biggest issues for us are the KB formatting as well as the display challenges within the new multi-panel config. Also as mentioned above, if you do have tables in your KBs, the answer highlighting doesn't work as well, if at all, and you cannot see the article content in the surfaced article (only the title). Although not a "defect", if you do choose to use Copilot's Answer Generation feature, have your staff give feedback on its accuracy. We turned Answer Generation off given it proved too risky for our lines of business to rely on [at this time] versus reviewing the highlighted answer or full KB. That's truly a "results & adoption varies by customer" feature.

Hi Brian,

Thanks so much for sharing your experience and insights with Knowledge Surfacing. That is incredibly helpful and encouraging to hear! Glad to hear that KS has been responsive is continuing to learn! I hope you don't mind a couple of additional questions:

• What was your experience with the initial effort required to get started with trigger rules and phrases? 
• Is there a specific KB article format/structure that performs better?
• Have you found that there is ongoing manual effort required to finetune responses and phrases?
• Are there any defects you have noticed with this feature? 

Thanks again for taking time to respond. Really appreciate it :)

Deepa

Greetings @Deepa Galaiya - Our experience has been great thus far with the [Agent Copilot] knowledge surfacing during voice interactions with minimal to no lag time. The only latency we noticed was based off the fact that Copilot seems to wait to do its KB article search once the Caller pauses their speech or stops talking altogether. The surfacing has been so good that it often continues to refine/requery the knowledge results as the conversation continues, which means the agent may have more than one KB article result to consider over time.

That said, it's important to note that the answers to your questions depend heavily on:

• Network Performance | Any latency in the home/office connectivity impacts overall Genesys Cloud performance.
• Well-Written KB Article Content | If the articles are vague, broad, or have any overlap, the system will have a harder time locating the appropriate KB article. Leveraging KB Article Phrases (where necessary) has helped us dial in KB surfacing.
• Clear Communication | I know...this seems like common sense, but unlike with a digital channel, if Copilot cannot understand what the caller is saying (e.g. mumbled words, slang, no relevant keywords, etc), it either will not surface anything or will not surface the right thing.
• Picking a Side Copilot Should Monitor | In our case we programmed our overall Copilot knowledge surfacing rule to only listen to the Caller side so KB articles were only surfaced based off what the Caller said, and didn't query every time the agent said something as well. Prior to that change we noticed the system was pinging & ponging KB article results, especially if the search terms it chose to use varied based on who was speaking.

------------------------------
Brian T. Jones | Ascension | Senior Specialist - Technology

Original Message:
Sent: 08-27-2025 20:38
From: Deepa Galaiya
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Thanks Jason - Do you know if many Genesys customers have implemented Genesys CoPilot Knowledge surfacing for the voice channel? 

------------------------------
Deepa Galaiya
Product Owner, Customer Interactions

Original Message:
Sent: 08-26-2025 11:52
From: Jason Kleitz
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Hello Deepa,

That's a great question. I'm genuinely curious to see what the Community's experience has been with the responsiveness of the platform.

------------------------------
Jason Kleitz
Online Community Manager/Moderator

Original Message:
Sent: 08-26-2025 04:13
From: Deepa Galaiya
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Hi Genesys Community,

Keen to hear from the community on your feedback or experience with Genesys CoPilot for Knowledge Surfacing especially around responsiveness and usability during live voice interactions. Has anyone experienced latency issues with CoPilot's real-time knowledge surfacing during voice interactions? It would be good to understand if there is any noticeable lag between the conversation and when knowledge articles are surfaced, or the way it searches to determine the right moment to trigger a search. This could make it challenging for agents to navigate the surfaced content while actively on a call. Thanks in advance :)

#ConversationalAI(Bots,AgentAssist,etc.)

------------------------------
Deepa Galaiya
Product Owner, Customer Interactions
------------------------------

Hi Brian, Apologies for the delayed response. Thanks for sharing your knowledge and experience!

The point about articles that are in table formats is really interesting. We use Live Pro as our Knowledge Management system, and all of our articles are in tabular format. I wonder if this could be a potential showstopper...

Does KS display an AI generated summary of the knowledge article in the Genesys UI?

Thanks again and have a great day!

Deepa

No problem, Deepa:

• Q: What was your experience with the initial effort required to get started with trigger rules and phrases? 
• A: For us it was minimal given we found sufficient success with the default "No other rule applies > search knowledge base" and adding a "Greeting" intent that automatically surfaced our "Opening Greeting" Canned response for the agent. Here's an example:

• Q: Is there a specific KB article format/structure that performs better?
• A: Great question that I'm sure other community members have various opinions on. We have tried both PORTRAIT and LANDSCAPE layouts. As clean as the table format is, it can be challenging to copy/paste/format those tables from our outside source documents, answer-highlighting doesn't work as well (if at all with tables), and the new multi-panel columnar format for the Agent Workspace scrunches up the table to the point where it's hard for our agents to scroll & read both horizontally and vertically. As such, we stuck with the vertical format given it plays better with the multi-panel config, is easier to format, & answer highlighting works fairly consistently.

• Q: Have you found that there is ongoing manual effort required to finetune responses and phrases?
• A: Anyone that says KB/responses/phrase/copilot management is not ongoing is lying to you, haha. That said, I will say if the KBs are well-written and tuned upfront, any ongoing effort should be minimal & tolerable, if not welcomed.

• Q: Are there any defects you have noticed with this feature? 
• A: The biggest issues for us are the KB formatting as well as the display challenges within the new multi-panel config. Also as mentioned above, if you do have tables in your KBs, the answer highlighting doesn't work as well, if at all, and you cannot see the article content in the surfaced article (only the title). Although not a "defect", if you do choose to use Copilot's Answer Generation feature, have your staff give feedback on its accuracy. We turned Answer Generation off given it proved too risky for our lines of business to rely on [at this time] versus reviewing the highlighted answer or full KB. That's truly a "results & adoption varies by customer" feature.

Hi Brian,

Thanks so much for sharing your experience and insights with Knowledge Surfacing. That is incredibly helpful and encouraging to hear! Glad to hear that KS has been responsive is continuing to learn! I hope you don't mind a couple of additional questions:

• What was your experience with the initial effort required to get started with trigger rules and phrases? 
• Is there a specific KB article format/structure that performs better?
• Have you found that there is ongoing manual effort required to finetune responses and phrases?
• Are there any defects you have noticed with this feature? 

Thanks again for taking time to respond. Really appreciate it :)

Deepa

Greetings @Deepa Galaiya - Our experience has been great thus far with the [Agent Copilot] knowledge surfacing during voice interactions with minimal to no lag time. The only latency we noticed was based off the fact that Copilot seems to wait to do its KB article search once the Caller pauses their speech or stops talking altogether. The surfacing has been so good that it often continues to refine/requery the knowledge results as the conversation continues, which means the agent may have more than one KB article result to consider over time.

That said, it's important to note that the answers to your questions depend heavily on:

• Network Performance | Any latency in the home/office connectivity impacts overall Genesys Cloud performance.
• Well-Written KB Article Content | If the articles are vague, broad, or have any overlap, the system will have a harder time locating the appropriate KB article. Leveraging KB Article Phrases (where necessary) has helped us dial in KB surfacing.
• Clear Communication | I know...this seems like common sense, but unlike with a digital channel, if Copilot cannot understand what the caller is saying (e.g. mumbled words, slang, no relevant keywords, etc), it either will not surface anything or will not surface the right thing.
• Picking a Side Copilot Should Monitor | In our case we programmed our overall Copilot knowledge surfacing rule to only listen to the Caller side so KB articles were only surfaced based off what the Caller said, and didn't query every time the agent said something as well. Prior to that change we noticed the system was pinging & ponging KB article results, especially if the search terms it chose to use varied based on who was speaking.

------------------------------
Brian T. Jones | Ascension | Senior Specialist - Technology

Original Message:
Sent: 08-27-2025 20:38
From: Deepa Galaiya
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Thanks Jason - Do you know if many Genesys customers have implemented Genesys CoPilot Knowledge surfacing for the voice channel? 

------------------------------
Deepa Galaiya
Product Owner, Customer Interactions

Original Message:
Sent: 08-26-2025 11:52
From: Jason Kleitz
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Hello Deepa,

That's a great question. I'm genuinely curious to see what the Community's experience has been with the responsiveness of the platform.

------------------------------
Jason Kleitz
Online Community Manager/Moderator

Original Message:
Sent: 08-26-2025 04:13
From: Deepa Galaiya
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Hi Genesys Community,

Keen to hear from the community on your feedback or experience with Genesys CoPilot for Knowledge Surfacing especially around responsiveness and usability during live voice interactions. Has anyone experienced latency issues with CoPilot's real-time knowledge surfacing during voice interactions? It would be good to understand if there is any noticeable lag between the conversation and when knowledge articles are surfaced, or the way it searches to determine the right moment to trigger a search. This could make it challenging for agents to navigate the surfaced content while actively on a call. Thanks in advance :)

#ConversationalAI(Bots,AgentAssist,etc.)

------------------------------
Deepa Galaiya
Product Owner, Customer Interactions
------------------------------

How are you getting your articles into a Genesys Cloud Knowledge Base for use with Agent Copilot and Bots?

Hi Brian, Apologies for the delayed response. Thanks for sharing your knowledge and experience!

The point about articles that are in table formats is really interesting. We use Live Pro as our Knowledge Management system, and all of our articles are in tabular format. I wonder if this could be a potential showstopper...

Does KS display an AI generated summary of the knowledge article in the Genesys UI?

Thanks again and have a great day!

Deepa

No problem, Deepa:

• Q: What was your experience with the initial effort required to get started with trigger rules and phrases? 
• A: For us it was minimal given we found sufficient success with the default "No other rule applies > search knowledge base" and adding a "Greeting" intent that automatically surfaced our "Opening Greeting" Canned response for the agent. Here's an example:

• Q: Is there a specific KB article format/structure that performs better?
• A: Great question that I'm sure other community members have various opinions on. We have tried both PORTRAIT and LANDSCAPE layouts. As clean as the table format is, it can be challenging to copy/paste/format those tables from our outside source documents, answer-highlighting doesn't work as well (if at all with tables), and the new multi-panel columnar format for the Agent Workspace scrunches up the table to the point where it's hard for our agents to scroll & read both horizontally and vertically. As such, we stuck with the vertical format given it plays better with the multi-panel config, is easier to format, & answer highlighting works fairly consistently.

• Q: Have you found that there is ongoing manual effort required to finetune responses and phrases?
• A: Anyone that says KB/responses/phrase/copilot management is not ongoing is lying to you, haha. That said, I will say if the KBs are well-written and tuned upfront, any ongoing effort should be minimal & tolerable, if not welcomed.

• Q: Are there any defects you have noticed with this feature? 
• A: The biggest issues for us are the KB formatting as well as the display challenges within the new multi-panel config. Also as mentioned above, if you do have tables in your KBs, the answer highlighting doesn't work as well, if at all, and you cannot see the article content in the surfaced article (only the title). Although not a "defect", if you do choose to use Copilot's Answer Generation feature, have your staff give feedback on its accuracy. We turned Answer Generation off given it proved too risky for our lines of business to rely on [at this time] versus reviewing the highlighted answer or full KB. That's truly a "results & adoption varies by customer" feature.

Hi Brian,

Thanks so much for sharing your experience and insights with Knowledge Surfacing. That is incredibly helpful and encouraging to hear! Glad to hear that KS has been responsive is continuing to learn! I hope you don't mind a couple of additional questions:

• What was your experience with the initial effort required to get started with trigger rules and phrases? 
• Is there a specific KB article format/structure that performs better?
• Have you found that there is ongoing manual effort required to finetune responses and phrases?
• Are there any defects you have noticed with this feature? 

Thanks again for taking time to respond. Really appreciate it :)

Deepa

Greetings @Deepa Galaiya - Our experience has been great thus far with the [Agent Copilot] knowledge surfacing during voice interactions with minimal to no lag time. The only latency we noticed was based off the fact that Copilot seems to wait to do its KB article search once the Caller pauses their speech or stops talking altogether. The surfacing has been so good that it often continues to refine/requery the knowledge results as the conversation continues, which means the agent may have more than one KB article result to consider over time.

That said, it's important to note that the answers to your questions depend heavily on:

• Network Performance | Any latency in the home/office connectivity impacts overall Genesys Cloud performance.
• Well-Written KB Article Content | If the articles are vague, broad, or have any overlap, the system will have a harder time locating the appropriate KB article. Leveraging KB Article Phrases (where necessary) has helped us dial in KB surfacing.
• Clear Communication | I know...this seems like common sense, but unlike with a digital channel, if Copilot cannot understand what the caller is saying (e.g. mumbled words, slang, no relevant keywords, etc), it either will not surface anything or will not surface the right thing.
• Picking a Side Copilot Should Monitor | In our case we programmed our overall Copilot knowledge surfacing rule to only listen to the Caller side so KB articles were only surfaced based off what the Caller said, and didn't query every time the agent said something as well. Prior to that change we noticed the system was pinging & ponging KB article results, especially if the search terms it chose to use varied based on who was speaking.

------------------------------
Brian T. Jones | Ascension | Senior Specialist - Technology

Original Message:
Sent: 08-27-2025 20:38
From: Deepa Galaiya
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Thanks Jason - Do you know if many Genesys customers have implemented Genesys CoPilot Knowledge surfacing for the voice channel? 

------------------------------
Deepa Galaiya
Product Owner, Customer Interactions

Original Message:
Sent: 08-26-2025 11:52
From: Jason Kleitz
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Hello Deepa,

That's a great question. I'm genuinely curious to see what the Community's experience has been with the responsiveness of the platform.

------------------------------
Jason Kleitz
Online Community Manager/Moderator

Original Message:
Sent: 08-26-2025 04:13
From: Deepa Galaiya
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Hi Genesys Community,

Keen to hear from the community on your feedback or experience with Genesys CoPilot for Knowledge Surfacing especially around responsiveness and usability during live voice interactions. Has anyone experienced latency issues with CoPilot's real-time knowledge surfacing during voice interactions? It would be good to understand if there is any noticeable lag between the conversation and when knowledge articles are surfaced, or the way it searches to determine the right moment to trigger a search. This could make it challenging for agents to navigate the surfaced content while actively on a call. Thanks in advance :)

#ConversationalAI(Bots,AgentAssist,etc.)

------------------------------
Deepa Galaiya
Product Owner, Customer Interactions
------------------------------

Hi George,

We have to get an API connected so that Genesys can ingest all of our KB articles from Livepro (our KM sysyem). 

Cheers,

Deepa

How are you getting your articles into a Genesys Cloud Knowledge Base for use with Agent Copilot and Bots?

Hi Brian, Apologies for the delayed response. Thanks for sharing your knowledge and experience!

The point about articles that are in table formats is really interesting. We use Live Pro as our Knowledge Management system, and all of our articles are in tabular format. I wonder if this could be a potential showstopper...

Does KS display an AI generated summary of the knowledge article in the Genesys UI?

Thanks again and have a great day!

Deepa

No problem, Deepa:

• Q: What was your experience with the initial effort required to get started with trigger rules and phrases? 
• A: For us it was minimal given we found sufficient success with the default "No other rule applies > search knowledge base" and adding a "Greeting" intent that automatically surfaced our "Opening Greeting" Canned response for the agent. Here's an example:

• Q: Is there a specific KB article format/structure that performs better?
• A: Great question that I'm sure other community members have various opinions on. We have tried both PORTRAIT and LANDSCAPE layouts. As clean as the table format is, it can be challenging to copy/paste/format those tables from our outside source documents, answer-highlighting doesn't work as well (if at all with tables), and the new multi-panel columnar format for the Agent Workspace scrunches up the table to the point where it's hard for our agents to scroll & read both horizontally and vertically. As such, we stuck with the vertical format given it plays better with the multi-panel config, is easier to format, & answer highlighting works fairly consistently.

• Q: Have you found that there is ongoing manual effort required to finetune responses and phrases?
• A: Anyone that says KB/responses/phrase/copilot management is not ongoing is lying to you, haha. That said, I will say if the KBs are well-written and tuned upfront, any ongoing effort should be minimal & tolerable, if not welcomed.

• Q: Are there any defects you have noticed with this feature? 
• A: The biggest issues for us are the KB formatting as well as the display challenges within the new multi-panel config. Also as mentioned above, if you do have tables in your KBs, the answer highlighting doesn't work as well, if at all, and you cannot see the article content in the surfaced article (only the title). Although not a "defect", if you do choose to use Copilot's Answer Generation feature, have your staff give feedback on its accuracy. We turned Answer Generation off given it proved too risky for our lines of business to rely on [at this time] versus reviewing the highlighted answer or full KB. That's truly a "results & adoption varies by customer" feature.

Hi Brian,

Thanks so much for sharing your experience and insights with Knowledge Surfacing. That is incredibly helpful and encouraging to hear! Glad to hear that KS has been responsive is continuing to learn! I hope you don't mind a couple of additional questions:

• What was your experience with the initial effort required to get started with trigger rules and phrases? 
• Is there a specific KB article format/structure that performs better?
• Have you found that there is ongoing manual effort required to finetune responses and phrases?
• Are there any defects you have noticed with this feature? 

Thanks again for taking time to respond. Really appreciate it :)

Deepa

Greetings @Deepa Galaiya - Our experience has been great thus far with the [Agent Copilot] knowledge surfacing during voice interactions with minimal to no lag time. The only latency we noticed was based off the fact that Copilot seems to wait to do its KB article search once the Caller pauses their speech or stops talking altogether. The surfacing has been so good that it often continues to refine/requery the knowledge results as the conversation continues, which means the agent may have more than one KB article result to consider over time.

That said, it's important to note that the answers to your questions depend heavily on:

• Network Performance | Any latency in the home/office connectivity impacts overall Genesys Cloud performance.
• Well-Written KB Article Content | If the articles are vague, broad, or have any overlap, the system will have a harder time locating the appropriate KB article. Leveraging KB Article Phrases (where necessary) has helped us dial in KB surfacing.
• Clear Communication | I know...this seems like common sense, but unlike with a digital channel, if Copilot cannot understand what the caller is saying (e.g. mumbled words, slang, no relevant keywords, etc), it either will not surface anything or will not surface the right thing.
• Picking a Side Copilot Should Monitor | In our case we programmed our overall Copilot knowledge surfacing rule to only listen to the Caller side so KB articles were only surfaced based off what the Caller said, and didn't query every time the agent said something as well. Prior to that change we noticed the system was pinging & ponging KB article results, especially if the search terms it chose to use varied based on who was speaking.

------------------------------
Brian T. Jones | Ascension | Senior Specialist - Technology

Original Message:
Sent: 08-27-2025 20:38
From: Deepa Galaiya
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Thanks Jason - Do you know if many Genesys customers have implemented Genesys CoPilot Knowledge surfacing for the voice channel? 

------------------------------
Deepa Galaiya
Product Owner, Customer Interactions

Original Message:
Sent: 08-26-2025 11:52
From: Jason Kleitz
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Hello Deepa,

That's a great question. I'm genuinely curious to see what the Community's experience has been with the responsiveness of the platform.

------------------------------
Jason Kleitz
Online Community Manager/Moderator

Original Message:
Sent: 08-26-2025 04:13
From: Deepa Galaiya
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Hi Genesys Community,

Keen to hear from the community on your feedback or experience with Genesys CoPilot for Knowledge Surfacing especially around responsiveness and usability during live voice interactions. Has anyone experienced latency issues with CoPilot's real-time knowledge surfacing during voice interactions? It would be good to understand if there is any noticeable lag between the conversation and when knowledge articles are surfaced, or the way it searches to determine the right moment to trigger a search. This could make it challenging for agents to navigate the surfaced content while actively on a call. Thanks in advance :)

#ConversationalAI(Bots,AgentAssist,etc.)

------------------------------
Deepa Galaiya
Product Owner, Customer Interactions
------------------------------

Do you already have that custom connector working?

I am just wondering whether the format of your original articles changes after synchronization to Genesys Cloud.

Hi George,

We have to get an API connected so that Genesys can ingest all of our KB articles from Livepro (our KM sysyem). 

Cheers,

Deepa

How are you getting your articles into a Genesys Cloud Knowledge Base for use with Agent Copilot and Bots?

Hi Brian, Apologies for the delayed response. Thanks for sharing your knowledge and experience!

The point about articles that are in table formats is really interesting. We use Live Pro as our Knowledge Management system, and all of our articles are in tabular format. I wonder if this could be a potential showstopper...

Does KS display an AI generated summary of the knowledge article in the Genesys UI?

Thanks again and have a great day!

Deepa

No problem, Deepa:

• Q: What was your experience with the initial effort required to get started with trigger rules and phrases? 
• A: For us it was minimal given we found sufficient success with the default "No other rule applies > search knowledge base" and adding a "Greeting" intent that automatically surfaced our "Opening Greeting" Canned response for the agent. Here's an example:

• Q: Is there a specific KB article format/structure that performs better?
• A: Great question that I'm sure other community members have various opinions on. We have tried both PORTRAIT and LANDSCAPE layouts. As clean as the table format is, it can be challenging to copy/paste/format those tables from our outside source documents, answer-highlighting doesn't work as well (if at all with tables), and the new multi-panel columnar format for the Agent Workspace scrunches up the table to the point where it's hard for our agents to scroll & read both horizontally and vertically. As such, we stuck with the vertical format given it plays better with the multi-panel config, is easier to format, & answer highlighting works fairly consistently.

• Q: Have you found that there is ongoing manual effort required to finetune responses and phrases?
• A: Anyone that says KB/responses/phrase/copilot management is not ongoing is lying to you, haha. That said, I will say if the KBs are well-written and tuned upfront, any ongoing effort should be minimal & tolerable, if not welcomed.

• Q: Are there any defects you have noticed with this feature? 
• A: The biggest issues for us are the KB formatting as well as the display challenges within the new multi-panel config. Also as mentioned above, if you do have tables in your KBs, the answer highlighting doesn't work as well, if at all, and you cannot see the article content in the surfaced article (only the title). Although not a "defect", if you do choose to use Copilot's Answer Generation feature, have your staff give feedback on its accuracy. We turned Answer Generation off given it proved too risky for our lines of business to rely on [at this time] versus reviewing the highlighted answer or full KB. That's truly a "results & adoption varies by customer" feature.

Hi Brian,

Thanks so much for sharing your experience and insights with Knowledge Surfacing. That is incredibly helpful and encouraging to hear! Glad to hear that KS has been responsive is continuing to learn! I hope you don't mind a couple of additional questions:

• What was your experience with the initial effort required to get started with trigger rules and phrases? 
• Is there a specific KB article format/structure that performs better?
• Have you found that there is ongoing manual effort required to finetune responses and phrases?
• Are there any defects you have noticed with this feature? 

Thanks again for taking time to respond. Really appreciate it :)

Deepa

Greetings @Deepa Galaiya - Our experience has been great thus far with the [Agent Copilot] knowledge surfacing during voice interactions with minimal to no lag time. The only latency we noticed was based off the fact that Copilot seems to wait to do its KB article search once the Caller pauses their speech or stops talking altogether. The surfacing has been so good that it often continues to refine/requery the knowledge results as the conversation continues, which means the agent may have more than one KB article result to consider over time.

That said, it's important to note that the answers to your questions depend heavily on:

• Network Performance | Any latency in the home/office connectivity impacts overall Genesys Cloud performance.
• Well-Written KB Article Content | If the articles are vague, broad, or have any overlap, the system will have a harder time locating the appropriate KB article. Leveraging KB Article Phrases (where necessary) has helped us dial in KB surfacing.
• Clear Communication | I know...this seems like common sense, but unlike with a digital channel, if Copilot cannot understand what the caller is saying (e.g. mumbled words, slang, no relevant keywords, etc), it either will not surface anything or will not surface the right thing.
• Picking a Side Copilot Should Monitor | In our case we programmed our overall Copilot knowledge surfacing rule to only listen to the Caller side so KB articles were only surfaced based off what the Caller said, and didn't query every time the agent said something as well. Prior to that change we noticed the system was pinging & ponging KB article results, especially if the search terms it chose to use varied based on who was speaking.

------------------------------
Brian T. Jones | Ascension | Senior Specialist - Technology

Original Message:
Sent: 08-27-2025 20:38
From: Deepa Galaiya
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Thanks Jason - Do you know if many Genesys customers have implemented Genesys CoPilot Knowledge surfacing for the voice channel? 

------------------------------
Deepa Galaiya
Product Owner, Customer Interactions

Original Message:
Sent: 08-26-2025 11:52
From: Jason Kleitz
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Hello Deepa,

That's a great question. I'm genuinely curious to see what the Community's experience has been with the responsiveness of the platform.

------------------------------
Jason Kleitz
Online Community Manager/Moderator

Original Message:
Sent: 08-26-2025 04:13
From: Deepa Galaiya
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Hi Genesys Community,

Keen to hear from the community on your feedback or experience with Genesys CoPilot for Knowledge Surfacing especially around responsiveness and usability during live voice interactions. Has anyone experienced latency issues with CoPilot's real-time knowledge surfacing during voice interactions? It would be good to understand if there is any noticeable lag between the conversation and when knowledge articles are surfaced, or the way it searches to determine the right moment to trigger a search. This could make it challenging for agents to navigate the surfaced content while actively on a call. Thanks in advance :)

#ConversationalAI(Bots,AgentAssist,etc.)

------------------------------
Deepa Galaiya
Product Owner, Customer Interactions
------------------------------

Two important points I'd also consider here with knowledge articles and Agent Copilot.

Titles and Phrasing

I see a lot of article titles written from a point of view of clicking on a list or searching. Copilot is searching based on a human conversation so consider the phrases that customers or agents are using. Topic Mining and transcript reviews can assist here. Customers may not understand business terms e.g. FNOL (First Notification of Loss) so that's less likely to help in the title but "My car was stolen" is more likely a phrase that would point in the right direction.

Content

Whilst it's easy to import from another source, consider if the article is written from the point of view of a call. A page of process steps may be better as a Next Best Action or future Guided Flows. If the agent has to read a long article maybe that is better delivered as a small e-learning session with the main points to remember in the article.

Do you already have that custom connector working?

I am just wondering whether the format of your original articles changes after synchronization to Genesys Cloud.

Hi George,

We have to get an API connected so that Genesys can ingest all of our KB articles from Livepro (our KM sysyem). 

Cheers,

Deepa

How are you getting your articles into a Genesys Cloud Knowledge Base for use with Agent Copilot and Bots?

Hi Brian, Apologies for the delayed response. Thanks for sharing your knowledge and experience!

The point about articles that are in table formats is really interesting. We use Live Pro as our Knowledge Management system, and all of our articles are in tabular format. I wonder if this could be a potential showstopper...

Does KS display an AI generated summary of the knowledge article in the Genesys UI?

Thanks again and have a great day!

Deepa

No problem, Deepa:

• Q: What was your experience with the initial effort required to get started with trigger rules and phrases? 
• A: For us it was minimal given we found sufficient success with the default "No other rule applies > search knowledge base" and adding a "Greeting" intent that automatically surfaced our "Opening Greeting" Canned response for the agent. Here's an example:

• Q: Is there a specific KB article format/structure that performs better?
• A: Great question that I'm sure other community members have various opinions on. We have tried both PORTRAIT and LANDSCAPE layouts. As clean as the table format is, it can be challenging to copy/paste/format those tables from our outside source documents, answer-highlighting doesn't work as well (if at all with tables), and the new multi-panel columnar format for the Agent Workspace scrunches up the table to the point where it's hard for our agents to scroll & read both horizontally and vertically. As such, we stuck with the vertical format given it plays better with the multi-panel config, is easier to format, & answer highlighting works fairly consistently.

• Q: Have you found that there is ongoing manual effort required to finetune responses and phrases?
• A: Anyone that says KB/responses/phrase/copilot management is not ongoing is lying to you, haha. That said, I will say if the KBs are well-written and tuned upfront, any ongoing effort should be minimal & tolerable, if not welcomed.

• Q: Are there any defects you have noticed with this feature? 
• A: The biggest issues for us are the KB formatting as well as the display challenges within the new multi-panel config. Also as mentioned above, if you do have tables in your KBs, the answer highlighting doesn't work as well, if at all, and you cannot see the article content in the surfaced article (only the title). Although not a "defect", if you do choose to use Copilot's Answer Generation feature, have your staff give feedback on its accuracy. We turned Answer Generation off given it proved too risky for our lines of business to rely on [at this time] versus reviewing the highlighted answer or full KB. That's truly a "results & adoption varies by customer" feature.

Hi Brian,

Thanks so much for sharing your experience and insights with Knowledge Surfacing. That is incredibly helpful and encouraging to hear! Glad to hear that KS has been responsive is continuing to learn! I hope you don't mind a couple of additional questions:

• What was your experience with the initial effort required to get started with trigger rules and phrases? 
• Is there a specific KB article format/structure that performs better?
• Have you found that there is ongoing manual effort required to finetune responses and phrases?
• Are there any defects you have noticed with this feature? 

Thanks again for taking time to respond. Really appreciate it :)

Deepa

Greetings @Deepa Galaiya - Our experience has been great thus far with the [Agent Copilot] knowledge surfacing during voice interactions with minimal to no lag time. The only latency we noticed was based off the fact that Copilot seems to wait to do its KB article search once the Caller pauses their speech or stops talking altogether. The surfacing has been so good that it often continues to refine/requery the knowledge results as the conversation continues, which means the agent may have more than one KB article result to consider over time.

That said, it's important to note that the answers to your questions depend heavily on:

• Network Performance | Any latency in the home/office connectivity impacts overall Genesys Cloud performance.
• Well-Written KB Article Content | If the articles are vague, broad, or have any overlap, the system will have a harder time locating the appropriate KB article. Leveraging KB Article Phrases (where necessary) has helped us dial in KB surfacing.
• Clear Communication | I know...this seems like common sense, but unlike with a digital channel, if Copilot cannot understand what the caller is saying (e.g. mumbled words, slang, no relevant keywords, etc), it either will not surface anything or will not surface the right thing.
• Picking a Side Copilot Should Monitor | In our case we programmed our overall Copilot knowledge surfacing rule to only listen to the Caller side so KB articles were only surfaced based off what the Caller said, and didn't query every time the agent said something as well. Prior to that change we noticed the system was pinging & ponging KB article results, especially if the search terms it chose to use varied based on who was speaking.

------------------------------
Brian T. Jones | Ascension | Senior Specialist - Technology

Original Message:
Sent: 08-27-2025 20:38
From: Deepa Galaiya
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Thanks Jason - Do you know if many Genesys customers have implemented Genesys CoPilot Knowledge surfacing for the voice channel? 

------------------------------
Deepa Galaiya
Product Owner, Customer Interactions

Original Message:
Sent: 08-26-2025 11:52
From: Jason Kleitz
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Hello Deepa,

That's a great question. I'm genuinely curious to see what the Community's experience has been with the responsiveness of the platform.

------------------------------
Jason Kleitz
Online Community Manager/Moderator

Original Message:
Sent: 08-26-2025 04:13
From: Deepa Galaiya
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Hi Genesys Community,

Keen to hear from the community on your feedback or experience with Genesys CoPilot for Knowledge Surfacing especially around responsiveness and usability during live voice interactions. Has anyone experienced latency issues with CoPilot's real-time knowledge surfacing during voice interactions? It would be good to understand if there is any noticeable lag between the conversation and when knowledge articles are surfaced, or the way it searches to determine the right moment to trigger a search. This could make it challenging for agents to navigate the surfaced content while actively on a call. Thanks in advance :)

#ConversationalAI(Bots,AgentAssist,etc.)

------------------------------
Deepa Galaiya
Product Owner, Customer Interactions
------------------------------

Hi Richard,

Thanks for sharing your top tips. For titles and phrasing, did you fine there was a lot of set up effort involved with matching customer language to the relevant knowledge- based articles? How did you manage this prework? I can imagine this must be ongoing. Do you have an effective feedback mechanism to guide your effort and is it easy or labour intensive?

Your comment about content is really interesting. I am struggling to differentiate use cases between NBA and Guided flows as well as the overlap with information that is contained in KB articles. Process steps populate in KB articles in our current state so there would be some rethinking required here.

Thanks again for sharing your experience.

Cheers,

Deepa 

Two important points I'd also consider here with knowledge articles and Agent Copilot.

Titles and Phrasing

I see a lot of article titles written from a point of view of clicking on a list or searching. Copilot is searching based on a human conversation so consider the phrases that customers or agents are using. Topic Mining and transcript reviews can assist here. Customers may not understand business terms e.g. FNOL (First Notification of Loss) so that's less likely to help in the title but "My car was stolen" is more likely a phrase that would point in the right direction.

Content

Whilst it's easy to import from another source, consider if the article is written from the point of view of a call. A page of process steps may be better as a Next Best Action or future Guided Flows. If the agent has to read a long article maybe that is better delivered as a small e-learning session with the main points to remember in the article.

Do you already have that custom connector working?

I am just wondering whether the format of your original articles changes after synchronization to Genesys Cloud.

Hi George,

We have to get an API connected so that Genesys can ingest all of our KB articles from Livepro (our KM sysyem). 

Cheers,

Deepa

How are you getting your articles into a Genesys Cloud Knowledge Base for use with Agent Copilot and Bots?

Hi Brian, Apologies for the delayed response. Thanks for sharing your knowledge and experience!

The point about articles that are in table formats is really interesting. We use Live Pro as our Knowledge Management system, and all of our articles are in tabular format. I wonder if this could be a potential showstopper...

Does KS display an AI generated summary of the knowledge article in the Genesys UI?

Thanks again and have a great day!

Deepa

No problem, Deepa:

• Q: What was your experience with the initial effort required to get started with trigger rules and phrases? 
• A: For us it was minimal given we found sufficient success with the default "No other rule applies > search knowledge base" and adding a "Greeting" intent that automatically surfaced our "Opening Greeting" Canned response for the agent. Here's an example:

• Q: Is there a specific KB article format/structure that performs better?
• A: Great question that I'm sure other community members have various opinions on. We have tried both PORTRAIT and LANDSCAPE layouts. As clean as the table format is, it can be challenging to copy/paste/format those tables from our outside source documents, answer-highlighting doesn't work as well (if at all with tables), and the new multi-panel columnar format for the Agent Workspace scrunches up the table to the point where it's hard for our agents to scroll & read both horizontally and vertically. As such, we stuck with the vertical format given it plays better with the multi-panel config, is easier to format, & answer highlighting works fairly consistently.

• Q: Have you found that there is ongoing manual effort required to finetune responses and phrases?
• A: Anyone that says KB/responses/phrase/copilot management is not ongoing is lying to you, haha. That said, I will say if the KBs are well-written and tuned upfront, any ongoing effort should be minimal & tolerable, if not welcomed.

• Q: Are there any defects you have noticed with this feature? 
• A: The biggest issues for us are the KB formatting as well as the display challenges within the new multi-panel config. Also as mentioned above, if you do have tables in your KBs, the answer highlighting doesn't work as well, if at all, and you cannot see the article content in the surfaced article (only the title). Although not a "defect", if you do choose to use Copilot's Answer Generation feature, have your staff give feedback on its accuracy. We turned Answer Generation off given it proved too risky for our lines of business to rely on [at this time] versus reviewing the highlighted answer or full KB. That's truly a "results & adoption varies by customer" feature.

Hi Brian,

Thanks so much for sharing your experience and insights with Knowledge Surfacing. That is incredibly helpful and encouraging to hear! Glad to hear that KS has been responsive is continuing to learn! I hope you don't mind a couple of additional questions:

• What was your experience with the initial effort required to get started with trigger rules and phrases? 
• Is there a specific KB article format/structure that performs better?
• Have you found that there is ongoing manual effort required to finetune responses and phrases?
• Are there any defects you have noticed with this feature? 

Thanks again for taking time to respond. Really appreciate it :)

Deepa

Greetings @Deepa Galaiya - Our experience has been great thus far with the [Agent Copilot] knowledge surfacing during voice interactions with minimal to no lag time. The only latency we noticed was based off the fact that Copilot seems to wait to do its KB article search once the Caller pauses their speech or stops talking altogether. The surfacing has been so good that it often continues to refine/requery the knowledge results as the conversation continues, which means the agent may have more than one KB article result to consider over time.

That said, it's important to note that the answers to your questions depend heavily on:

• Network Performance | Any latency in the home/office connectivity impacts overall Genesys Cloud performance.
• Well-Written KB Article Content | If the articles are vague, broad, or have any overlap, the system will have a harder time locating the appropriate KB article. Leveraging KB Article Phrases (where necessary) has helped us dial in KB surfacing.
• Clear Communication | I know...this seems like common sense, but unlike with a digital channel, if Copilot cannot understand what the caller is saying (e.g. mumbled words, slang, no relevant keywords, etc), it either will not surface anything or will not surface the right thing.
• Picking a Side Copilot Should Monitor | In our case we programmed our overall Copilot knowledge surfacing rule to only listen to the Caller side so KB articles were only surfaced based off what the Caller said, and didn't query every time the agent said something as well. Prior to that change we noticed the system was pinging & ponging KB article results, especially if the search terms it chose to use varied based on who was speaking.

------------------------------
Brian T. Jones | Ascension | Senior Specialist - Technology

Original Message:
Sent: 08-27-2025 20:38
From: Deepa Galaiya
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Thanks Jason - Do you know if many Genesys customers have implemented Genesys CoPilot Knowledge surfacing for the voice channel? 

------------------------------
Deepa Galaiya
Product Owner, Customer Interactions

Original Message:
Sent: 08-26-2025 11:52
From: Jason Kleitz
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Hello Deepa,

That's a great question. I'm genuinely curious to see what the Community's experience has been with the responsiveness of the platform.

------------------------------
Jason Kleitz
Online Community Manager/Moderator

Original Message:
Sent: 08-26-2025 04:13
From: Deepa Galaiya
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Hi Genesys Community,

Keen to hear from the community on your feedback or experience with Genesys CoPilot for Knowledge Surfacing especially around responsiveness and usability during live voice interactions. Has anyone experienced latency issues with CoPilot's real-time knowledge surfacing during voice interactions? It would be good to understand if there is any noticeable lag between the conversation and when knowledge articles are surfaced, or the way it searches to determine the right moment to trigger a search. This could make it challenging for agents to navigate the surfaced content while actively on a call. Thanks in advance :)

#ConversationalAI(Bots,AgentAssist,etc.)

------------------------------
Deepa Galaiya
Product Owner, Customer Interactions
------------------------------

Hi Deepa,

One or two phrases is likely to help a lot. You can always use the Intent Generation to give you ideas of useful phrases. This becomes easier once you add new knowledge but can take time for an existing FAQ list. Variations in knowledge enable you to reformat for voice. That may be putting bullet points at the top or using a script page. Guided flows for processes will be very useful for more complex articles if it is to follow a step by step process.

Hi Richard,

Thanks for sharing your top tips. For titles and phrasing, did you fine there was a lot of set up effort involved with matching customer language to the relevant knowledge- based articles? How did you manage this prework? I can imagine this must be ongoing. Do you have an effective feedback mechanism to guide your effort and is it easy or labour intensive?

Your comment about content is really interesting. I am struggling to differentiate use cases between NBA and Guided flows as well as the overlap with information that is contained in KB articles. Process steps populate in KB articles in our current state so there would be some rethinking required here.

Thanks again for sharing your experience.

Cheers,

Deepa 

Two important points I'd also consider here with knowledge articles and Agent Copilot.

Titles and Phrasing

I see a lot of article titles written from a point of view of clicking on a list or searching. Copilot is searching based on a human conversation so consider the phrases that customers or agents are using. Topic Mining and transcript reviews can assist here. Customers may not understand business terms e.g. FNOL (First Notification of Loss) so that's less likely to help in the title but "My car was stolen" is more likely a phrase that would point in the right direction.

Content

Whilst it's easy to import from another source, consider if the article is written from the point of view of a call. A page of process steps may be better as a Next Best Action or future Guided Flows. If the agent has to read a long article maybe that is better delivered as a small e-learning session with the main points to remember in the article.

Do you already have that custom connector working?

I am just wondering whether the format of your original articles changes after synchronization to Genesys Cloud.

Hi George,

We have to get an API connected so that Genesys can ingest all of our KB articles from Livepro (our KM sysyem). 

Cheers,

Deepa

How are you getting your articles into a Genesys Cloud Knowledge Base for use with Agent Copilot and Bots?

Hi Brian, Apologies for the delayed response. Thanks for sharing your knowledge and experience!

The point about articles that are in table formats is really interesting. We use Live Pro as our Knowledge Management system, and all of our articles are in tabular format. I wonder if this could be a potential showstopper...

Does KS display an AI generated summary of the knowledge article in the Genesys UI?

Thanks again and have a great day!

Deepa

No problem, Deepa:

• Q: What was your experience with the initial effort required to get started with trigger rules and phrases? 
• A: For us it was minimal given we found sufficient success with the default "No other rule applies > search knowledge base" and adding a "Greeting" intent that automatically surfaced our "Opening Greeting" Canned response for the agent. Here's an example:

• Q: Is there a specific KB article format/structure that performs better?
• A: Great question that I'm sure other community members have various opinions on. We have tried both PORTRAIT and LANDSCAPE layouts. As clean as the table format is, it can be challenging to copy/paste/format those tables from our outside source documents, answer-highlighting doesn't work as well (if at all with tables), and the new multi-panel columnar format for the Agent Workspace scrunches up the table to the point where it's hard for our agents to scroll & read both horizontally and vertically. As such, we stuck with the vertical format given it plays better with the multi-panel config, is easier to format, & answer highlighting works fairly consistently.

• Q: Have you found that there is ongoing manual effort required to finetune responses and phrases?
• A: Anyone that says KB/responses/phrase/copilot management is not ongoing is lying to you, haha. That said, I will say if the KBs are well-written and tuned upfront, any ongoing effort should be minimal & tolerable, if not welcomed.

• Q: Are there any defects you have noticed with this feature? 
• A: The biggest issues for us are the KB formatting as well as the display challenges within the new multi-panel config. Also as mentioned above, if you do have tables in your KBs, the answer highlighting doesn't work as well, if at all, and you cannot see the article content in the surfaced article (only the title). Although not a "defect", if you do choose to use Copilot's Answer Generation feature, have your staff give feedback on its accuracy. We turned Answer Generation off given it proved too risky for our lines of business to rely on [at this time] versus reviewing the highlighted answer or full KB. That's truly a "results & adoption varies by customer" feature.

Hi Brian,

Thanks so much for sharing your experience and insights with Knowledge Surfacing. That is incredibly helpful and encouraging to hear! Glad to hear that KS has been responsive is continuing to learn! I hope you don't mind a couple of additional questions:

• What was your experience with the initial effort required to get started with trigger rules and phrases? 
• Is there a specific KB article format/structure that performs better?
• Have you found that there is ongoing manual effort required to finetune responses and phrases?
• Are there any defects you have noticed with this feature? 

Thanks again for taking time to respond. Really appreciate it :)

Deepa

Greetings @Deepa Galaiya - Our experience has been great thus far with the [Agent Copilot] knowledge surfacing during voice interactions with minimal to no lag time. The only latency we noticed was based off the fact that Copilot seems to wait to do its KB article search once the Caller pauses their speech or stops talking altogether. The surfacing has been so good that it often continues to refine/requery the knowledge results as the conversation continues, which means the agent may have more than one KB article result to consider over time.

That said, it's important to note that the answers to your questions depend heavily on:

• Network Performance | Any latency in the home/office connectivity impacts overall Genesys Cloud performance.
• Well-Written KB Article Content | If the articles are vague, broad, or have any overlap, the system will have a harder time locating the appropriate KB article. Leveraging KB Article Phrases (where necessary) has helped us dial in KB surfacing.
• Clear Communication | I know...this seems like common sense, but unlike with a digital channel, if Copilot cannot understand what the caller is saying (e.g. mumbled words, slang, no relevant keywords, etc), it either will not surface anything or will not surface the right thing.
• Picking a Side Copilot Should Monitor | In our case we programmed our overall Copilot knowledge surfacing rule to only listen to the Caller side so KB articles were only surfaced based off what the Caller said, and didn't query every time the agent said something as well. Prior to that change we noticed the system was pinging & ponging KB article results, especially if the search terms it chose to use varied based on who was speaking.

------------------------------
Brian T. Jones | Ascension | Senior Specialist - Technology

Original Message:
Sent: 08-27-2025 20:38
From: Deepa Galaiya
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Thanks Jason - Do you know if many Genesys customers have implemented Genesys CoPilot Knowledge surfacing for the voice channel? 

------------------------------
Deepa Galaiya
Product Owner, Customer Interactions

Original Message:
Sent: 08-26-2025 11:52
From: Jason Kleitz
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Hello Deepa,

That's a great question. I'm genuinely curious to see what the Community's experience has been with the responsiveness of the platform.

------------------------------
Jason Kleitz
Online Community Manager/Moderator

Original Message:
Sent: 08-26-2025 04:13
From: Deepa Galaiya
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Hi Genesys Community,

Keen to hear from the community on your feedback or experience with Genesys CoPilot for Knowledge Surfacing especially around responsiveness and usability during live voice interactions. Has anyone experienced latency issues with CoPilot's real-time knowledge surfacing during voice interactions? It would be good to understand if there is any noticeable lag between the conversation and when knowledge articles are surfaced, or the way it searches to determine the right moment to trigger a search. This could make it challenging for agents to navigate the surfaced content while actively on a call. Thanks in advance :)

#ConversationalAI(Bots,AgentAssist,etc.)

------------------------------
Deepa Galaiya
Product Owner, Customer Interactions
------------------------------

Hi Richard, Thanks for your response. We have an established KB system with 1000s of articles so will need to think through a strategy to make this successful. Forgive this naive question... You suggested using 'intent generation' - Is this an existing Genesys capability? We have a list of known topics on why customers call us based on IVR selection and predefined wrap codes but it would be useful to have this in the customers own language/ phrase.  

Hi Deepa,

One or two phrases is likely to help a lot. You can always use the Intent Generation to give you ideas of useful phrases. This becomes easier once you add new knowledge but can take time for an existing FAQ list. Variations in knowledge enable you to reformat for voice. That may be putting bullet points at the top or using a script page. Guided flows for processes will be very useful for more complex articles if it is to follow a step by step process.

Hi Richard,

Thanks for sharing your top tips. For titles and phrasing, did you fine there was a lot of set up effort involved with matching customer language to the relevant knowledge- based articles? How did you manage this prework? I can imagine this must be ongoing. Do you have an effective feedback mechanism to guide your effort and is it easy or labour intensive?

Your comment about content is really interesting. I am struggling to differentiate use cases between NBA and Guided flows as well as the overlap with information that is contained in KB articles. Process steps populate in KB articles in our current state so there would be some rethinking required here.

Thanks again for sharing your experience.

Cheers,

Deepa 

Two important points I'd also consider here with knowledge articles and Agent Copilot.

Titles and Phrasing

I see a lot of article titles written from a point of view of clicking on a list or searching. Copilot is searching based on a human conversation so consider the phrases that customers or agents are using. Topic Mining and transcript reviews can assist here. Customers may not understand business terms e.g. FNOL (First Notification of Loss) so that's less likely to help in the title but "My car was stolen" is more likely a phrase that would point in the right direction.

Content

Whilst it's easy to import from another source, consider if the article is written from the point of view of a call. A page of process steps may be better as a Next Best Action or future Guided Flows. If the agent has to read a long article maybe that is better delivered as a small e-learning session with the main points to remember in the article.

Do you already have that custom connector working?

I am just wondering whether the format of your original articles changes after synchronization to Genesys Cloud.

Hi George,

We have to get an API connected so that Genesys can ingest all of our KB articles from Livepro (our KM sysyem). 

Cheers,

Deepa

How are you getting your articles into a Genesys Cloud Knowledge Base for use with Agent Copilot and Bots?

Hi Brian, Apologies for the delayed response. Thanks for sharing your knowledge and experience!

The point about articles that are in table formats is really interesting. We use Live Pro as our Knowledge Management system, and all of our articles are in tabular format. I wonder if this could be a potential showstopper...

Does KS display an AI generated summary of the knowledge article in the Genesys UI?

Thanks again and have a great day!

Deepa

No problem, Deepa:

• Q: What was your experience with the initial effort required to get started with trigger rules and phrases? 
• A: For us it was minimal given we found sufficient success with the default "No other rule applies > search knowledge base" and adding a "Greeting" intent that automatically surfaced our "Opening Greeting" Canned response for the agent. Here's an example:

• Q: Is there a specific KB article format/structure that performs better?
• A: Great question that I'm sure other community members have various opinions on. We have tried both PORTRAIT and LANDSCAPE layouts. As clean as the table format is, it can be challenging to copy/paste/format those tables from our outside source documents, answer-highlighting doesn't work as well (if at all with tables), and the new multi-panel columnar format for the Agent Workspace scrunches up the table to the point where it's hard for our agents to scroll & read both horizontally and vertically. As such, we stuck with the vertical format given it plays better with the multi-panel config, is easier to format, & answer highlighting works fairly consistently.

• Q: Have you found that there is ongoing manual effort required to finetune responses and phrases?
• A: Anyone that says KB/responses/phrase/copilot management is not ongoing is lying to you, haha. That said, I will say if the KBs are well-written and tuned upfront, any ongoing effort should be minimal & tolerable, if not welcomed.

• Q: Are there any defects you have noticed with this feature? 
• A: The biggest issues for us are the KB formatting as well as the display challenges within the new multi-panel config. Also as mentioned above, if you do have tables in your KBs, the answer highlighting doesn't work as well, if at all, and you cannot see the article content in the surfaced article (only the title). Although not a "defect", if you do choose to use Copilot's Answer Generation feature, have your staff give feedback on its accuracy. We turned Answer Generation off given it proved too risky for our lines of business to rely on [at this time] versus reviewing the highlighted answer or full KB. That's truly a "results & adoption varies by customer" feature.

Hi Brian,

Thanks so much for sharing your experience and insights with Knowledge Surfacing. That is incredibly helpful and encouraging to hear! Glad to hear that KS has been responsive is continuing to learn! I hope you don't mind a couple of additional questions:

• What was your experience with the initial effort required to get started with trigger rules and phrases? 
• Is there a specific KB article format/structure that performs better?
• Have you found that there is ongoing manual effort required to finetune responses and phrases?
• Are there any defects you have noticed with this feature? 

Thanks again for taking time to respond. Really appreciate it :)

Deepa

Greetings @Deepa Galaiya - Our experience has been great thus far with the [Agent Copilot] knowledge surfacing during voice interactions with minimal to no lag time. The only latency we noticed was based off the fact that Copilot seems to wait to do its KB article search once the Caller pauses their speech or stops talking altogether. The surfacing has been so good that it often continues to refine/requery the knowledge results as the conversation continues, which means the agent may have more than one KB article result to consider over time.

That said, it's important to note that the answers to your questions depend heavily on:

• Network Performance | Any latency in the home/office connectivity impacts overall Genesys Cloud performance.
• Well-Written KB Article Content | If the articles are vague, broad, or have any overlap, the system will have a harder time locating the appropriate KB article. Leveraging KB Article Phrases (where necessary) has helped us dial in KB surfacing.
• Clear Communication | I know...this seems like common sense, but unlike with a digital channel, if Copilot cannot understand what the caller is saying (e.g. mumbled words, slang, no relevant keywords, etc), it either will not surface anything or will not surface the right thing.
• Picking a Side Copilot Should Monitor | In our case we programmed our overall Copilot knowledge surfacing rule to only listen to the Caller side so KB articles were only surfaced based off what the Caller said, and didn't query every time the agent said something as well. Prior to that change we noticed the system was pinging & ponging KB article results, especially if the search terms it chose to use varied based on who was speaking.

------------------------------
Brian T. Jones | Ascension | Senior Specialist - Technology

Original Message:
Sent: 08-27-2025 20:38
From: Deepa Galaiya
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Thanks Jason - Do you know if many Genesys customers have implemented Genesys CoPilot Knowledge surfacing for the voice channel? 

------------------------------
Deepa Galaiya
Product Owner, Customer Interactions

Original Message:
Sent: 08-26-2025 11:52
From: Jason Kleitz
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Hello Deepa,

That's a great question. I'm genuinely curious to see what the Community's experience has been with the responsiveness of the platform.

------------------------------
Jason Kleitz
Online Community Manager/Moderator

Original Message:
Sent: 08-26-2025 04:13
From: Deepa Galaiya
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Hi Genesys Community,

Keen to hear from the community on your feedback or experience with Genesys CoPilot for Knowledge Surfacing especially around responsiveness and usability during live voice interactions. Has anyone experienced latency issues with CoPilot's real-time knowledge surfacing during voice interactions? It would be good to understand if there is any noticeable lag between the conversation and when knowledge articles are surfaced, or the way it searches to determine the right moment to trigger a search. This could make it challenging for agents to navigate the surfaced content while actively on a call. Thanks in advance :)

#ConversationalAI(Bots,AgentAssist,etc.)

------------------------------
Deepa Galaiya
Product Owner, Customer Interactions
------------------------------

With Speech and Text Analytics you have Intent Miner and Topic Miner to look at historical phrasing used.

In Copilot you have a section Natural Language Understanding where you can ad Intents and utterances can be created via AI which could be re-used. However doing that for 1000s of articles would be a challenge.

Hi Richard, Thanks for your response. We have an established KB system with 1000s of articles so will need to think through a strategy to make this successful. Forgive this naive question... You suggested using 'intent generation' - Is this an existing Genesys capability? We have a list of known topics on why customers call us based on IVR selection and predefined wrap codes but it would be useful to have this in the customers own language/ phrase.  

Hi Deepa,

One or two phrases is likely to help a lot. You can always use the Intent Generation to give you ideas of useful phrases. This becomes easier once you add new knowledge but can take time for an existing FAQ list. Variations in knowledge enable you to reformat for voice. That may be putting bullet points at the top or using a script page. Guided flows for processes will be very useful for more complex articles if it is to follow a step by step process.

Hi Richard,

Thanks for sharing your top tips. For titles and phrasing, did you fine there was a lot of set up effort involved with matching customer language to the relevant knowledge- based articles? How did you manage this prework? I can imagine this must be ongoing. Do you have an effective feedback mechanism to guide your effort and is it easy or labour intensive?

Your comment about content is really interesting. I am struggling to differentiate use cases between NBA and Guided flows as well as the overlap with information that is contained in KB articles. Process steps populate in KB articles in our current state so there would be some rethinking required here.

Thanks again for sharing your experience.

Cheers,

Deepa 

Two important points I'd also consider here with knowledge articles and Agent Copilot.

Titles and Phrasing

I see a lot of article titles written from a point of view of clicking on a list or searching. Copilot is searching based on a human conversation so consider the phrases that customers or agents are using. Topic Mining and transcript reviews can assist here. Customers may not understand business terms e.g. FNOL (First Notification of Loss) so that's less likely to help in the title but "My car was stolen" is more likely a phrase that would point in the right direction.

Content

Whilst it's easy to import from another source, consider if the article is written from the point of view of a call. A page of process steps may be better as a Next Best Action or future Guided Flows. If the agent has to read a long article maybe that is better delivered as a small e-learning session with the main points to remember in the article.

Do you already have that custom connector working?

I am just wondering whether the format of your original articles changes after synchronization to Genesys Cloud.

Hi George,

We have to get an API connected so that Genesys can ingest all of our KB articles from Livepro (our KM sysyem). 

Cheers,

Deepa

How are you getting your articles into a Genesys Cloud Knowledge Base for use with Agent Copilot and Bots?

Hi Brian, Apologies for the delayed response. Thanks for sharing your knowledge and experience!

The point about articles that are in table formats is really interesting. We use Live Pro as our Knowledge Management system, and all of our articles are in tabular format. I wonder if this could be a potential showstopper...

Does KS display an AI generated summary of the knowledge article in the Genesys UI?

Thanks again and have a great day!

Deepa

No problem, Deepa:

• Q: What was your experience with the initial effort required to get started with trigger rules and phrases? 
• A: For us it was minimal given we found sufficient success with the default "No other rule applies > search knowledge base" and adding a "Greeting" intent that automatically surfaced our "Opening Greeting" Canned response for the agent. Here's an example:

• Q: Is there a specific KB article format/structure that performs better?
• A: Great question that I'm sure other community members have various opinions on. We have tried both PORTRAIT and LANDSCAPE layouts. As clean as the table format is, it can be challenging to copy/paste/format those tables from our outside source documents, answer-highlighting doesn't work as well (if at all with tables), and the new multi-panel columnar format for the Agent Workspace scrunches up the table to the point where it's hard for our agents to scroll & read both horizontally and vertically. As such, we stuck with the vertical format given it plays better with the multi-panel config, is easier to format, & answer highlighting works fairly consistently.

• Q: Have you found that there is ongoing manual effort required to finetune responses and phrases?
• A: Anyone that says KB/responses/phrase/copilot management is not ongoing is lying to you, haha. That said, I will say if the KBs are well-written and tuned upfront, any ongoing effort should be minimal & tolerable, if not welcomed.

• Q: Are there any defects you have noticed with this feature? 
• A: The biggest issues for us are the KB formatting as well as the display challenges within the new multi-panel config. Also as mentioned above, if you do have tables in your KBs, the answer highlighting doesn't work as well, if at all, and you cannot see the article content in the surfaced article (only the title). Although not a "defect", if you do choose to use Copilot's Answer Generation feature, have your staff give feedback on its accuracy. We turned Answer Generation off given it proved too risky for our lines of business to rely on [at this time] versus reviewing the highlighted answer or full KB. That's truly a "results & adoption varies by customer" feature.

Hi Brian,

Thanks so much for sharing your experience and insights with Knowledge Surfacing. That is incredibly helpful and encouraging to hear! Glad to hear that KS has been responsive is continuing to learn! I hope you don't mind a couple of additional questions:

• What was your experience with the initial effort required to get started with trigger rules and phrases? 
• Is there a specific KB article format/structure that performs better?
• Have you found that there is ongoing manual effort required to finetune responses and phrases?
• Are there any defects you have noticed with this feature? 

Thanks again for taking time to respond. Really appreciate it :)

Deepa

Greetings @Deepa Galaiya - Our experience has been great thus far with the [Agent Copilot] knowledge surfacing during voice interactions with minimal to no lag time. The only latency we noticed was based off the fact that Copilot seems to wait to do its KB article search once the Caller pauses their speech or stops talking altogether. The surfacing has been so good that it often continues to refine/requery the knowledge results as the conversation continues, which means the agent may have more than one KB article result to consider over time.

That said, it's important to note that the answers to your questions depend heavily on:

• Network Performance | Any latency in the home/office connectivity impacts overall Genesys Cloud performance.
• Well-Written KB Article Content | If the articles are vague, broad, or have any overlap, the system will have a harder time locating the appropriate KB article. Leveraging KB Article Phrases (where necessary) has helped us dial in KB surfacing.
• Clear Communication | I know...this seems like common sense, but unlike with a digital channel, if Copilot cannot understand what the caller is saying (e.g. mumbled words, slang, no relevant keywords, etc), it either will not surface anything or will not surface the right thing.
• Picking a Side Copilot Should Monitor | In our case we programmed our overall Copilot knowledge surfacing rule to only listen to the Caller side so KB articles were only surfaced based off what the Caller said, and didn't query every time the agent said something as well. Prior to that change we noticed the system was pinging & ponging KB article results, especially if the search terms it chose to use varied based on who was speaking.

------------------------------
Brian T. Jones | Ascension | Senior Specialist - Technology

Original Message:
Sent: 08-27-2025 20:38
From: Deepa Galaiya
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Thanks Jason - Do you know if many Genesys customers have implemented Genesys CoPilot Knowledge surfacing for the voice channel? 

------------------------------
Deepa Galaiya
Product Owner, Customer Interactions

Original Message:
Sent: 08-26-2025 11:52
From: Jason Kleitz
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Hello Deepa,

That's a great question. I'm genuinely curious to see what the Community's experience has been with the responsiveness of the platform.

------------------------------
Jason Kleitz
Online Community Manager/Moderator

Original Message:
Sent: 08-26-2025 04:13
From: Deepa Galaiya
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Hi Genesys Community,

Keen to hear from the community on your feedback or experience with Genesys CoPilot for Knowledge Surfacing especially around responsiveness and usability during live voice interactions. Has anyone experienced latency issues with CoPilot's real-time knowledge surfacing during voice interactions? It would be good to understand if there is any noticeable lag between the conversation and when knowledge articles are surfaced, or the way it searches to determine the right moment to trigger a search. This could make it challenging for agents to navigate the surfaced content while actively on a call. Thanks in advance :)

#ConversationalAI(Bots,AgentAssist,etc.)

------------------------------
Deepa Galaiya
Product Owner, Customer Interactions
------------------------------

Do you already have that custom connector working?

I am just wondering whether the format of your original articles changes after synchronization to Genesys Cloud.

------------------------------
George Ganahl GCCX-AI, GCP, GCSME
Technical Adoption Champion
Genesys
2024 Community Member of the Year

Original Message:
Sent: 09-05-2025 20:40
From: Deepa Galaiya
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Hi George,

We have to get an API connected so that Genesys can ingest all of our KB articles from Livepro (our KM sysyem). 

Cheers,

Deepa

------------------------------
Deepa Galaiya
Product Owner, Customer Interactions

Original Message:
Sent: 09-05-2025 13:08
From: George Ganahl
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

How are you getting your articles into a Genesys Cloud Knowledge Base for use with Agent Copilot and Bots?

------------------------------
George Ganahl GCCX-AI, GCP, GCSME
Technical Adoption Champion
Genesys
2024 Community Member of the Year

Original Message:
Sent: 09-04-2025 21:31
From: Deepa Galaiya
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Hi Brian, Apologies for the delayed response. Thanks for sharing your knowledge and experience!

The point about articles that are in table formats is really interesting. We use Live Pro as our Knowledge Management system, and all of our articles are in tabular format. I wonder if this could be a potential showstopper...

Does KS display an AI generated summary of the knowledge article in the Genesys UI?

Thanks again and have a great day!

Deepa

------------------------------
Deepa Galaiya
Product Owner, Customer Interactions

Original Message:
Sent: 08-28-2025 17:25
From: Brian Jones
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

No problem, Deepa:

• Q: What was your experience with the initial effort required to get started with trigger rules and phrases? 
• A: For us it was minimal given we found sufficient success with the default "No other rule applies > search knowledge base" and adding a "Greeting" intent that automatically surfaced our "Opening Greeting" Canned response for the agent. Here's an example:

• Q: Is there a specific KB article format/structure that performs better?
• A: Great question that I'm sure other community members have various opinions on. We have tried both PORTRAIT and LANDSCAPE layouts. As clean as the table format is, it can be challenging to copy/paste/format those tables from our outside source documents, answer-highlighting doesn't work as well (if at all with tables), and the new multi-panel columnar format for the Agent Workspace scrunches up the table to the point where it's hard for our agents to scroll & read both horizontally and vertically. As such, we stuck with the vertical format given it plays better with the multi-panel config, is easier to format, & answer highlighting works fairly consistently.

• Q: Have you found that there is ongoing manual effort required to finetune responses and phrases?
• A: Anyone that says KB/responses/phrase/copilot management is not ongoing is lying to you, haha. That said, I will say if the KBs are well-written and tuned upfront, any ongoing effort should be minimal & tolerable, if not welcomed.

• Q: Are there any defects you have noticed with this feature? 
• A: The biggest issues for us are the KB formatting as well as the display challenges within the new multi-panel config. Also as mentioned above, if you do have tables in your KBs, the answer highlighting doesn't work as well, if at all, and you cannot see the article content in the surfaced article (only the title). Although not a "defect", if you do choose to use Copilot's Answer Generation feature, have your staff give feedback on its accuracy. We turned Answer Generation off given it proved too risky for our lines of business to rely on [at this time] versus reviewing the highlighted answer or full KB. That's truly a "results & adoption varies by customer" feature.

------------------------------
Brian T. Jones | Ascension | Senior Specialist - Technology

Original Message:
Sent: 08-28-2025 05:20
From: Deepa Galaiya
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Hi Brian,

Thanks so much for sharing your experience and insights with Knowledge Surfacing. That is incredibly helpful and encouraging to hear! Glad to hear that KS has been responsive is continuing to learn! I hope you don't mind a couple of additional questions:

• What was your experience with the initial effort required to get started with trigger rules and phrases? 
• Is there a specific KB article format/structure that performs better?
• Have you found that there is ongoing manual effort required to finetune responses and phrases?
• Are there any defects you have noticed with this feature? 

Thanks again for taking time to respond. Really appreciate it :)

Deepa

------------------------------
Deepa Galaiya
Product Owner, Customer Interactions

Original Message:
Sent: 08-27-2025 23:59
From: Brian Jones
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Greetings @Deepa Galaiya - Our experience has been great thus far with the [Agent Copilot] knowledge surfacing during voice interactions with minimal to no lag time. The only latency we noticed was based off the fact that Copilot seems to wait to do its KB article search once the Caller pauses their speech or stops talking altogether. The surfacing has been so good that it often continues to refine/requery the knowledge results as the conversation continues, which means the agent may have more than one KB article result to consider over time.

That said, it's important to note that the answers to your questions depend heavily on:

• Network Performance | Any latency in the home/office connectivity impacts overall Genesys Cloud performance.
• Well-Written KB Article Content | If the articles are vague, broad, or have any overlap, the system will have a harder time locating the appropriate KB article. Leveraging KB Article Phrases (where necessary) has helped us dial in KB surfacing.
• Clear Communication | I know...this seems like common sense, but unlike with a digital channel, if Copilot cannot understand what the caller is saying (e.g. mumbled words, slang, no relevant keywords, etc), it either will not surface anything or will not surface the right thing.
• Picking a Side Copilot Should Monitor | In our case we programmed our overall Copilot knowledge surfacing rule to only listen to the Caller side so KB articles were only surfaced based off what the Caller said, and didn't query every time the agent said something as well. Prior to that change we noticed the system was pinging & ponging KB article results, especially if the search terms it chose to use varied based on who was speaking.

------------------------------
Brian T. Jones | Ascension | Senior Specialist - Technology

Original Message:
Sent: 08-27-2025 20:38
From: Deepa Galaiya
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Thanks Jason - Do you know if many Genesys customers have implemented Genesys CoPilot Knowledge surfacing for the voice channel? 

------------------------------
Deepa Galaiya
Product Owner, Customer Interactions

Original Message:
Sent: 08-26-2025 11:52
From: Jason Kleitz
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Hello Deepa,

That's a great question. I'm genuinely curious to see what the Community's experience has been with the responsiveness of the platform.

------------------------------
Jason Kleitz
Online Community Manager/Moderator

Original Message:
Sent: 08-26-2025 04:13
From: Deepa Galaiya
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Hi Genesys Community,

Keen to hear from the community on your feedback or experience with Genesys CoPilot for Knowledge Surfacing especially around responsiveness and usability during live voice interactions. Has anyone experienced latency issues with CoPilot's real-time knowledge surfacing during voice interactions? It would be good to understand if there is any noticeable lag between the conversation and when knowledge articles are surfaced, or the way it searches to determine the right moment to trigger a search. This could make it challenging for agents to navigate the surfaced content while actively on a call. Thanks in advance :)

#ConversationalAI(Bots,AgentAssist,etc.)

------------------------------
Deepa Galaiya
Product Owner, Customer Interactions
------------------------------

@George - Good question. We haven't got the API connector yet. We have recently implemented Auto summary.and predictive wrapcodes and are researching what comes next so that we can realise value from our CoPilot tokens. I wonder if Genesys retains the original structure of the KB articles when it ingests the content or if it reformats it to suit the Genesys UI.

------------------------------
Deepa Galaiya
Product Owner, Customer Interactions

Original Message:
Sent: 09-07-2025 08:18
From: George Ganahl
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Do you already have that custom connector working?

I am just wondering whether the format of your original articles changes after synchronization to Genesys Cloud.

------------------------------
George Ganahl GCCX-AI, GCP, GCSME
Technical Adoption Champion
Genesys
2024 Community Member of the Year

Original Message:
Sent: 09-05-2025 20:40
From: Deepa Galaiya
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Hi George,

We have to get an API connected so that Genesys can ingest all of our KB articles from Livepro (our KM sysyem). 

Cheers,

Deepa

------------------------------
Deepa Galaiya
Product Owner, Customer Interactions

Original Message:
Sent: 09-05-2025 13:08
From: George Ganahl
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

How are you getting your articles into a Genesys Cloud Knowledge Base for use with Agent Copilot and Bots?

------------------------------
George Ganahl GCCX-AI, GCP, GCSME
Technical Adoption Champion
Genesys
2024 Community Member of the Year

Original Message:
Sent: 09-04-2025 21:31
From: Deepa Galaiya
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Hi Brian, Apologies for the delayed response. Thanks for sharing your knowledge and experience!

The point about articles that are in table formats is really interesting. We use Live Pro as our Knowledge Management system, and all of our articles are in tabular format. I wonder if this could be a potential showstopper...

Does KS display an AI generated summary of the knowledge article in the Genesys UI?

Thanks again and have a great day!

Deepa

------------------------------
Deepa Galaiya
Product Owner, Customer Interactions

Original Message:
Sent: 08-28-2025 17:25
From: Brian Jones
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

No problem, Deepa:

• Q: What was your experience with the initial effort required to get started with trigger rules and phrases? 
• A: For us it was minimal given we found sufficient success with the default "No other rule applies > search knowledge base" and adding a "Greeting" intent that automatically surfaced our "Opening Greeting" Canned response for the agent. Here's an example:

• Q: Is there a specific KB article format/structure that performs better?
• A: Great question that I'm sure other community members have various opinions on. We have tried both PORTRAIT and LANDSCAPE layouts. As clean as the table format is, it can be challenging to copy/paste/format those tables from our outside source documents, answer-highlighting doesn't work as well (if at all with tables), and the new multi-panel columnar format for the Agent Workspace scrunches up the table to the point where it's hard for our agents to scroll & read both horizontally and vertically. As such, we stuck with the vertical format given it plays better with the multi-panel config, is easier to format, & answer highlighting works fairly consistently.

• Q: Have you found that there is ongoing manual effort required to finetune responses and phrases?
• A: Anyone that says KB/responses/phrase/copilot management is not ongoing is lying to you, haha. That said, I will say if the KBs are well-written and tuned upfront, any ongoing effort should be minimal & tolerable, if not welcomed.

• Q: Are there any defects you have noticed with this feature? 
• A: The biggest issues for us are the KB formatting as well as the display challenges within the new multi-panel config. Also as mentioned above, if you do have tables in your KBs, the answer highlighting doesn't work as well, if at all, and you cannot see the article content in the surfaced article (only the title). Although not a "defect", if you do choose to use Copilot's Answer Generation feature, have your staff give feedback on its accuracy. We turned Answer Generation off given it proved too risky for our lines of business to rely on [at this time] versus reviewing the highlighted answer or full KB. That's truly a "results & adoption varies by customer" feature.

------------------------------
Brian T. Jones | Ascension | Senior Specialist - Technology

Original Message:
Sent: 08-28-2025 05:20
From: Deepa Galaiya
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Hi Brian,

Thanks so much for sharing your experience and insights with Knowledge Surfacing. That is incredibly helpful and encouraging to hear! Glad to hear that KS has been responsive is continuing to learn! I hope you don't mind a couple of additional questions:

• What was your experience with the initial effort required to get started with trigger rules and phrases? 
• Is there a specific KB article format/structure that performs better?
• Have you found that there is ongoing manual effort required to finetune responses and phrases?
• Are there any defects you have noticed with this feature? 

Thanks again for taking time to respond. Really appreciate it :)

Deepa

------------------------------
Deepa Galaiya
Product Owner, Customer Interactions

Original Message:
Sent: 08-27-2025 23:59
From: Brian Jones
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Greetings @Deepa Galaiya - Our experience has been great thus far with the [Agent Copilot] knowledge surfacing during voice interactions with minimal to no lag time. The only latency we noticed was based off the fact that Copilot seems to wait to do its KB article search once the Caller pauses their speech or stops talking altogether. The surfacing has been so good that it often continues to refine/requery the knowledge results as the conversation continues, which means the agent may have more than one KB article result to consider over time.

That said, it's important to note that the answers to your questions depend heavily on:

• Network Performance | Any latency in the home/office connectivity impacts overall Genesys Cloud performance.
• Well-Written KB Article Content | If the articles are vague, broad, or have any overlap, the system will have a harder time locating the appropriate KB article. Leveraging KB Article Phrases (where necessary) has helped us dial in KB surfacing.
• Clear Communication | I know...this seems like common sense, but unlike with a digital channel, if Copilot cannot understand what the caller is saying (e.g. mumbled words, slang, no relevant keywords, etc), it either will not surface anything or will not surface the right thing.
• Picking a Side Copilot Should Monitor | In our case we programmed our overall Copilot knowledge surfacing rule to only listen to the Caller side so KB articles were only surfaced based off what the Caller said, and didn't query every time the agent said something as well. Prior to that change we noticed the system was pinging & ponging KB article results, especially if the search terms it chose to use varied based on who was speaking.

------------------------------
Brian T. Jones | Ascension | Senior Specialist - Technology

Original Message:
Sent: 08-27-2025 20:38
From: Deepa Galaiya
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Thanks Jason - Do you know if many Genesys customers have implemented Genesys CoPilot Knowledge surfacing for the voice channel? 

------------------------------
Deepa Galaiya
Product Owner, Customer Interactions

Original Message:
Sent: 08-26-2025 11:52
From: Jason Kleitz
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Hello Deepa,

That's a great question. I'm genuinely curious to see what the Community's experience has been with the responsiveness of the platform.

------------------------------
Jason Kleitz
Online Community Manager/Moderator

Original Message:
Sent: 08-26-2025 04:13
From: Deepa Galaiya
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Hi Genesys Community,

Keen to hear from the community on your feedback or experience with Genesys CoPilot for Knowledge Surfacing especially around responsiveness and usability during live voice interactions. Has anyone experienced latency issues with CoPilot's real-time knowledge surfacing during voice interactions? It would be good to understand if there is any noticeable lag between the conversation and when knowledge articles are surfaced, or the way it searches to determine the right moment to trigger a search. This could make it challenging for agents to navigate the surfaced content while actively on a call. Thanks in advance :)

#ConversationalAI(Bots,AgentAssist,etc.)

------------------------------
Deepa Galaiya
Product Owner, Customer Interactions
------------------------------

Deepa - In response to this...

"We were looking into real time analytics providers such as Operative Intelligence, Qualtrics and Call Journey but now I am wondering how much overlap there might be between Genesys features and what other vendors offer noting that the platform features might not be in real time but they might essentially offer the same insights."

...before spending any of your hard-earned time/money on 3rd party products/services, I would strongly recommend you schedule time with your Genesys CS/TAM team to discuss further. In my experience, they do an outstanding job (free of charge) understanding your need, identifying the right application to address the need, and helping you create an action plan accordingly (whether the solution is Genesys or not). That ongoing consultation [should be] included in your contract 😉

------------------------------
Brian T. Jones | Ascension | Senior Specialist - Technology

Original Message:
Sent: 09-09-2025 07:59
From: Deepa Galaiya
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

@George - Good question. We haven't got the API connector yet. We have recently implemented Auto summary.and predictive wrapcodes and are researching what comes next so that we can realise value from our CoPilot tokens. I wonder if Genesys retains the original structure of the KB articles when it ingests the content or if it reformats it to suit the Genesys UI.

------------------------------
Deepa Galaiya
Product Owner, Customer Interactions

Original Message:
Sent: 09-07-2025 08:18
From: George Ganahl
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Do you already have that custom connector working?

I am just wondering whether the format of your original articles changes after synchronization to Genesys Cloud.

------------------------------
George Ganahl GCCX-AI, GCP, GCSME
Technical Adoption Champion
Genesys
2024 Community Member of the Year

Original Message:
Sent: 09-05-2025 20:40
From: Deepa Galaiya
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Hi George,

We have to get an API connected so that Genesys can ingest all of our KB articles from Livepro (our KM sysyem). 

Cheers,

Deepa

------------------------------
Deepa Galaiya
Product Owner, Customer Interactions

Original Message:
Sent: 09-05-2025 13:08
From: George Ganahl
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

How are you getting your articles into a Genesys Cloud Knowledge Base for use with Agent Copilot and Bots?

------------------------------
George Ganahl GCCX-AI, GCP, GCSME
Technical Adoption Champion
Genesys
2024 Community Member of the Year

Original Message:
Sent: 09-04-2025 21:31
From: Deepa Galaiya
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Hi Brian, Apologies for the delayed response. Thanks for sharing your knowledge and experience!

The point about articles that are in table formats is really interesting. We use Live Pro as our Knowledge Management system, and all of our articles are in tabular format. I wonder if this could be a potential showstopper...

Does KS display an AI generated summary of the knowledge article in the Genesys UI?

Thanks again and have a great day!

Deepa

------------------------------
Deepa Galaiya
Product Owner, Customer Interactions

Original Message:
Sent: 08-28-2025 17:25
From: Brian Jones
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

No problem, Deepa:

• Q: What was your experience with the initial effort required to get started with trigger rules and phrases? 
• A: For us it was minimal given we found sufficient success with the default "No other rule applies > search knowledge base" and adding a "Greeting" intent that automatically surfaced our "Opening Greeting" Canned response for the agent. Here's an example:

• Q: Is there a specific KB article format/structure that performs better?
• A: Great question that I'm sure other community members have various opinions on. We have tried both PORTRAIT and LANDSCAPE layouts. As clean as the table format is, it can be challenging to copy/paste/format those tables from our outside source documents, answer-highlighting doesn't work as well (if at all with tables), and the new multi-panel columnar format for the Agent Workspace scrunches up the table to the point where it's hard for our agents to scroll & read both horizontally and vertically. As such, we stuck with the vertical format given it plays better with the multi-panel config, is easier to format, & answer highlighting works fairly consistently.

• Q: Have you found that there is ongoing manual effort required to finetune responses and phrases?
• A: Anyone that says KB/responses/phrase/copilot management is not ongoing is lying to you, haha. That said, I will say if the KBs are well-written and tuned upfront, any ongoing effort should be minimal & tolerable, if not welcomed.

• Q: Are there any defects you have noticed with this feature? 
• A: The biggest issues for us are the KB formatting as well as the display challenges within the new multi-panel config. Also as mentioned above, if you do have tables in your KBs, the answer highlighting doesn't work as well, if at all, and you cannot see the article content in the surfaced article (only the title). Although not a "defect", if you do choose to use Copilot's Answer Generation feature, have your staff give feedback on its accuracy. We turned Answer Generation off given it proved too risky for our lines of business to rely on [at this time] versus reviewing the highlighted answer or full KB. That's truly a "results & adoption varies by customer" feature.

------------------------------
Brian T. Jones | Ascension | Senior Specialist - Technology

Original Message:
Sent: 08-28-2025 05:20
From: Deepa Galaiya
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Hi Brian,

Thanks so much for sharing your experience and insights with Knowledge Surfacing. That is incredibly helpful and encouraging to hear! Glad to hear that KS has been responsive is continuing to learn! I hope you don't mind a couple of additional questions:

• What was your experience with the initial effort required to get started with trigger rules and phrases? 
• Is there a specific KB article format/structure that performs better?
• Have you found that there is ongoing manual effort required to finetune responses and phrases?
• Are there any defects you have noticed with this feature? 

Thanks again for taking time to respond. Really appreciate it :)

Deepa

------------------------------
Deepa Galaiya
Product Owner, Customer Interactions

Original Message:
Sent: 08-27-2025 23:59
From: Brian Jones
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Greetings @Deepa Galaiya - Our experience has been great thus far with the [Agent Copilot] knowledge surfacing during voice interactions with minimal to no lag time. The only latency we noticed was based off the fact that Copilot seems to wait to do its KB article search once the Caller pauses their speech or stops talking altogether. The surfacing has been so good that it often continues to refine/requery the knowledge results as the conversation continues, which means the agent may have more than one KB article result to consider over time.

That said, it's important to note that the answers to your questions depend heavily on:

• Network Performance | Any latency in the home/office connectivity impacts overall Genesys Cloud performance.
• Well-Written KB Article Content | If the articles are vague, broad, or have any overlap, the system will have a harder time locating the appropriate KB article. Leveraging KB Article Phrases (where necessary) has helped us dial in KB surfacing.
• Clear Communication | I know...this seems like common sense, but unlike with a digital channel, if Copilot cannot understand what the caller is saying (e.g. mumbled words, slang, no relevant keywords, etc), it either will not surface anything or will not surface the right thing.
• Picking a Side Copilot Should Monitor | In our case we programmed our overall Copilot knowledge surfacing rule to only listen to the Caller side so KB articles were only surfaced based off what the Caller said, and didn't query every time the agent said something as well. Prior to that change we noticed the system was pinging & ponging KB article results, especially if the search terms it chose to use varied based on who was speaking.

------------------------------
Brian T. Jones | Ascension | Senior Specialist - Technology

Original Message:
Sent: 08-27-2025 20:38
From: Deepa Galaiya
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Thanks Jason - Do you know if many Genesys customers have implemented Genesys CoPilot Knowledge surfacing for the voice channel? 

------------------------------
Deepa Galaiya
Product Owner, Customer Interactions

Original Message:
Sent: 08-26-2025 11:52
From: Jason Kleitz
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Hello Deepa,

That's a great question. I'm genuinely curious to see what the Community's experience has been with the responsiveness of the platform.

------------------------------
Jason Kleitz
Online Community Manager/Moderator

Original Message:
Sent: 08-26-2025 04:13
From: Deepa Galaiya
Subject: Genesys CoPilot Knowledge Surfacing for Voice Channel

Hi Genesys Community,

Keen to hear from the community on your feedback or experience with Genesys CoPilot for Knowledge Surfacing especially around responsiveness and usability during live voice interactions. Has anyone experienced latency issues with CoPilot's real-time knowledge surfacing during voice interactions? It would be good to understand if there is any noticeable lag between the conversation and when knowledge articles are surfaced, or the way it searches to determine the right moment to trigger a search. This could make it challenging for agents to navigate the surfaced content while actively on a call. Thanks in advance :)

#ConversationalAI(Bots,AgentAssist,etc.)

------------------------------
Deepa Galaiya
Product Owner, Customer Interactions
------------------------------