The Teams performance report allows you to analyze the team's performance within a specific service group. It consolidates individual and collective data, helping managers identify bottlenecks, response patterns, workload, and agent efficiency.

To access the report:

1. Access the Analytics menu.
2. Select Sales > Teams performance.
3. Apply the desired filters. You can filter by the following criteria:
   1. Group: Defines which service team will be analyzed.
   2. Period: Determines the time interval to be considered (e.g., yesterday, last 7 days, custom period).
   3. Users: Allows viewing only specific agents or the entire group simultaneously.

Based on these filters, the system automatically calculates:

• Group First Response Time
• Group Response Time
• Individual Agent Performance Table
• Heatmap - Response Time by Hour

How the calculations work

Below, see how the report calculates the two main metrics displayed at the top: First Response Time and Response Time.

These values always represent the median of individual times recorded in the conversations.

First Response Time

It is the time your team takes to respond to a new contact for the first time - it only considers new conversations opened in the period.

Calculation:
For each new conversation created in the period, the system measures:

conversation creation → agent's first response
It gathers all the times from the new conversations and calculates the median.

Important rules:

• Respects the agent's working hours. For example, if a client sends a message on Sunday night, the first response time will start counting on Monday morning or at the beginning of the agent's configured shift.
• Does not consider quick responses.
• Does not consider responses coming from the bot flow.

Response Time

It is the time your team takes to respond during an ongoing conversation - considering all interactions between client and agent, and not just the first response.

Calculation:
For each interaction where the client sends a message and, afterward, the agent responds, the system records this interval:

client message → agent response
It gathers all these times and calculates the median. 

Important rules:

• Does not pause outside business hours. The calculation considers the complete interval between the client's message and the agent's response, even if part of that time occurs outside the shift.
• Ignores automatic responses.
• Considers all interactions, not just the first one.

Team performance Table (by user)

The table summarizes the behavior of each consultant/agent within the group and the filtered period. It presents volume and productivity metrics.

Total contacts

Quantity of contacts/clients, counting any action taken by the consultant/group.

Total conversations

Total number of conversations in which the user participated - including new conversations, ongoing conversations, and reopened conversations.

New conversations

Quantity of conversations started or assigned to the consultant/agent for the first time in the filtered period.

Closed conversations

Number of conversations closed by the consultant/agent in the period. If the agent did not close conversations, the value may appear as 0 - common when the service is still ongoing.

Average time to close (Closes in)

Average time between the start of the conversation and the closure done by the agent.

• If the agent did not close conversations, the value appears as 0s.
• When the time is very high (e.g., 2d 22h), it means that some conversations remained open for long periods before closure.

Average user response time (Responds in)

It is the average time this agent takes to respond, considering all messages they sent in the filtered conversations.

This calculation takes into account only the working hours of the consultant or the group they belong to - meaning periods outside business hours are not included.

Messages sent

Total quantity of messages sent by the agent in the period.

Heatmap - Response time per hour

The heatmap shows the distribution of response times throughout the day, hour by hour, considering the filtered period.

It helps you identify:

• Times with best performance, where responses are fastest (lighter colors).
• Critical times, in which response time increases significantly (darker colors).
• Demand peaks, when many contacts are entering simultaneously.
• Situations outside business hours, which usually raise response times or generate long periods without a response.

How to interpret:

• Each cell represents one hour of the day.

• The color indicates the average response time for messages sent during that hour.

Understand it in practice

Filters: Group Line 1 • Period: Yesterday • Users: All

Group Performance Table:

How did the system arrive at the group results?

✔️ First Response Time = 2m16s
The system analyzed  all new conversations opened by the 5 agents (7 in total). For each conversation, it calculated the interval between the conversation creation and the agent's first response. Then, it gathered all these times and extracted the median - resulting in 2m16s.

✔️ Response Time = 1m30s
In this case, the system gathered all response intervals from all conversations (client message → agent response). From this complete list, it calculated the median again, arriving at 1m30s.

Why don't these top values appear in the table?

Because the numbers at the top are calculated with different methods and with broader datasets than those presented in the table per agent.

In the example above:

• The top shows 1m30s, which is the median of all response times for the entire group.
• The table, on the other hand, shows, per agent, the column Responds in, which is the individual average for each agent.

That is why one agent appears with 43s, another with 13m51s, another with 3h02m, and another with 2d22h. These averages reflect individual performances, influenced by volume, pace, and conversation distribution.

The overall median, in turn, is less impacted by very long or very short times and, therefore, remains at 1m30s, even when some agents have longer intervals throughout the day.