I wanted to share my experiment querying data with AI, that I shared on LinkedIn this week. I'd love to get some feedback from the community on whether this would be useful and if so how I could improve it.

The experiment uses Claude Desktop and my Genesys Cloud MCP Server to allow me to ask questions of data in the Genesys Cloud org, like:

• Find me a conversation from queue X last week with the lowest sentiment score. Then tell me why the sentiment score was so low and how the agent could have prevented this.
• List the topics discussed on queue X yesterday
• Create a report of contact on queue X over the past month

This is what it looks like:

Do people think this would be useful? What type of data would you most likely want to ask AI about, or make adhoc reports on?

#AnalyticsAPI

------------------------------
Lucas Woodward
OVO Energy Ltd

https://www.linkedin.com/in/lucas-woodward-the-dev
------------------------------

@Lucas - Great application of AI tools outside Genesys to mine the data inside Genesys. Hopefully Genesys is watching this thread because [given Genesys also uses Claude among other AI engines], they could logically add a feature to the new AI Studio that allows admins/users to ask Genesys AI (name TBD, haha) to run these same queries of itself [about our respective orgs]. For example, imagine asking Genesys AI to:

• (To the points in your example) Create a queue/skill/agent/topic/survey/evaluation performance report by simply providing it filter and output parameters in the prompt.
• Summarize the performance of X bot, including the top X intents & utterances, identify gaps, and then provide a list of improvement opportunities.
• Create a table that breaks our token usage down by feature, division (or queue, reports to, whatever makes sense), and use that data to provide a monthly trend report that estimates our token consumption over the next year.
• Summarize the AI Insights for the last X days of interactions, and group the results by queue (or program or whatever makes sense) so we can see a trend of Call Reasons, Resolutions, & Action Required.
• Create an Executive Summary of the last 30 days of the Operational Console, and list improvement opportunities it found based on trends.
• So many more examples I could add...

Yes, these are broad interpretations/queries, but that's intentional given the beholder could get as granular as they want within each one thanks to the Genesys data being easily/directly accessible. We're not asking Genesys AI to reinvent the wheel. Instead we're asking it to help us use the same ingredients with its advanced technology to more efficiently/effectively make the wheel faster, better, stronger, or even eliminate the need for it all together given it can help us fly!

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

Original Message:
Sent: 07-04-2025 18:06
From: Lucas Woodward
Subject: Experiment querying data with AI

I wanted to share my experiment querying data with AI, that I shared on LinkedIn this week. I'd love to get some feedback from the community on whether this would be useful and if so how I could improve it.

The experiment uses Claude Desktop and my Genesys Cloud MCP Server to allow me to ask questions of data in the Genesys Cloud org, like:

• Find me a conversation from queue X last week with the lowest sentiment score. Then tell me why the sentiment score was so low and how the agent could have prevented this.
• List the topics discussed on queue X yesterday
• Create a report of contact on queue X over the past month

This is what it looks like:

Do people think this would be useful? What type of data would you most likely want to ask AI about, or make adhoc reports on?

#AnalyticsAPI

------------------------------
Lucas Woodward
OVO Energy Ltd

https://www.linkedin.com/in/lucas-woodward-the-dev
------------------------------

@Lucas - Great application of AI tools outside Genesys to mine the data inside Genesys. Hopefully Genesys is watching this thread because [given Genesys also uses Claude among other AI engines], they could logically add a feature to the new AI Studio that allows admins/users to ask Genesys AI (name TBD, haha) to run these same queries of itself [about our respective orgs]. For example, imagine asking Genesys AI to:

• (To the points in your example) Create a queue/skill/agent/topic/survey/evaluation performance report by simply providing it filter and output parameters in the prompt.
• Summarize the performance of X bot, including the top X intents & utterances, identify gaps, and then provide a list of improvement opportunities.
• Create a table that breaks our token usage down by feature, division (or queue, reports to, whatever makes sense), and use that data to provide a monthly trend report that estimates our token consumption over the next year.
• Summarize the AI Insights for the last X days of interactions, and group the results by queue (or program or whatever makes sense) so we can see a trend of Call Reasons, Resolutions, & Action Required.
• Create an Executive Summary of the last 30 days of the Operational Console, and list improvement opportunities it found based on trends.
• So many more examples I could add...

Yes, these are broad interpretations/queries, but that's intentional given the beholder could get as granular as they want within each one thanks to the Genesys data being easily/directly accessible. We're not asking Genesys AI to reinvent the wheel. Instead we're asking it to help us use the same ingredients with its advanced technology to more efficiently/effectively make the wheel faster, better, stronger, or even eliminate the need for it all together given it can help us fly!

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

Original Message:
Sent: 07-04-2025 18:06
From: Lucas Woodward
Subject: Experiment querying data with AI

I wanted to share my experiment querying data with AI, that I shared on LinkedIn this week. I'd love to get some feedback from the community on whether this would be useful and if so how I could improve it.

The experiment uses Claude Desktop and my Genesys Cloud MCP Server to allow me to ask questions of data in the Genesys Cloud org, like:

• Find me a conversation from queue X last week with the lowest sentiment score. Then tell me why the sentiment score was so low and how the agent could have prevented this.
• List the topics discussed on queue X yesterday
• Create a report of contact on queue X over the past month

This is what it looks like:

Do people think this would be useful? What type of data would you most likely want to ask AI about, or make adhoc reports on?

#AnalyticsAPI

------------------------------
Lucas Woodward
OVO Energy Ltd

https://www.linkedin.com/in/lucas-woodward-the-dev
------------------------------

Happy to contribute. I sent you a connect request via LinkedIn, and will let you know if others come to mind.

@Lucas - Great application of AI tools outside Genesys to mine the data inside Genesys. Hopefully Genesys is watching this thread because [given Genesys also uses Claude among other AI engines], they could logically add a feature to the new AI Studio that allows admins/users to ask Genesys AI (name TBD, haha) to run these same queries of itself [about our respective orgs]. For example, imagine asking Genesys AI to:

• (To the points in your example) Create a queue/skill/agent/topic/survey/evaluation performance report by simply providing it filter and output parameters in the prompt.
• Summarize the performance of X bot, including the top X intents & utterances, identify gaps, and then provide a list of improvement opportunities.
• Create a table that breaks our token usage down by feature, division (or queue, reports to, whatever makes sense), and use that data to provide a monthly trend report that estimates our token consumption over the next year.
• Summarize the AI Insights for the last X days of interactions, and group the results by queue (or program or whatever makes sense) so we can see a trend of Call Reasons, Resolutions, & Action Required.
• Create an Executive Summary of the last 30 days of the Operational Console, and list improvement opportunities it found based on trends.
• So many more examples I could add...

Yes, these are broad interpretations/queries, but that's intentional given the beholder could get as granular as they want within each one thanks to the Genesys data being easily/directly accessible. We're not asking Genesys AI to reinvent the wheel. Instead we're asking it to help us use the same ingredients with its advanced technology to more efficiently/effectively make the wheel faster, better, stronger, or even eliminate the need for it all together given it can help us fly!

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

Original Message:
Sent: 07-04-2025 18:06
From: Lucas Woodward
Subject: Experiment querying data with AI

I wanted to share my experiment querying data with AI, that I shared on LinkedIn this week. I'd love to get some feedback from the community on whether this would be useful and if so how I could improve it.

The experiment uses Claude Desktop and my Genesys Cloud MCP Server to allow me to ask questions of data in the Genesys Cloud org, like:

• Find me a conversation from queue X last week with the lowest sentiment score. Then tell me why the sentiment score was so low and how the agent could have prevented this.
• List the topics discussed on queue X yesterday
• Create a report of contact on queue X over the past month

This is what it looks like:

Do people think this would be useful? What type of data would you most likely want to ask AI about, or make adhoc reports on?

#AnalyticsAPI

------------------------------
Lucas Woodward
OVO Energy Ltd

https://www.linkedin.com/in/lucas-woodward-the-dev
------------------------------