Migrating from On-Premise Callback to Mindful Callback

We know - name changes can be hard. It's also hard to re-wire our brains to speak in Mindful terms when we've lived in the On-Premise world for so long. So we created this terminology matrix to help you understand what the On-Premise Callback functionality translates to in Mindful Callback.

The platforms share a lot of similar features and functionality. The names are different, but the concepts are the same.

Application

Configuration

Reports

Dashboards

User Roles

 Comparing ECBT and EWT

Similarities between EWT and ECBT

For both EWT and ECBT, the calculation can essentially be reduced to (Data Point 1) * (Data Point 2) = Wait Time. Both systems have different data points available for their respective calculations, but the general approach is the same.

On-Premise Callback Calculation

In On-Premise Callback, the EWT calculation for a queue was pretty straightforward and not too challenging to remember and explain to clients.

The ECBT calculation in Mindful Callback is not as simple as the EWT calculation in On-Premise Callback, but the same principles are used to determine the expected wait time.

• On-Premise Callback only uses two data points in the calculation, but Mindful Callback uses several more.
• Some of the data points in Mindful Callback use arrays to determine the value for the calculation, just like On-Premise Callback.
• Mindful Callback also counts the number of calls in queue, just like On-Premise Callback.

Arrays and Answered Events

On-Premise Callback and Mindful Callback both use the time delta between answered events as part of their wait time calculations. On-Premise Callback called this Per Call Wait Time (PCWT). Mindful Callback refers to this as Average Seconds Per Agent Answer.

On-Premise Callback and Mindful Callback both drop samples into an array:

The following steps show the process in action.

1. Detect Answered events.
2. Calculate the time delta for each answered event.
3. Drop the delta as a sample into the array.
4. Calculate the Average Seconds Per Agent Answer data point using a rolling average of the samples in the array.

While On-Premise Callback uses a single array to calculate EWT, Mindful Callback actually uses multiple arrays. The additional arrays cannot be directly compared to a function of On-Premise Callback.

We discuss arrays in detail in the ECBT Calculation module in the Mindful Callback Foundations course.

Array Settings

On-Premise Callback had several configurable settings available to change the function of the array. For example, you could:

• Use an average or a median.
• Change the number of samples allowed.
• Configure the length of time that a sample will remain in the array.

In Mindful Callback, ECBT settings can be configured by Mindful staff in an attempt to correct overestimation or underestimation in the ECBT calculation for a Call Target. Mindful staff can:

• Set the minimum timeframe for the rolling average calculation of the average seconds per agent answer metric.
• Customize the calculation of the average seconds per agent answer metric.

Tracking Calls

On-Premise Callback and Mindful Callback both rely on the number of calls being tracked to help in the wait time calculation. The stage of the call plays a critical role (inbound IVR, pending callback, priority queue).

Tracking Calls in On-Premise Callback

On-Premise Callback uses the number of total calls being tracked as part of the EWT calculation.

The following four stages in the journey of a call in On-Premise Callback were added together to get the Total Calls data point used in the EWT calculation.

Tracking Calls in Mindful Callback

Mindful Callback does not use a "total calls" concept as part of the ECBT calculation, because it doesn't have visibility into all stages of a call's life cycle. Instead, it looks at the number of calls at various points in the journey. These points in the journey are referred to as call stages or phases.

Let's use the same four stages we used to describe On-Premise Callback to draw a comparison for Mindful Callback.

1. Inbound IVR

Callers selecting a callback or interacting on inbound menus for On-Premise Callback were referred to as Inbound IVR calls. Mindful Callback uses the term Registering Phase for calls in this stage.

In On-Premise Callback, these calls counted towards the EWT calculation. The number of calls at this stage do not count towards the ECBT calculation in Mindful Callback.

2. Holding Queue

Mindful Callback has no concept of a holding queue and does not monitor calls queuing at a normal priority. Callers are still presented with an option to speak to the next available agent or choose a callback (ASAP or scheduled). The offer can occur in one of two places.

• Method 1: Ask the caller in Mindful Callback. This method involves Mindful Callback asking the caller if they would like to receive a callback or wait in queue for the next available agent. Callers that choose to hold are routed back to the ACD but no longer tracked by Mindful Callback. Callers that choose to receive a callback will continue to be tracked by Mindful Callback.
• Method 2: Ask the caller on the ACD. This method involves the client asking the caller on their ACD if they would like to receive a callback or wait in queue for the next available agent. There is no message sent to Mindful Callback about a caller choosing to hold and the call is not tracked by Mindful Callback. Callers that choose a callback are routed to the Mindful Callback application to register, and these calls will continue to be tracked.

3. Virtual Queue

It is better to think of this as two different queues within Mindful Callback; one for Pending calls, and one for calls on an internal waitlist.

ASAP callbacks waiting to dial fall into one of two "queues:"

• Pending is for all ASAP callbacks that have not been dialed, for which the current time is still before the Minimum Callback Delay. This is where all ASAP callbacks begin.
• ASAP callbacks in Pending will move to the waitlist when the current time is past the Minimum Callback Delay for the Call Target.

Mindful Callback will use the number of pending calls, the number of waitlisted calls, or the sum of both in the ECBT calculation.

4. Priority Queue

A high-priority queue in Mindful Callback is just like a priority queue in On-Premise Callback. It is where customers wait on the ACD for an agent after receiving their callback. The number of calls in the priority queue is used to help calculate ECBT for a Call Target.

New phase: Callers talking to agents

Mindful Callback keeps track of customers who received their callback and are actively talking to an agent. This differs from On-Premise Callback, which stopped tracking calls when they were answered by an agent. The number of callers talking to agents after receiving their callback is used to calculate ECBT for a Call Target.

Recap

• Mindful Callback does not have a concept of the "total number of calls." Instead, Mindful Callback considers the number of calls at different stages.
  • Pending ASAP callbacks
  • Waitlisted ASAP callbacks
  • Pending ASAP callbacks + waitlisted callbacks
  • Calls in a high-priority queue
  • Callers talking to agents after a callback
• We still track answered events.
• Arrays are still there.
• Mindful staff can configure ECBT settings.

We have a training module available that takes a deep dive into the ECBT calculation. Contact the Mindful Training team for details.

Operation Modes are used in On-Premise Callback to shape the inbound caller experience. Although there is no concept of Operation Modes in Mindful Callback, we can achieve the same results by enabling various settings in the application.

This section will show you how to configure Mindful Callback to imitate On-Premise Operation Modes.

Normal Mode

Normal mode provides callers with the option to:

• receive an ASAP callback
• schedule a callback for a later time
• continue to wait on hold
• transition to a messaging conversation

Normal Mode in Mindful Callback

Use the following combination of settings to imitate Normal mode in Mindful Callback:

Quick access: Callback > Call Targets > Your Call Target > General tab

1. In the Registration Settings section, make sure the Callback offer checkbox is selected. Only Mindful staff have access to this setting, so you'll need to contact the Support team if it is currently deselected.

2. In the Callback Scheduling Settings section, select the Voice Scheduled Callbacks checkbox.

Quick access: Callback > Call Targets > Your Call Target > Messaging tab

3. Select the Voice to Messaging checkbox. This will provide callers with the option to transition to a messaging conversation.

Call routing with On-Premise Callback is built to have the ACD "ask" On-Premise Callback through a CTI connection if a caller should be routed to it to receive some defined inbound treatment or be routed to queue. The recommendation for clients was to modify their call routing logic to place the "ask" of On-Premise Callback immediately after segmentation and just prior to queuing the call to agents. This process was referred to as a Route Request.

On-Premise Callback would consider several factors to determine whether a call should be routed to it for inbound treatment. One of these factors was the Turn-On-Threshold. This was the minimum EWT for a queue before a call was considered eligible for inbound treatment.

Route Request in On-Premise Callback

So how does the Route Request come into play with Callback? There is no configuration value of a Turn-On-Threshold in Callback and no CTI Route Request that can be performed. The client will monitor their own queue conditions and determine whether a caller should be offered the option for a callback. One of the factors in consideration is whether to offer based on EWT, this is where the concept of a Turn-On-Threshold exists and is set by the client as part of their call routing logic. The client will define the rules on their ACD to determine when to offer callbacks to callers.

There are two methods available to offer the option for a callback or to remain on hold:

1. A Callback solution can be built to have Callback ask the customer if they want to choose hold or receive a callback.
2. A Callback solution can be built to have the ACD ask the customer if they want to choose hold or receive a callback.

Recapping the Offer Strategies

Here are the key points to remember:

1. The client determines when to offer a callback to callers. There is no Route Request with Callback
2. The concept of a Turn-On-Threshold does exist, but there is no configuration in Callback for this.
3. Any time a call is routed to Callback, it will always be presented with an option of an ASAP callback. Scheduled callbacks and transition to messaging are optional.
4. When a choose hold option is offered during call routing logic on the client's ACD, the best practice would be to leave the option off in Callback. It is ultimately the customers' decision here so if they want it on in Callback, then turn it on.

On-Premise Callback allows you to define Business Hours for each queue. Mindful Callback requires the same configuration, but it is referred to as Hours of Operation.

Hours of Operation in Mindful Callback provide the same functionality as Business Hours in On-Premise Callback, with one major difference: in Mindful Callback, you can specify Hours of Operation for the entire Organization while creating unique Hours of Operation for each Call Target.

Both applications provide options for changing the callback offer settings towards the end of the day to minimize the impact of inbound call traffic as the close of business hours approaches. On-Premise Callback allows you to change the behavior of outbound dialing towards the end of business hours, but Mindful Callback does not.

Similarities

Before discussing the end-of-day treatment, let's see where the two applications work the same:

• Inbound call treatment: During business hours (Hours of Operation), customers can register new callbacks or transition to a messaging conversation, depending on the Call Target or queue configuration.
• Scheduled callback times: Both applications allow you to specify how many scheduled callbacks are allowed per 15-minute interval within the configured business hours.

Smart Operation Mode vs. End of Day Handling

On-Premise Callback and Mindful Callback handle end-of-day traffic in a similar way, although the configuration and available options are different.

In On-Premise Callback, you can use the Smart Operation Mode feature to automatically change the Operation Mode when the EWT extends beyond the end of Business Hours. In Mindful Callback, you can enable End of Day Handling features to inform callers that the business is closing soon and offer them the choice to wait on hold or register a callback for the next business day.

To learn more about End of Day Handling in Mindful Callback, see Call Target End of Day Handling and Best Practices for End of Day Handling.

Invalid Responses

On-Premise Callback has configuration available for nearly 20 different places in the inbound call flow to manage invalid responses.

On the other hand, Mindful Callback does not have any configurable options for invalid responses on the inbound or outbound call flow. Callers are typically allowed between two and four invalid responses, depending on where they are in the call flow.

Inbound Invalid Responses

During an inbound call, a caller would get 2-4 invalid response attempts before being transferred to queue or disconnected.

• If the Callback Offer checkbox is enabled, the caller will be transferred to queue after the maximum number of invalid attempts is reached
• If the Callback Offer checkbox is disabled, the caller will be disconnected after the maximum number of invalid attempts is reached.

Outbound Invalid Responses

During an outbound call, a caller will get 2-4 invalid response attempts before the call is disconnected. The callback will be retried and rescheduled until the retry attempts have been met.

Mindful Callback offers a simplified experience for callers who already have a pending ASAP or Scheduled callback. These callers will be informed that they have already registered for a callback, and the call will be disconnected.

There is no option to:

• Cancel a callback and transfer to the holding queue
• Transfer callers to the priority queue after canceling the callback if their remaining wait time was within a defined time period

On-Premise Callback and Mindful Callback both offer the same two experiences for customers after they receive their callback:

• Customer first - Customer is dialed first
• Agent first - Agent is dialed first

These are referred to as Queue Modes in On-Premise Callback and Callback Strategies in Callback.

Customer First

The Customer first Callback Strategy follows the same caller experience as On-Premise Callback's Predictive Queue Mode.

1. Customer is called when it is their turn in queue.
2. Customer accepts their callback.
3. Customer is transferred into an ACD queue at top priority to wait for an agent.

Agent First

The Agent first Callback Strategy follows the same caller experience as On-Premise Callback's Agent Priority Operation Mode.

1. A call is placed into an ACD queue at top priority.
2. An agent answers the call and is told to wait as a callback is in flight.
3. The callback to the customer is made.
4. The caller and agent are merged together.

In On-Premise Callback, there are about five different types of retry counts to help reschedule callbacks after a failed attempt to get a customer on the phone.

In Mindful Callback, there is a single retry attempt count to rule them all. Callback doesn't determine if the callback failed because of a busy signal or no answer; a failed callback is simply a failed callback.

Once a callback reaches the maximum number of attempts, the callback is canceled.

Configuring Retries

You can configure the retry attempts for each Call Target in General > Callback Strategy. There are only two settings to configure:

• Retries - The total number of retries to launch the callback if the callback fails.
• Retry Delay - Number of minutes in between each retry attempt

On-Premise Callback contains a large number of customizable menu options for how to present menu options to incoming callers.

This has been greatly scaled back in Mindful Callback. The Callback menu options allow callers to quickly make their choice, and continue with their interaction.

Menu Options

The menu options presented to callers can contain any or all of the following options:

On-Premise Callback offers configuration to allow callbacks to dial from a client's telephony environment and rules to prevent the registration of callback phone numbers. Let's review some of the ways that phone numbers and calls are handled in Mindful Callback based on rules implemented in On-Premise Callback.

In this section, we compare and contrast real-time and historical reporting features in the On-Premise Callback to features in the Mindful Callback Dashboard.

Real-Time Reporting

The On-Premise Callback real-time reporting engine was called Dashboards. Clients relied on this heavily to get a current pulse on conditions in the call center when using On-Premise Callback. The Mindful Callback real-time reporting engine, Callback Status, provides the same metrics known and loved by On-Premise Callback clients.

This table breaks down the metrics with in Dashboards showing the comparable metric in Callback Status.

The Global Snapshot in On-Premise Callback provide a clients a list of pending ASAP, retry, and scheduled callbacks for a Queue. What was loved most by clients was the ability to cancel pending ASAP, retry, and scheduled callbacks. The Call Detail feature in Callback provides the same functionality to clients.

The Global Snapshot did allow clients to delete all the pending callbacks for a queue. One small difference is there is no ability to cancel all callbacks for a Call Target with Call Detail.

Historical Reports

On-Premise Callback had multiple Performance Reports and System Reports available for clients to leverage to understand the impact the solution provide their call center. When it comes down to it, clients loved certain data points available on different reports. Callback captures several of these data points.

This table provides data points that appear on various performance reports for On-Premise Callback showing a corresponding data point in Callback.

This next table provides data points that appear on various performance reports for On-Premise Callback showing a corresponding data point in Callback.

Readerboard Statistics and the Mindful API

In On-Premise Callback, the Readerboard Adapter provides real-time statistics to be displayed on agent dashboards, wall-screen readerboards, and anywhere such metrics are needed.

In Mindful, the Metrics API provides similar statistics via a secure RESTful API. The Metrics API provides many of the same stats as the Readerboard Adapter and a few new stats that are only relevant in the Mindful ecosystem.

For complete documentation on the Mindful API, see the Mindful API Reference article.

See the table below for a comparison of Readerboard Adapter and Mindful Metrics stats.

New Metrics in the Mindful API

The Mindful API provides a few additional metrics only relevant Mindful Callback. These metrics capture statistics that are not available in On-Premise Callback due to the differences in the two services.

A: ECBT is the next evolution in calculation when a customer can expect their callback.

One of the services Mindful Callback provides is the calculation of Estimated Callback Time (ECBT). The system accurately predicts the time a customer can expect to receive their callback. ECBT is the next evolution in estimating the time when a customer can talk to an agent.

EWT took into consideration the information from the Holding Queue. Now, with ECBT, at any given moment, the calculation looks at many data sources, like talk-time metrics (Agent speaking with customer), waitlist forecasts, and the configuration of each Call Target. The system dynamically uses different formulas to calculate ECBT based on different scenarios. For more information, see Understanding Estimated Callback Time (ECBT).

Q: How should we use Smart Rules?

A: Smart Rules can be utilized to automate processes, such as:

• Disabling callback completely at the end of the day
• Increasing aggressiveness of a Call Target to clear calls at the end of the day or before a large drop in agent count

Q: We use Screen Pop. Is anything changing with that integration?

A: We still have Screen Pop integration.

We still offer Screen Pop integration, but it isn't a quick flip of a switch to turn it on. You'll need to work with Mindful to set up the integration.

In the call flow, if you pass ANI to your contact center to trigger Screen Pop, you're in luck! The ANI of the agent leg of the callback will always be the customer's callback number.

If information needs to pass through Mindful Callback and back to the contact center, configuring SIP headers is the easiest way to accomplish this. If you are not using SIP, we can pass data using Mindful Datastore. For UCCE/ICM environments, a CTI Event Relay can be set up if requested. 

Q: We use Readerboard and/or Real-time Adapter in On-Premise Callback. Is there something similar in Mindful Callback?

A: Get real-time data with the Metrics API endpoint.

The Metrics endpoint in the Mindful API suite will provide the same real-time data that can be found on the Callback Status page and will provide high-level information from the Call Details page in Mindful Callback. You can request Metrics data for specific Call Targets, or from all Call Targets at once. 

Q: How do we connect calls to agents?

A: In On-Premise Callback, connecting a call to an Agent required no interaction by the Agent. In Mindful Callback, Agents need to press 1 in order to accept the call connection. Changing the Agent's pool habits can be difficult, so there are a couple of available options:

• You can have your developer team build a tool on your end that passes the DTMF of 1 when the Agent is available.
• You can create a tool or script (using an integration) to connect the Agent to the call as soon as the Agent is available.
• Additionally, we offer a Voice Recognition feature where anything recognized as Speech or an Agent speaking a specific word(s) will connect the call.

Q: Is the Mindful platform SOC 2 certified?

A: Yes, it is.

Q: Can we have a copy of the SOC 2 certification?

A: Yes, please talk with your Sales Team who will connect you with the security team to deliver that to you via DocuSign.

Q: Is penetration testing conducted?

A: Yes it is, twice a year.

Q: Is vulnerability testing conducted?

A: Both internal and external vulnerabilities are tested. Timeframes can be requested from your Sales Team.

Q: If we request a security questionnaire be filled out by Mindful, how long will it take?

A: Mindful has a 5 business day SLA for security questionnaires.

Q: Are there servers/hardware to maintain?

A: Based entirely in the cloud, the hardware dependencies, server, and database maintenance are removed.

Q: How are upgrades handled?

A: All updates happen automatically, including new features. Notification of releases and outages can be subscribed to for ease of monitoring.

Q: Do upgrades on Edge devices affect Mindful?

A: Mindful is almost completely agnostic to all versioning of other vendors and their updates.

Q: Is the platform high availability (HA)?

A: Yes, always.

Q: What are best practices for installation? 

A: TLS and SRTP are recommended protocols for security, especially when passing user data.

Q: Are any hardware/servers required? What about database requirements? 

A: Based entirely in the cloud, there are no hardware dependencies and therefore no server security maintenance nor service accounts are required. There is no database requirement, with everything needed for callback stored within the platform. The exception to this could be the passing of user data in certain environments.

Q: How are upgrades handled? 

A: All updates happen automatically with rare/planned service interruption, including security updates and new feature deployment. There is no additional cost for new features available within your Platform Tier.

Q: How are we notified of new releases and outages? 

A: Notification of releases and outages can be subscribed to for ease of monitoring.

Q: What happens when our other infrastructure pieces require an update? 

A: Mindful is almost completely agnostic to all versioning of other vendors and their updates.

Q: Who controls and manages the call routing?

A: You do! The client is in full control of routing both inside and outside of Mindful, with all routing options in Mindful available to the client. There is also more flexibility about when and how callback is offered.

Q: What license/resource limitations are pre-set?

A: Licenses are replaced by "conversations" in Mindful but are not limited anywhere in the software. From a technical perspective, there is no resource throttling. For more information on conversations and how those are billed/tracked, please contact your Sales Team.

Q: What are the firewall requirements? 

A: It will be necessary to whitelist Mindful's web interfaces, APIs, SIP and RTP proxies over an extended port range.

Q: Will returning customers notice a difference?

A: Returning customers will be affected the least of all by migration. Prompts will be in the same voice they recognize, the path to Callback is incredibly similar, and the feature set has everything they are used to, PLUS more options (Text Scheduling/Notifications) than they had before.

Q: What if we choose to offer Callback on our IVR What additional work does this create?

A: Prompting must be added to include the Callback Offer. Additionally, some thought must be given to the best way to inform customers of hold times in order to encourage them to accept the Callback. Speaking the maximum wait time is suggested, as take rates are increased when some expectation of wait time is delivered. The Callback expectation can be refined by announcing ECBT after Callback has been chosen.

It is likely you are already doing a check in the routing for business hours - is the call center open/is it closing soon, whether an agent is available (with a decision to route directly to hold if so) and a check of EWT or longest call waiting. These should definitely be incorporated into the routing decision before sending a call to Mindful for treatment.

Q: Will we need to record new voice prompts?

A: This will depend on whether you use your own voice talent or not.

• You use your own voice talent: Yes, you will need to record a full set of Mindful voice prompts. There are far fewer than were used in On-Premise Callback, so the work/expense should be far less.
• You use the same GM Voices talent that Mindful used on On-Premise Callback and now in Mindful Callback: No, you do not need to record new files. We can offer some branding and even record your Caller ID number in the proper voice through our GM Voices contract with no fee. Discussion with your Integration Team is suggested for full best use.

Q: How should we utilize customized media sets? 

A: Customized media sets can be used in many ways!

• Initial callback prompts can include branding - the customer will hear your company name at the front of the callback
• Music files to match what is played on your IVR
• Callback CID prompt - if recorded/uploaded and enabled in the Call Target settings, this file will play in place of the more robotic delivery of each digit
• Additional content can be included, like a notice of call recording or a survey request. These can be added in a few places, and a discussion with Support can pinpoint the best location. Attention should be paid to whether the prompt that is being added to is one that loops or not, and if this longer prompt could have a negative impact on the call flow.

Q: Are we limited to just Callback or Scheduled Callbacks? 

A: Each Call Target can have as many or as few options for the customer as required. It is possible to only offer Callback, although a best practice is to include Scheduled Callback whenever possible.

Q: How can we inform the customer that the Callback is from our company?

A: Your company ANI is displayed and branding of prompts is still available, and with fewer voice files, the process is easier with Mindful. Work with your Integration Engineer to see what your options are.

Q: What can we do to mitigate long return call hold times on smaller queues? 

A: Utilizing the Agent First strategy on queues with wildly fluctuating AHT or with limited staffing can remove return call hold time and provide a customer-centric experience.

Q: How is End of Day handled?

A: There are several options to handle the end-of-day timeframe gracefully, including but not limited to a return to hold option, scheduling for the next day, and allowing registered calls to process with after-hours agents.

Q: Can we add other options/channels later? 

A: Yes! It is incredibly easy to enable Notifications, Automations, Digital Callbacks, and Digital Intents, and they can be enabled at any time. Reach out to Cloud Support and they will work with you to set up any of these additional free features.

Q: How do we submit feature requests? What is the best way to let Mindful know what features would be the most helpful? 

A: Using the Submit Feedback button in the lower left corner of the UI directly inputs your request into our tracking software. As this is attached to your organization and you can best speak to how the feature would benefit your business, submitting this way carries the most weight.

Q: What is the difference between Business Units and Reporting Categories? 

A: Business Units are used to lock the User Role of Viewer into seeing a specific subset of Call Targets. This is all about what a User sees or doesn't see in the UI.

Reporting Categories allow for the grouping of Call Targets into their own report. The choice then becomes whether you d like to see the report of an individual Call Target or a group of Call Targets. Once a Reporting Category is created, it is visible in the Filter by Call Target or Category dropdown under the list of Call Targets.

Q: What is the difference between disabling Callback Processing and disabling Callback Registration?

A: Callback Processing is the act of dialing the callback. Disabling this process will allow customers to request a callback but will stop all callbacks from launching. You might use this if there is a temporary issue with dialing out or with the agent s desktop application. Disabling this for a long period will build up undialed callbacks that will either need to be dialed or canceled.

Callback Registration is the first step in Callback and is when the customer's information is collected to be able to place the Callback later. Disabling this process will allow Callbacks that are already registered to be dialed but doesn't allow any new Callbacks into the system. This would be used only in emergency and would be better handled by stopping the flow of calls to Mindful from the ACD, as Mindful will play a message to the caller and then disconnect.