Engage Digital | Quality value reports
Quality value reports allow you to get information on the quality of agent responses and service level in your contact center. There are a variety of reports that represent the most common data that you want to analyze. You can find quality value reports by navigating to Analytics > Quality value and selecting a specific report.
Business hours
You may want to view reports that take in account your company’s hours of operation, or business hours. When you switch to business hours, the response and conversation times are only taken into account between the opening and closing hours that you defined in Admin > Settings > Business hours.
Suppose that your business hours are from 9am to 6pm. If a customer messages you at 8am and an agent replies to them at 9:30am, the first response time for the conversation will be 1h30min outside business hours. Switching to business hours only, the first response time goes from 1h30min to 30min only (9h30min - 9h = 30min).
If you don’t want to take into account the time elapsed outside business hours, you can enable business hours by using the switch button that is available for the following reports: Average first response time, Average response time, and Service level.
First reply resolution report
This report gives, in absolute value, the number of replies needed before agents can close a conversation, over the selected period. The table shows, per agent, the number of replies needed before agents can close a conversation, over the selected period.
This report allows you to compare the proportion of conversations that, at the time of the consultation, are closed after receiving a single agent response compared to the proportion of conversations that required multiple agent responses. The higher the rate of closed conversations in a single agent response, the better the quality of your service: agents are able to resolve client conversations in a single response and do not need to rephrase or reformulate several times.
You can filter by disposition categories to identify which contact the reasons for customer requests require longer conversations than others.
This report is especially useful for email channels, as email is a private channel where more time can be spent by the agent to compose messages and provide more information to a customer. However, this report is hardly useful for social networks, and not useful at all for chat channels. Indeed, solving an issue on those last two channels often requires more than a single agent reply.
Interpreting the results: identifying how many conversations are closed with a single reply
For example, a team solves customer inquiries with only one reply in more than half the conversations. Out of the 5 conversations that were created over the period selected, 3 have been closed after only 1 reply — and are still closed at the time of the consultation.
Client satisfaction report
The Client satisfaction report shows you the ratings from client satisfaction surveys configured in the system. You can create client satisfaction surveys in SurveyGizmo, and configure those surveys to be sent to customers when a conversation is closed. To configure surveys, you navigate to Admin > Digital > Surveys.
The report shows the client satisfaction ratings as selected by a customer in the SurveyGizmo survey the customer receives after their conversation is closed. Ratings are shown for each agent based on that agent’s closed conversations, as well as the total ratings for surveys for all agents with closed conversations.
If an agent has low client satisfaction ratings, you should verify the agent is not closing conversations too early.
First contact resolution report
The First contact resolution report shows you the absolute number of closed conversations with or without reopening. The table shows, per agent, the number of closed conversations with or without reopening over the selected period.
This report allows you to compare the proportion of conversations that, at the time of the consultation, are closed without reopening compared to those that have been closed and reopened. The higher the rate of closed conversations without reopening, the better the quality of your service Agents are able to resolve customer inquiries durably and customers do not need to write again.
This report also allows you to diagnose agents’ difficulties in working through messages. For example, if an agent has a large amount of closed operations with reopenings, they probably close their conversations too early, that is, before the customer’s inquiry is solved. Also, if you have enabled customer satisfaction surveys, your agents must avoid closing conversations too early, as a survey is sent to the customer once the agent closes the conversation.
Interpreting the results: examining the rate of closed conversations without reopening
For example, suppose that your data shows that of the 5 conversations that were created over the period selected and handled by agents, 4 were closed by agents, and then reopened one or several times after a new message was received. This shows that conversations are being closed too early before the conversation is solved.
Rate of contact report
The Rate of contact report gives, in absolute value, the number of discussion threads without conversations as well as threads with at least one conversation. The table shows the number of children without conversations, with at least one conversation, over the selected period, the previous period, as well as the evolution between the two.
The rate of contact is the ratio of the number of threads with at least one conversation over the total number of threads. A discussion thread with no conversation is a thread in which no agent has engaged messages, meaning that no conversation was opened. The messages from a thread with no conversation therefore have a status of either New or Archived.
The rate of contact report is most appropriate for use with chat channels, as a discussion thread for a chat channel can only contain a single conversation associated with a client, whereas on social media channels, a discussion thread oftens leads to several conversations (one per client). As a result, not replying to a customer has a stronger negative impact on the rate of contact on chat channels than on social media channels. For social channels, we recommend that you calculate the engagement rate from the Messages by status report.
Interpreting results: viewing the abandonment and engagement rates
For example, suppose that over a selected period, your report shows you have an engagement rate of 63% and an abandonment rate of 38% on an Engage Chat channel. The engagement rate of 63% represents threads with at least one conversation, while the abandonment rate of 38% reflects threads where no conversation was opened by an agent.
Average queue time report
The Average queue time report gives the average queue time for the selected period from a historical perspective. This time is calculated based on the interactions that are accepted by agents, and establishes the time difference between interactions that appeared in the queue (created or transferred) and interactions accepted by an agent.
As the report is calculated based on the interactions accepted by agents, it will show you the historical view of how long interactions have been in the queue before being accepted by an agent. The report won't show you the state of your current queue.
The table shows the queue time for the interactions associated with the channel group in the system, such as ‘Async’ or ‘Realtime’. Data is reported for the current and previous periods, and relates to the selected period of time on the top of the page. The growth is the difference between the values in the current and previous periods. As the growth of queue time is a red flag of the agents' performance, it is marked with red color.
Interpreting results: viewing the abandonment and engagement rates
For example, suppose that over a selected period, the average queue time is 4mn55s. For the channel ‘Async’ the previous period the queue time is 2mn32s but there are no accepted interactions for the current period. For this reason growth has not been calculated yet. For the channel ‘Realtime’ there is a growth of the queue time, equal to 41.25%, in comparison with the previous period, and this increase is highlighted in red.
Average first response time report
The Average first response time report shows you the average response time for the selected period. This time corresponds to the time between the first client message and the first agent response of the conversation. The response times are calculated with respect to when the customer message is received and when the agent replies — not when the agent reads the message for the first time and the time they reply to it.
On chat channels, the average response time corresponds to the average customer wait time to chat with an agent.
If the average response time is far from your target, possible causes are the team is understaffed, agents have replied to old messages, or agents have replied to a new message from a thread whose old messages had been archived. For the third cause, It is always the time elapsed between the first customer message of the thread and the first agent reply that is taken into account.
If the average first response time differs significantly when switching to business hours, you may want to consider extending your opening hours so as to decrease the average response time.
Interpreting the results
Over the selected period, the average response time is 46mn32s. The first agent in the table has an average response time of 12h32mn and sent 4 first responses.
Average response time report
The Average response time report shows you the average response time over the selected period between each customer message and each agent reply within a conversation. The detail shows the average time per agent as well as the number of replies per conversation created over the selected period.
The response time is the time that separates a customer message from an agent reply. For each conversation, the system determines an average response time. The average response time displayed in this report is the average of the average response times for all conversations that have been created over the selected period.
Response times are associated with conversations and to the agent in charge of a conversation. Messages that are sent to the customer with no conversation (automated messages, bot messages, or messages sent directly from the channel, for instance) are not taken into account for the calculation of the average response time. If several agents answer a customer, the last agent who handled the conversation inherits all the response times.
If the average response time is far from your target, the possible causes are the team is understaffed, agents have replied to old messages or agents have replied to a new message from a thread whose old messages had been ignored. For the third cause, it is always the time elapsed between the first customer message of the thread and the first agent reply that is taken into account.
Interpreting results: comparing agents’ average response times
You can see that for conversations opened today, agents are replying within 76mn36s on average. Filtering the agents’ results by descending order allows you to identify the agents with the highest average response times. For the first two agents, their average response times are very different from the average, which is due to the fact that, on the current day, they answered old messages from previous days.
Average handling time (Routing mode only) report
The Average handling time report shows — per agent — the time spent on an ongoing interaction, with or without a conversation, as well as the total time spent on ongoing digital interactions. The overall figure is an average of the time spent per conversation and per agent.
The data provided by this report makes it possible to determine the average time an agent spends responding to a client (‘with conversations’) and the time spent on interactions you’re not replying to (‘without conversation’). For more details, see the Conversation times section below.
If an agent switches from one interaction to another (from interaction 1 to interaction 2), the count stops for interaction 1 and starts for interaction 2. This is the most reliable way of measuring productivity. Note that if an interaction was transferred or if several agents replied, the time spent by each agent who worked on the interaction will be counted.
To get the average handling time per conversation, you can generate an export of conversations, adding the field handling_time. The global time on the page is an average per conversation and per agent. If transfers between agents are many, the total conversation time is longer.
The table contains the following data:
- Average: The average time spent on interactions with conversations (total with conversation divided by the conversations count).
- Total with conversations: The total time spent on the interactions to which a reply was sent. This is visible only if you filter the report per conversation.
- Conversations count: The number of conversations handled by the agent over the selected period. Note that several agents may have worked on the same conversation. The total time may therefore be higher than the conversations count over the period. This is visible only if you filter the report per conversation.
- Total without conversation: The time spent on interactions without a conversation (archiving, categorizing, transferring). This is visible only if you filter the report per conversation.
- Total: The total with conversations + total without conversations.
Conversation times
As soon as you switch from one interaction to another, or close an interaction, the tool will check whether the interaction has a conversation or not, and allocate the time either in the ‘with conversations’ or ‘without conversation’ timers.
Here are some examples:
- Without conversation: You accept the interaction, read the message, categorize and close the interaction. You spent 30 seconds on these actions. By the time you close the interaction, there has been no open conversation (you have not responded to the client), so we assign 30 seconds to the without conversation time
- With conversation: You accept the interaction, read the message, reply to the customer, and close the interaction. You spent 5 minutes on these actions. By the time you close the interaction, a conversation has been opened (you have responded to the client), so we assign 5 minutes to the with conversation time
Mixing ‘without conversation’ / ‘with conversation’
You accept interaction 1, read the message for 45 seconds. You switch from interaction 1 to interaction 2. As you exit interaction 1, the timer for this interaction stops. The tool checks whether there is an open conversation or not and allocates the time accordingly. Here you did not answer the client (no conversation) so the 45 seconds spent on the interaction will go into the ‘without conversation’ timer.
You return to interaction 1 to answer, and spend 3 minutes on this interaction. By the time you close the interaction, a conversation has been opened (you have sent a reply to the client), so we assign 3 minutes in ‘with conversations.’
To sum up, there will be 45 seconds ‘without conversation’ and 3 minutes ‘with conversation’ on this interaction.
Please note: This behavior only occurs if you switch interactions during processing. Otherwise, case 1 or 2 applies.
An ongoing interaction is an interaction that is selected, and whose content is visible on screen. As long as an interaction is selected in the Routing mode, it remains active, even if the agent is working in another window.
Interpreting results: tracking agents’ activity and performance
First, you can have a look at the average per agent to find out which agents spend more time (on average) on active interactions.
Be careful when comparing agent data: a level 2 agent will spend more time on a conversation than a level 1 agent dispatching interactions by categorizing them. The majority of the handling time for a level 1 agent will therefore go to his total without conversation, while the majority of the handling time for a level 2 agent will go to his total with conversations.
It is also necessary to check the conversations count per agent. Indeed, an agent can have a very low average (and therefore very good) but a small number of conversations.
Measuring an agent’s actual working time
For the second agent in the list, his actual working time is 38m19s. This represents the total time spent categorizing the interaction, answering it, transferring it to another agent, as well as the time spent thinking about a reply (whether a conversation is open or not).
One contact resolution
The One contact resolution report shows the quality of resolution of digital interactions, also known as interactions, by agent, as measured by the ratio of the number of interactions closed without a customer contacting your business again through the same channel before a threshold, to the total number of all closed interactions for the same customer in the same channel. The One contact resolution threshold is defined in Settings > Account settings. By default it is 3 days.
The report also shows the average value of the resolution rate for all or selected agents for the current period.
The table contains the following data:
- The names of the agents who processed interactions in the selected period
- Previous period: The ratio of conversations closed without a customer contacting your business again through the same channel, to the total number of all closed conversations for the same customer in the same channel, for the previous period
- Current period: The ratio of interactions closed without a customer contacting your business again through the same channel, to the total number of all closed interactions for the same customer in the same channel, for the current period
- Growth: The percentage change between the previous and current periods
Interpreting results: evaluating operational results
You can use the One contact resolution report along with the Average handling time report to see the speed and quality of how agents handle interactions. While the Average handling time report shows you how long a unit of work takes, the One contact resolution report shows whether this unit of work resolved a customer's problem.
If an agent closes interactions rapidly and customers end up contacting you again for the same issue, the Average handling time will be low, but the One contact resolution ratio will be poor. If an agent spends a long time on interactions, the One contact resolution ratio will be good, but the Average Handling time will be high.
For these reasons, you can use the Average handling time report to help control the duration of
the agent’s interaction with the customer at an expected level. You can use the One contact resolution report to check the quality of the issue resolution. The expected One contact resolution ratio is between 0.75 and 1.
For example:
- Agent A has 10 closed interactions where 9 of them are successful. The ratio for Agent A is 9/10
- Agent B has 5 closed interactions where 2 of them are successful. The ratio for Agent B = 2/5
- The indicator above the table will show ((9/10)+(2/5))/2 = 0.65
Comparison to First contact resolution report
The following are the differences between the One contact resolution and First contact resolution reports:
| First contact resolution | One contact resolution |
| Customer metric | Operational metric
|
| Conversation level | Interaction level |
| Measures the reopening of conversations | Measures the opening of new interactions |
Service level report
The Service level report shows in absolute terms the distribution of service level thresholds depending on the event type (conversation opened, first reply, conversation closed). You can customize your Service Level thresholds by going to Admin > Settings > Account Settings. In the Account settings configuration page, you can find Service Level thresholds under Analytics configuration.
The table contains the following data:
- Opened: Time elapsed between the first customer message and the agent’s click on Engage
- First reply: Time elapsed between the first customer message and the first agent reply
- Closed: Time elapsed between the first customer message and the last agent reply (if the conversation was reopened, only the last closure is taken into account)
The table allows you to have a view per agent.
Interpreting results: making sure that your service’s SLA is met
This report allows you to have an accurate view of the distribution of conversations depending on the service level thresholds you have set in Admin.
© 1999-2022 RingCentral, Inc. Todos os direitos reservados.