Callback strategy
Explore the various settings for callback strategies and learn how to optimize them for your specific needs.
First Party Called
The First Party Called setting determines whether the system will call the customer or the agent before getting the other party on the line. When set to Customer, the system will call the customer first and wait to dial the agent until the customer is on the line. When set to Agent, the system dials an agent first and has the agent confirm they are ready to accept the callback before calling the customer.
Each option introduces unique advantages and disadvantages, and choosing the best option will depend on the specific needs and resources of the team of agents connected to a particular Call Target. For example, if you choose Agent, the customer experience can improve since an agent will be ready and waiting the moment the customer answers the callback. However, this can keep your agents on the phone for a longer time in each interaction.
Callback Registration Experience
The following diagrams describe the default voice experience for an Agent First and Customer First callback.
Customer First Voice Experience
Agent First Voice Experience
Rescheduling Options
When a customer opts to reschedule their callback, the ECBT options are announced by saying "For 10 minutes from now, press 1. For 20 minutes from now, press 2…", up to 9 for 90 minutes. All ECBT times are rounded up. The reschedule option is only available if the Offer Rescheduled Callback setting is enabled.
Example
If the current wait time is 34 minutes, callers will hear options to choose 4 for 40 minutes, 5 for 50 minutes, and so on, up to 9 for 90 minutes. Callers would not hear the options to choose 1, 2, or 3 since those times would be less than the current wait time.
Minimum Callback Delay
Minimum Callback Delay sets a minimum amount of time that must pass before a callback can be launched. You can use the Minimum Callback Delay to prevent the system from attempting to launch callbacks too quickly after they are registered.
Under normal circumstances, the current ECBT is not immediate, due to the agent handle time and the number of callbacks currently waiting in the queue. However, if you find that callbacks are being dialed too quickly due to a low ECBT or call volume, you might consider increasing the value of this setting to increase the time between callbacks.
- The default value for Minimum Callback Delay is 10 seconds.
- If you use Confirmation notifications, set the value to at least two minutes (00:02:00). If the value is less than two minutes, Confirmation notifications may not be sent.
- The countdown for Minimum Callback Delay begins when the Call Target registers an incoming call. To identify the exact time that the countdown begins, look for the Registering:CallIncoming event in the Call Detail report.
Max Active Calls
The Max Active Calls setting allows you to limit the number of callbacks that can be active simultaneously. A callback is considered active the moment that an agent is engaged. This means that the agent is on the line and ready to speak with the customer. When one of the two parties disconnects, that callback is no longer considered active.
- To identify the exact moment that a callback becomes active, look for the Connecting:AgentAnswered event in a Call Detail report.
- To identify the moment that a callback is removed from the Max Active Calls count, look for Talking:PartiesTalking:AgentHangup or Talking:PartiesTalking:CustomerHangup.
If a pending callback is ready to be launched when the Max Active Calls limit is reached, the callback will be added to the waitlist to be dialed as soon as possible. This can cause callbacks to slow down and increase the ECBT, but it can be useful for smaller teams who may have fewer agents than necessary to keep up with the demand for callbacks.
By default, the value is set to unlimited. This means there is no limit to the number of calls that can be active at once.
- We recommend leaving Max Active Calls at the default value of unlimited under normal circumstances. This setting is meant to address trunking capacity limitations at the call center.
- You might consider using Max Active Calls for multi-skilled agent groups, or groups of less than 15 agents, to maintain your desired pacing for holding calls.
- To set an initial value for Max Active Calls, we recommend dividing the number of Call Targets in your Organization by the number of agents servicing callbacks. From there, monitor the results and make small changes as necessary to achieve your desired call pacing.
Max Queue Depth and Max Active Calls can impact the dialing pacing of callbacks.
When a callback cannot be dialed due to the current queue conditions (based on the Max Queue Depth and/or Max Active Calls settings), it is added to an internal waitlist. Two separate waitlists are used, with different levels of priority, based on the callback type.
ASAP callbacks are added to their own waitlist, while scheduled callbacks and retries are combined into a second waitlist with a higher priority.
Setting different priorities for different types of callbacks helps to ensure that customer expectations are met as closely as possible.
- Customers expect scheduled callbacks at a specific time, and they may have received a Reminder Notification to reinforce the scheduled time, so it is important to dial those callbacks as close to the scheduled time as possible.
- Retries are callbacks that were intended to be answered already, and it's a best practice to dial retries at a regular interval.
- For ASAP callbacks, customers may have been quoted the Estimated Callback Time (ECBT) with a +/- 20% buffer, or they may not have been quoted an estimate at all. Punctuality is still essential, but less important than scheduled callbacks or retries.
Max Queue Depth
The Max Queue Depth setting applies when Wait for Live Agent is enabled and the Customer First Callback Strategy is being used.
Max Queue Depth represents the number of simultaneous callbacks awaiting an agent that can be placed into the priority queue at once. When the maximum number is reached, no additional callbacks will be placed for the associated Call Target until the number decreases. Any additional callbacks due to be processed will be added to the internal waitlist until the number of currently queued callbacks decreases below the Max Queue Depth value.
By default, the value is set to unlimited. This means that no limit will be applied to the number of callbacks that can be sent to the contact center.
If you need to make adjustments to Max Queue Depth, we recommend only making small changes and monitoring the impact over time before adjusting it again.
- After making a change, analyze the impact on the average wait time for the holding queue to determine whether you may need to adjust the Max Queue Depth value further.
- If the average wait time is under 30 seconds, no adjustment is needed. If you feel that your agents can handle more traffic, you might consider increasing the value.
- If the average wait time is greater than 30 seconds, consider lowering the value of Max Queue Depth.
- If Max Queue Depth is set to 1 (the lowest possible value) and your customers still experience long wait times, then there is likely an issue with managing traffic at the call center.
Caller ID (CID) and Use Custom Audio
These options determine the Caller ID number that will be displayed and/or spoken to customers in various scenarios.
- Caller ID (CID): Enter the number you would like to be displayed on customer's Caller ID. This value can also be used in text message notifications to remind a customer of the number to expect.
- Federal and some state laws prohibit the broadcasting of a phone number that does not accept incoming calls.
- A valid phone number must be used for the Caller ID or any custom CID audio.
A maximum of 1,024 characters can be entered into the Caller ID (CID) field.
- Use Custom CID Audio: Select this checkbox to use a custom audio file to read off the expected Caller ID when confirming a callback has been registered. By default, Mindful uses a text-to-speech engine to read the number to customers.
- When selected, remember to upload a custom audio prompt in the Media Set assigned to the Call Target.
Customer Dial Timeout
Quick access: Callback > Call Targets > Your Call Target > General tab > Callback Strategy
The Customer Dial Timeout setting provides an option to configure the number of seconds your system will maintain an outbound dial before disconnecting the call. Increasing this value allows your customers more time to answer their callbacks, but it can have the effect of increasing the total time that a callback is active in the system if multiple retry attempts fail.
The default value for the timeout is 120 seconds (two minutes), with a minimum value of five seconds. Calls that breached the timeout can be seen in the Call Detail report with an Outcome of Unavailable, with a Retry Reason of customer_leg_timeout.
Number For Extension Only Callbacks
The Number for Extension Only Callbacks setting takes effect when Allow Extension or Phone Number is enabled in the Registration section. If you opt to allow callers to enter an extension rather than a complete phone number, you'll need to enter a phone number to be dialed prior to the extension. Mindful will use this number to launch the callback and will dial the extension once the call is connected.
To learn more about extension-only callbacks, see Registration Settings.
Post Dial DTMF (Callback)
The Post Dial DTMF (Callback) setting allows you to specify a string of digits to be pulsed after the Callback Number is dialed. This is useful if the Callback Number leads into an Interactive Voice Response (IVR) system that requires some additional input, such as a menu option or an internal extension.
- The Post Dial DTMF setting only applies to calls made to your contact center, but there is another option available to dial an extension for calls to customers (the Allow Extension or Phone Number setting).
- Post Dial DTMF applies to both the Callback Number and the Callback Failover Number.
- This setting does not apply to the Choose Hold Number.
- Post Dial DTMF applies to both Customer First and Agent First interactions, but the digits are only sent on the agent leg of the call.
A maximum of 1,024 characters can be entered into the Post Dial DTMF field.
Formatting the Post Dial DTMF String
You can add pauses into the Post Dial DTMF string, if needed, to navigate your IVR properly. This forces the system to pause to allow your IVR to transition to a different menu or speak additional instructions before continuing.
Include a w character in the Post Dial DTMF string anytime a pause is required, as in the following example.
Example
Contact Center Phone Number = 800-555-1234Post Dial DTMF = 1w2413
- The system dials the Contact Center Phone Number, 800-555-1234.
- The contact center IVR answers the call and waits for input.
- The system pulses 1, then pauses for a moment (due to the w).
- The IVR enters a different menu after receiving the 1, then expects an extension number to be entered.
- The system pulses 2413, the remainder of the Post Dial DTMF string.
- The IVR routes the call to the correct extension.
On-deck Callback (Agent First)
The On-deck Callback option only applies to Agent First interactions. This option configures the system to call the customer before the agent confirms they are available. The agent is still called first, but there is less of a delay before reaching out to the customer.
In a normal Agent First interaction, the system waits until an agent is on the line and has indicated they are ready to accept a call before launching the callback to the customer. While this is advantageous to the customer, it increases the average amount of time agents spend on the phone, since they must wait while the call is dialed. The On-deck Callback option increases the efficiency of the Agent First callback strategy by retrieving the customer before the agent has answered the call.
Prompt Customer to Confirm Callback (Agent First)
When enabled, the Prompt Customer to Confirm Callback option gives customers the opportunity to confirm they are ready to speak with an agent after they have answered their callback. This setting is enabled by default. If the checkbox is deselected, the caller will not be asked to confirm they are available and will immediately be connected to an agent.
Leave Voicemail on Early Hangup (Agent First)
The Leave Voicemail on Early Hangup setting will only apply when the Prompt Customer to Confirm Callback setting is disabled.
Offer Rescheduled Callback
Offer Rescheduled Callback allows customers to reschedule their callback for a later time. An additional voice prompt will be played after a customer has answered a callback but before the customer has indicated they are ready to speak with an agent. The agent is released from the call when rescheduling is selected.
This feature is enabled by default for the Customer FirstCallback Strategy. If you wish to disable callback rescheduling for a Call Target, make sure the Offer Rescheduled Callback checkbox is deselected.
This feature is disabled by default for the Agent FirstCallback Strategy. If you wish to enable callback rescheduling for a Call Target using the Agent First Callback Strategy, the Prompt Customer to Confirm Callback option must be selected. This will then allow you to use Prompt Customer to Reschedule Callback.
For both Callback Strategies, after a customer has pressed 3 to indicate they would like to reschedule, they will hear the following additional prompts and options.
Customer Prompt Loops
The Customer Prompt Loops setting allows you to control the number of times the system will present the menu options to callers after a callback is made. The confirmation voice prompt plays once immediately after the callback is connected, and will repeat after several moments of silence, based on this setting. Callers can be presented with the following options:
- Speak to an Agent (Always On)
- Cancel the Callback (Always On)
- Re-schedule the Callback (Optional)
- By default, Customer Prompt Loops is set to 2 for Agent First Call Targets and 4 for Customer First Call Targets.
- The maximum value allowed is 10. An error will be displayed on the screen if you attempt to enter a higher value, and you will not be able to save the Call Target until the invalid value is removed.
Wait For Live Agent (Customer First)
The Wait for Live Agent setting forces the system to wait until an agent confirms they are available before connecting a customer call. Wait for Live Agent only applies to the Customer First Callback Strategy. This setting helps Callback to differentiate between customers waiting in a holding queue and customers talking to agents. This differentiation helps to calculate a more accurate ECBT.
When Wait for Live Agent is disabled, all callbacks in are considered to be in the Talking phase as soon as they are connected to your contact center.
Example
The following example illustrates the customer experience with Wait for Live Agent enabled. Note that any hold music or other prompts that would play on the ACD will not occur while the customer waits for a live agent.
We recommend keeping Wait for Live Agent enabled for all Call Targets using the Customer First Callback Strategy. Waiting for agent answer confirmation allows Mindful Callback to identify the precise moment that a conversation between a customer and agent begins. This helps Callback to calculate accurate statistics for Return Call Hold Time and Talk Time.
If you do not wish to require your agents to confirm they are available via voice or DTMF input, it may be possible to automate the agent answer confirmation. To see if this can be done in your integration, contact the Mindful Support team.
Agent Answer Recognition (Customer First)
The Agent Answer Recognition toggle switch determines what is required from agents to indicate they are ready to accept a callback.
- When set to Voice, agents will hear a customizable audio prompt instructing them to say specific words or phrases to accept.
- When set to DTMF, agents will be prompted to press 1 to accept.
- The Agent Answer Recognition feature must be enabled by Mindful staff prior to use. Otherwise, the toggle switch and its related settings will not be accessible.
- This setting only applies when First Party Called is set to Customer and Wait for Live Agent is selected. Voice detection is not available for Agent First Call Targets.
- This feature only detects speech in the English language at this time.
- When the Priority Queue Timeout is set to Infinite, Agent Answer Recognition cannot be set to Voice. An error message will appear if you attempt to enable voice recognition with an unlimited timeout. This ensures that a timeout is always in place if a technical issue prevents agent voice recognition.
Whether Agent Answer Recognition is set to DTMF or Voice, the concept of requiring confirmation before connecting customers allows agents time to review screen-pop data or other information about a callback before beginning a conversation.
It is possible to automate the process of agents accepting calls by automatically playing an audio prompt (containing voice or DTMF) from the agent side when an agent connects to a call. We recommend the following best practices to ensure the audio prompt is recognized by Mindful (especially in Avaya environments):
- Record the audio file in a soundproof environment, such as a sound booth, to avoid background chatter.
- Include one second or more of leading silence in the audio file. You may need to adjust the leading silence based on your environment.
- If you are using DTMF (Agent Answer Recognition = DTMF), note that the audio quality is more important than the volume of the tone. Mindful has a DTMF tone audio file available (tone.wav) to assist with DTMF recognition.
Customizing Voice Recognition
Beneath the Agent Answer Recognition selector, you will see several fields that impact the voice recognition feature:
- Allow Any Speech Recognition
- Recognition Keywords
- Confidence Level
See the sections below for more information on these settings.
Customizing Audio for Speech Recognition
The Agent Answer Recognition setting will determine which audio prompt is played to agents to tell them how to accept a callback.
- When set to DTMF, the Prompt Agent to Connect (DTMF) audio prompt will be played, which says "To engage the caller, press 1" by default.
- When set to Voice, the Prompt Agent to Connect (Voice) audio prompt will be played. This prompt says "To connect to the caller, say 'Connect'" by default.
If you add or remove Recognition Keywords from the list, or if you enable the Allow any Speech Recognition setting, we recommend updating the Prompt Agent to Connect (Voice) audio prompt to advise agents of the complete list of recognizable words or phrases.
For example, if you added the word Ready to the list of Recognition Keywords, you might update the audio prompt to say "To connect to the caller, say 'Connect' or 'Ready'."
For complete instructions on updating audio prompts, see How to Manage Audio Files.
Allow Any Speech Recognition
You can use the Allow any Speech Recognition check box to instruct the speech recognition engine to ignore the configured Recognition Keywords and simply recognize any speech from agents. When selected, agents can respond with any speech when prompted to connect to a caller. This setting is only available when Agent Answer Recognition is set to Voice.
When Allow any Speech Recognition is enabled, agent speech will be positively detected under either of two conditions:
- An agent speaks any word or phrase that is not configured in the Recognition Keywords list.
- Agent voice is detected but the required Confidence Level is not reached.
This setting can help to address any issues with low volume from the agent side, and also avoids the need to train or instruct agents on the exact keywords to use.
If you enable the Allow any Speech Recognition setting, consider updating the Prompt Agent to Connect (Voice) audio prompt to advise agents that they can respond with any speech to connect to a customer. By default, the audio file plays "To connect to the caller, say 'Connect'."
Recognition Keywords
When the Agent Answer Recognition radio button is set to Voice, you can use the Recognition Keywords list to define the words or phrases that agents can speak to accept a callback.
Adding and Removing Keywords
To add new keywords or phrases, type a word or phrase into the text field. Press Enter when finished to add the word or phrase to the list. Only one word or phrase can be added at a time, but you can continue to add words and phrases in the text field to expand the list.
We recommend using single words or short phrases as Recognition Keywords.
In addition, it can be helpful to add variations of your intended keywords. For example, if one of your keywords is "yes," you might consider adding "yeah" to catch variations in agent speech.
By default, the keyword connect is configured in the list. The default keyword connect aligns with the default audio file Prompt Agent to Connect (Voice), which plays "To connect to the caller, say 'Connect'.".
If you add or remove Recognition Keywords from the list, we recommend updating the Prompt Agent to Connect (Voice) audio file to advise agents of the complete list of recognizable words or phrases.
For example, if you added the word ready into the list of Recognition Keywords, you might update the audio file to play "To connect to the caller, say 'Connect' or 'Ready'."
For complete instructions on updating audio prompts, see How to manage audio files.
Confidence Level
With Agent Answer Recognition set to Voice, you can set a threshold for positively identifying that an agent has spoken a keyword with the Confidence Level setting. A default value of 80 is used for the Confidence Level setting unless you specify otherwise. This means that the speech recognition engine must reach 80% confidence that a keyword has been spoken before it identifies a match.
When the speech detection engine recognizes that an agent has spoken a keyword, it assigns a confidence score as a percentage value. This value represents the level of certainty that was reached in recognizing the keyword. Higher values indicate clearer and more precise speech that was more easily recognized, while lower values indicate that the speech detection engine was less certain that the spoken word truly matched a keyword.
Configuration Tips
- Lower the value to loosen the confidence requirement and positively identify agent speech more often. Be aware that you will risk increasing the potential for false positives, where the system recognizes an incorrect word or phrase as a match.
- Increase the value to tighten the confidence requirement for more precision, with the risk of missing some instances of correct keywords that were not entirely clear.
- We recommend testing agent speech at various levels of clarity before and after making changes to the Confidence Level. This can help you to fully understand the impacts of raising or lowering the value.
Priority Queue Timeout (Customer First)
The Priority Queue Timeout setting controls the maximum number of seconds allowed for a callback to be sitting in a priority queue before being timed out. When the timeout is reached, the waiting customer will be automatically connected to an agent. This setting applies when Wait for Live Agent is enabled and the Customer First Callback Strategy is being used.
By default, the value is set to Infinite. Setting a value here essentially enables the feature. When Agent Answer Recognition is set to Voice, the Priority Queue Timeout slider cannot be set to Infinite. This ensures that a timeout is always in place if a technical issue prevents agent voice recognition.
We recommend using Priority Queue Timeout to ensure that no customer waits on hold longer than two minutes in the high-priority holding queue. Consider your Average Speed of Answer (ASA) metric when determining an ideal value for your call center. You can use one of the two options below as a starting point, then make small adjustments as needed if you still receive Priority Queue Timed Out alerts.
- Option 1: Use the formula ASA * 1.2 to set an initial value. This takes the ASA metric and adds a 20% buffer to allow a little extra time.
- Option 2: Set the value to the longest hold time included in your ASA metric, excluding any large outliers.
Retry Delay
The Retry Delay setting allows you to slow down the pace at which the system retries failed callback attempts. With a higher value, retries will be attempted further apart, allowing more time for customers to become available after missing one or more callback attempts. The value can be set anywhere from zero seconds to two hours.
Retries per Callback
The Retries per Callback setting determines how many times a callback will be reattempted when the first attempt to reach the customer is unsuccessful. Unsuccessful callback attempts can occur when:
- the caller does not answer after a number of ring attempts, determined by the Customer Dial Timeout setting.
- an answering machine or voicemail intercepts the call.
- a busy signal is encountered indicating that the line is engaged.
- the call results in a telephone carrier intercept because the phone number is invalid, disconnected, or the carrier has other issues completing the call.
You can set the number of callback retries anywhere from 0 to 40. Once the maximum number of retries has been reached, the callback will be marked as unsuccessful on the Call Detail report and will not be attempted again. Additionally, the Retry Callback button will no longer be available on the Call Detail page for that callback request.
Retry After Hours
When the Retry After Hours checkbox is selected, the Call Target will continue to make retry attempts even if they would be made outside of business hours. When the checkbox is deselected, no further retries will be made once the end of business hours is reached.
- The Retry After Hours setting only applies when the Retries field is configured with a value greater than 0.
- The first attempt for a callback can always be made after hours, regardless of the Retry After Hours setting. To learn more about managing first attempts, see Best Practices for End of Day Handling.
Country Codes
Country codes can be used to configure Call Targets for international dialing. This setting allows you to configure country code prefixes that will be appended to a customer's ANI but not announced to the customer when confirming their callback phone number.
A maximum of 1,024 characters can be entered into the Country Codes field.
International dialing options (Country Codes, Spoken ANI Prefix, Require Country Code, and Minimum/Maximum Callback Number Length) will only be visible to customers who are authorized for non-US dialing. Contact a Mindful representative for help with international dialing
You can list multiple codes in the Country Codes field by separating each code with a comma. The first country code provided in the list will be used when no ANI is available. For example, if you have entered 358,44,39 and no ANI is available, the system will automatically append 358 at the beginning of the phone number that the customer entered when they requested a callback. If you believe that a specific country code will be used more often than others, consider listing it as the first number in this field.