Pre-Built Analytics
Aisera provides out-of-the-box analytics dashboards that are available immediately so customers can derive insights and value as they go live.
After you log into the Aisera Admin UI, you will find the Analytics page below the Home page. Choose Analytics, and then Pre-built Analytics from the left navigation menu.
Pre-Built Analytics provides out-of-the-box dashboards that metrics related to Requests, Sessions, Users, App Performance and many more. Users can select various filters and can even save custom filters as per individual needs.
Types of Dashboards
Each of the tabs on the Pre-Built Analytics page is a dashboard of analytics that are collected or calculated by the Aisera platform.
The dashboard tabs include:
Requests, Sessions, Approval, Users, Knowledge, Tickets, App Performance, Search, Conversational Assist, and Business Metrics.
The information on each of these dashboards is described in the table below.
Dashboard Name
Description
The Requests (Conversation Requests) are defined as a series of messages exchanged between the Users and AiseraGPT application (App).
The Request dashboard captures the insights into the conversation requests that users have with the app.
It focuses on metrics such as requests that were successfully answered (and not answered) by the app, escalated to agents, user drop-offs, intents that helped resolve the request and that did not help, identify knowledge gaps, etc.
A session is considered to have started as soon as a user sends a message and ends when the user indicates that the problem is solved, abandons the chat, or is escalated to an agent.
A session can have multiple conversation requests.
Hence, the number of total requests >= no of sessions.
The Sessions dashboard captures the insights of sessions created by users on the virtual assistant i.e. it displays insights into the conversation initiations by users when interacting with the virtual assistant in a selected time period.
The Users dashboard captures insights on the type of users interacting with the virtual assistance and focuses on metrics such as who are the Top users, New users, Live users, VIP users (only applicable for ServiceNow), their Locations, platform..etc.
These insights help understand user trends and focus on improving user experience thereby improving user retention.
The Knowledge dashboard captures the count of knowledge documents that were used in all conversations with corresponding Click-Through-Rate (CTR).
Note: CTR is taken into consideration for events when/if the users have clicked on the document title or clicked on the “read more” option.
The Tickets dashboard captures the insights of all the tickets that have been ingested from various endpoints into Aisera.
It focuses on the metrics such as Total tickets created, MTTR, priorities, types..etc
The App Performance dashboard captures the overall insights of the VA and is designed from a VA admin’s perspective.
It focuses on helping admins improve overall VA performance and captures matrices such as how many intents are active and helpful, total fulfilments and how many are active & helpful, conversation requests that were understood by the bot vs not understood..etc.
Tenant Configuration - Resolution Rate
One of the primary KPIs providing value to our customers is Resolution Rate. You are required to set up the Tenant Level Configuration, shown below, to capture data that the Aisera platform uses to calculate the Resolution Rate.
You can browse the configuration shown above by clicking on:
Settings > Configuration, and then selecting “Resolution Rate” as shown in the image above.
Non-Flow Resolution considered as:
Use non-flow resolution to use a default fulfillment status of “Resolved” or “Abandoned”.
If set to “Resolved”, the conversation will be considered as “Resolved” if the bot found the relevant Knowledge Article and served as the response. If the user explicitly clicks on the thumbs-down icon after Non-Flow Resolution is set to Resolved, the conversation will be considered as unresolved.
If set to “Abandoned”, the conversation will be considered as “Abandoned” if there is no response from the user. If the user explicitly clicks on the thumbs-up icon, the conversation will be considered resolved. If the user clicks on the thumbs-down icon, the conversation will be unresolved.
Pre-Built Analytics
Aisera provides out-of-the-box analytics dashboards readily available for customers to derive insights and value as they go live.
Navigation:
Admin UI → Analytics → Pre-Built
The Pre-Built Analytics page is divided into x tabs, that represent analytics data categories.
This document will discuss each of the analytics metrics on these tabs.
Requests Tab (Conversation Resolution Analytics)
At the top of the Pre-Built Analytics page, you will see a section of Key Performance Indicators or analytics that are used across various dashboards. This will give you a foundational understanding of the Aisera analytics capabilities.
Total Requests
This metric captures the total number of conversation requests that users had with the chatbot.
Resolution Rate
Percentage of conversation requests that were successfully answered by AiseraGPT.
Note:
A conversation is resolved if the user has clicked yes or if there is no response from the user, such as if the user dropped off (this is configurable).
Answers are displayed only if the confidence score is above the threshold set by the Conversational AI model.
Not Helpful Rate - Thumbs Down (Negative) Feedback
Percentage of conversation requests that did not have appropriate fulfillment. These are the conversations where the user has explicitly clicked thumbs down. Irrespective of the fulfillment, the feedback from the user was negative (thumbs down).
Not helpful requests - Indicates requests where the bot found an intent, and responds back with the solution but the user clicks thumbs down.
Application/bot users can add context by selecting any of the following options as the reason for the Thumbs Down response:
Completely Inaccurate
Missing Relevant Information
Difficult to Follow
Something Else
Unresolved Requests by Type
The analytics for these Feedback options will show up as Unresolved Requests by Type on the Requests tab.

The Resolution Type column allows you to see the feedback comments for unresolved requests (where the bot user has entered a Thumbs Down response), without clicking on each request to see feedback on the Request Details page.
To see the Resolution Type column: 1. Navigate to the Analytics >Pre-Built Analytics >Requests tab. 2. Click on Not Helpful Rate to get to the Unresolved Requests view. 3. Filter by Resolution Type and select Not Helpful - Something Else. 4. In the filtered view, click the ellipsis icon (3 dots) to Add/Remove Columns. 5. Add the Resolution Type column to your view.
Abandoned Rate
Captures the percentage of requests where users did not respond or dropped off.
Escalation Rate
The percentage of total conversation requests that were not answered by the Aisera application or where the user specifically clicked “No” on the feedback option. This indicates the requests where users have given negative feedback to the bot response and then reached out to Support, created a ticket, or gave a live hand-off to the agent.
Escalated requests
Indicates requests where users have given negative feedback to the bot's response and then reached out to support or created a ticket.
Assisted Rate
Captures the requests where users have directly reached out to Support or created a ticket. The number of assisted requests divided by the number of relevant requests (resolved, unresolved, assisted, abandoned).
Resolution Rate Over Time
This metric shows the percentage of conversations between users and the chatbot and that the queries were answered and resolved by the chatbot without being escalated/transferred to Support Agents.
Note:
Answers are displayed only if the confidence score is above the threshold set by the Conversational AI model.
A conservation is said to be resolved if the user has clicked yes or if there is no response from the user, such as if the user dropped off.
It is possible to filter conversations by user feedback (see below).
Average First Resolution Rate Over Time
The Prebuilt Analytics window now includes the Average First Resolution Rate Over Time metric in the Requests tab.

This metric measures the time taken in seconds from when a user initiates an actionable request or query (excluding casual conversations) until the bot provides a first relevant, actionable, correct response.
Escalation Rate over Time
Percentage trend of conversation requests that were not answered by Aisera or where the user specifically clicked “No” on the feedback option for the selected time period.
Requests over Time
Breakdown of Resolved vs Unresolved conversation count over time.
Resolved Requests by Category
Breakdown of Resolved Conversations by Categories with the ability to drill down to sub-categories and intents.
Unresolved Requests by Category
Breakdown of Unresolved Conversations by Categories with the ability to drill down to sub-categories and intents.
Top Resolved Intents
Breakdown of top intents from resolved conversations.
Top Unresolved Intents
Breakdown of top intents from unresolved conversations.
Top Intents by Sentiment
Breakdown of top Intents and user sentiments observed in the conversation with the virtual assistant.
Unresolved Requests by Type
Categorize unresolved conversations based on six distinct types,
Correct Answer
Not Understood
Completely Inaccurate
Missing Information
Hard to Understand
Not Helpful - Something else

Sessions Tab (Conservation Session Analytics)

Total Sessions
Captures the total number of user sessions that took place in a given time period.
The session is considered to have started as soon as a user sends a message and ends when the user indicates that the problem is solved, abandons the chat, or is escalated to an agent.
CSAT/ESAT
Captures the average Customer Satisfaction survey score for a given time period.
Average Daily Sessions
Captures the average number of daily sessions that took place.
Live Sessions
Captures number of live sessions where the virtual assistant is being engaged.
Average Session Duration
Captures time from the minute the user first engaged with the virtual assistant to the minute there was no activity or the session timed out.
Total Request:
Captures the total number of conversations that took place.
Average Daily Requests
Captures the average number of daily conversations that took place.
Average Conversation Duration
Calculates and stores the average time from the beginning of all conversations to the end of all conversations.
Average Session Request
Captures the average number of conversations per session.
Average User Session
The number of conversation sessions that have occurred on average per user.
Sessions Over Time
Captures total count of user sessions over the selected time period.
CSAT/ESAT Over Time
Captures the trend of the CSAT/ESAT for the selected time period.

Requests by CSAT
Captures the CSAT/ESAT score based on the number of conversations/requests.

CSAT/ESAT surveys over Time
Captures the CSAT/ESAT score trend for user sessions over time.

Top Intents
This metric shows the trend of Top Intents that have been discovered by the bot in the conversation between all the users and their interaction with the bot.
Upon hovering the mouse on any of the intents, the graph shows the length of time that this intent was used, as a percentage of a given time period.
When you click on a specific intent, the diagram displays additional details for that intent, such as the associated Fulfillment, Resolution Type, and Date.
Intents Usage and Engagement
Captures and stores the number of intents per user session or request within a selected time period.
Flows Completion
Captures the details of the workflow/fulfillment and shows the number of Requests where the flow was helpful/completed vs where it wasn't.
User Analytics Tab
Total Active Users
Captures the count of total users that have engaged with the virtual assistant for a given time period. “Total Users” has been renamed to “Total Active Users”.
New Users
Captures the count of new users that have engaged with the virtual assistant for a given time period. This metric will only be populated if a user’s details do not already exist within the Aisera application when the user logs in to the bot.
Live Users
Captures the count of live users engaging with the virtual assistant.
VIP Users
Captures the total number of VIP users that have engaged with a virtual assistant for a given time period.
This metric is specific to user data ingested from ServiceNow. This metric gets populated when the VIP users from ServiceNow initiate a conversation with the virtual assistant.
Users Over Time
Captures the trend of users (Total and New users) interacting with the virtual assistant for a selected time period.
Total Returning Users Over Time
The Total Returning Users Over Time metric in Users tab of the Prebuilt Analytics window depicts the number of returning users in a given period of time.

This allows you to determine whether the bot is receiving a lot of repeat visitors.
Top Users
Captures breakdown of top users (users who have engaged the most) with the virtual assistant.
User Locations
Captures the number of users from various locations interacting with the virtual assistant during the selected time period.
Device Type
Captures the user device types that interact with the virtual assistant.
Device Platforms
Captures the user’s platform type that interacts with the virtual assistant.

Web Page Sources
Captures the web page data from users who are in conversation with the virtual assistant.
Browsers
Captures the user's browser type when they are interacting with the virtual assistant.
Knowledge Analytics Tab
Total Knowledge document
Captures the total number of documents that have been crawled/ingested from the Source system for the selected date range
Knowledge Article Utilization Rate
Calculates and stores the percentage of the total number of Knowledge Articles that have been served by the bot at least once for the selected date range.
Average Click-Through Rate
Calculates and stores the average click-through rate based on the click-through rates of the articles that have been clicked by the users after they were served by the bot.
Average Click Through Rate (CTR) =
CTR for KB1 + CTR for KB2 +… CTR for KBN / Total KBs
Top Knowledge Gaps
Breakdown of top knowledge gaps from unresolved conversations.
Top Documents in Requests by Click-Through
Captures the number of Knowledge Documents that were used in all conversations with corresponding Click-Through-Rates (CTR).
Note: CTR is assumed for events when the user has clicked on the document title or clicked on the “read more” option.
Ticket Analytics Tab
Total Tickets
Captures count of total tickets created for a given time period.
Active Tickets
Captures count of tickets created from an active system of record for a given time period. You can also view active ticket data by choosing SOR > Tickets and then selecting the chart you want to view from the pull-down menu in the upper-right corner of the Tickets window.
The Bubble Chart enhancement allows you to view Ticket data as a bubble chart.

The Bar Chart enhancement allows you to view Ticket data as a horizontal bar chart.

MTTR Tab (Mean Time To Resolution)
Captures mean time to resolution of tickets that includes the time from when the ticket was first submitted and recorded in the system of record to the time it was successfully resolved and marked in the system of record.
Tickets over Time
Captures ticket count trend over a period of time.
MTTR Over Time
Captures the MTTR trend over a period of time.
Tickets by Priority Over Time
Captures the number of tickets created in a selected time period, broken down by priorities.
Tickets by Type
Captures the number of tickets created, broken down by Ticket Type.
Tickets by Priority
Captures the number of tickets created, broken down by priorities, as defined in the system of record.

App Performance Tab

Total Intents
This metric displays the total number of active intents in the application that correctly calculate (understand) the user's request. Upon clicking on the metric, you will see the list of all the Intents created in the system.
Total Entities
This metric shows the total number of Entities that have been created in the system that are active. Upon clicking the graphic for this field,, you will see the list of all the Entities created in your Aisera System.
Total Fulfillments
This metric displays the total number of Fulfillments created in the system that are active. Active fulfillments include action flows, knowledge flows, service catalog items, and custom messages in the application.
Total Phrases
This metric shows the total number of phrases that are created in the system and are active.
When you click on the graphic for this metric, you will see the list of all the phrases created in the application.
Intent Utilization
This metric presents the percentage of the total number of intents that have been utilized in all of the Requests. Upon clicking on the graphic for this metric, you will see the list of all the intents that have been utilized/invoked within the application.
Entity Utilization
This metric shows the percentage of the total number of Entities that have been utilized for all user requests to date. If you click on this metric, you will see the list of all the Entities that have been utilized/invoked within the application.
Fulfillment Utilization
This metric captures the percentage of the total action-flows/fulfilments that have been utilized in serving the requests.
Top Intents (Intents that are Used the Most)
This metric shows the trend of Top Intents that have been discovered by the bot in all of the user conversations with the bot.
If you hover the cursor over any of the intent fields, you will see the percentage of the total conversation time that the specific Intent was detected.
If you click on a specific intent, you will see additional details for that intent, such as related Fulfillment, Resolution Type, and Date.
Top Entities
This metric displays the trend of Top Entities that have been discovered in all of the conversations between the users and the bot. Upon hovering the mouse on any of the Entities, you will see details like the name of the Entity and the number of requests it was discovered in. Upon clicking on one of the Entities, you will see additional details like associated Fulfillment, Resolution Type, and Date.
Understood Rate
The percentage of total conversation requests that were understood by the virtual assistant. ‘Understood’ is defined here as requests for which the virtual assistant was able to identify the Intents.
Not Understood Rate
The percentage of the total conversation requests that were not understood by the virtual assistant. ‘Not Understood’ is defined here as requests for which the virtual assistant was not able to identify the Intents.
Avg Confidence
The average confidence that the conversion requests will be fulfilled.
Understood vs Not Understood Requests
This metric shows the number of requests that were understood by the bot and those that weren’t understood for a given time period.
Upon hovering your cursor over this metric,, it shows details such as Request Count, Date, and Type.
Upon clicking the widget, you will see the list of all the Requests that were Understood and Not Understood by the bot.
Total Knowledge Articles
This metric shows the total number of knowledge articles that were used in intent active fulfillments.
Total Knowledge Flows
This graph shows the number of knowledge flows used in intent active fulfillments.
Total Action Flows
This metric shows the number of action flows used in intents active fulfillments. Total Service Catalog
This metric displays the number of service catalog items used in intents active fulfillments.
Total Custom Messages
This metric shows the number of custom messages used in intents active fulfillments.
Knowledge Article Utilization
This metric gives you the percentage of the total number of Knowledge Articles utilized in the requests.
Knowledge Flow Utilization
This metric shows the number of knowledge flows utilized in the requests.
Action Flow Utilization
This metric shows the number of action flows utilized in the requests.
Service Catalog Utilization
This metric shows the number of service catalog items utilized in the requests.
Custom Message Utilization
This metric shows the number of custom messages utilized in the requests.
Fulfillments Breakdown
This pie chart displays the count as well as the percentage of the fulfillment breakdown that was either Supervised or Unsupervised.

Request Fulfillments by Source
This pie chart shows the percentage of the total fulfillments per source.
Top Fulfillments (Fulfillments Used the Most)
This pie chart displays the fulfillments that are used most often in conversation requests.
Business Metrics Tab
User Productivity Savings / Total Time Saved for Employees
Defines how the bot saves productive working time for employees who use the bot.
Each issue resolved is assumed to save employees 60 minutes on average.
Total Employee Productivity Gained = Number of issues resolved * 60 minutes/issue resolved
Note -This metric is 60 minutes by default but can be re-configured in the AISX UI
Agent Impact
Measures the amount of time saved or gained by Agents when their workload is decreased due to workload handled by the bot.
Illustrates how the bot eliminates work for the service desk and does the equivalent work of full-time agents.
Helps visualize how issues accelerated are trending over time (for the selected time period)
Avoided Cost
Defines avoided support costs for the organization calculated by multiplying Agent Cost by Agent Impact.
Average amount of time to resolve an issue (minutes)
Agent Capacity = the number of issues resolved per day
Cost per issue
The above business metrics have the following AiseraGPT Configuration:
For metrics about workflows, see Workflow Analytics.
Last updated