ORECX CALL RECORDING BLOG

Conversational analytics is the name of the game today in the call recording world. According to leading conversational analytics provider, CallMiner, "Conversational analytics delivers clear insight into the mindset of the customer by monitoring and analyzing their behavior and emotion during every interaction. By categorizing, tagging, and scoring 100% of customer interactions, customer conversation analytics reveals the opinions, desires, and needs driving customer choices and gives companies a deeper understanding of what is required to create exceptional customer experiences."

More and more businesses, large and small, are deploying call center analytics these days, and you want to be sure your current or future call recorder primes you for (rather than prevents you from) effectively implementing conversational analytics. 

Here are four key features your call recorder must have to easily transition you to conversational/speech analytics:

1. Dual channel, high fidelity stereo audio capture is imperative as it enables the recorder to isolate each party on the call onto a distinct line so overtalk is not an issue. Sound quality is also high fidelity, leading to much more accurate transcription and analytics results. 

2. Simultaneous audio recordings for both post-call analytics and real-time streaming. You may opt for post-call analytics today and then want real-time analytics tomorrow. Your call recorder must accommodate recording delivery for both use cases. Most recorders do not.

3. Audio file segmentation to easily identify calls that involve transfers from one agent to another. Many cloud business VoIP recorders offer transferred calls as one long single audio file, which can disrupt call segment distinction for quality assurance and compliance purposes. Other advanced call recording features enable the linking of the disparate segments to a single full interaction through a unique "interaction ID" to keep the complete call together.

4. REST API for open, easy audio data sharing with transcription and analytics engines. This enables real-time analytics. Many recorders require lengthy and complicated integration with outside engines to receive the recording, transcribe it, and then process analytics findings.

Are you ready for analytics? Does your call recorder have these critical functions to prime your business to deploy analytics successfully and efficiently? If not, now is the time to consider changing recording vendors. 

Agent compliance infractions can lead to unwanted penalties for your organization. Consider this:

• Dodd-Frank Act penalties can amount to $1 million or more
• PCI Compliance penalties can be $5,000 per month per agent
• MiFID II infractions can total $10.8M or 2% of your company's annual revenue
• HIPAA infractions can cost up to $50,000
• CMS Final Rule 2023 penalties can amount to $2,007,500 per hospital

Each of these regulations requires certain call recording requirements.

Dodd-Frank Act (U.S. financial services regulation)

• All communications relating to pre-execution trade information must be recorded completely and accurately, including telephone, voicemail, instant messaging, chats, email, and mobile.
• Records need to be uniformly time stamped – A record of the date and time, to the nearest minute, must be on every record.
• Trading records need to be identifiable and searchable by transaction.
• All records must be stored securely and readily accessible. 

-->Oreka TR provides detailed audit trails, time stamping, multi-criteria searching, and secure storage of your calls to help you comply.
PCI-DSS (payment card industry regulation)

• No cardholder data (cardholder name, expiration date, PAN, etc.) should ever be stored unless it’s necessary to meet the needs of your business.
• No sensitive authentication data (SAD), which includes card validation codes (CVV2, CVC2, CID, or CAV2), personal identification numbers and/or full magnetic stripe data, may be stored in a digital, audio or video format (such as WAV or MP3) after authorization, even if encrypted.

--> Oreka TR can pause both screen and audio via API or web user interface while credit card numbers are being received over the phone.

MiFID II (European Union financial services regulation)

• Record all calls which will/may result in transactions.
• Notify the customer that the conversation is being recorded.
• Store all communications for a minimum of 5 years.

-->Oreka TR can record 100% of calls or configure the system to selectively record only certain calls; enable users to easily search for, retrieve and playback specific calls based on multi-criteria searching, and store all communications for a minimum of five years, or for any duration you choose.

HIPAA (U.S. health insurance regulation)

• Call centers need to encrypt and secure all customer data.
• Organizations must maintain records and proof of call recording consent during patient interactions.
• No cardholder data (cardholder name, expiration date, PAN, etc.) should ever be stored unless it’s necessary to meet the needs of your business, and no sensitive authentication data (SAD), which includes card validation codes (CVV2, CVC2, CID, or CAV2), personal identification numbers and/or full magnetic stripe data, may be stored in a digital, audio or video format (such as WAV or MP3) after authorization, even if encrypted.

-->Oreka TR will pause both screen and audio recording via API or web user interface while credit card numbers are being received over the phone. This way, no numbers are stored anywhere on the recording system.

CMS Final Rule 2023 (U.S. Medicare & Medicaid services regulation)

• Agents and brokers need to record calls in their entirety during the enrollment process.
• Calls must be securely retained for 10 years.

Let OrecX help your organization adequately comply with relevant regulations and help you mitigate avoidable penalties.

Try OrecX today for free for 30 days!

Picking the right call recording vendor can greatly impact your customer service levels, agent performance scores and even your compliance penalties. Not all recording solutions are created equal, so you need key questions to ask prospective vendors when you are looking for a new system or considering upgrading your existing one to ensure you are procuring the best solutions for your needs.

Here are the top 10 questions to ask each potential recording vendor:

1. Do you offer call recording, screen recording, quality monitoring and mobile recording software? You want a modular-based solution that offers all four to support your organization's needs today and tomorrow. Some solutions come all-in-one and charge you for those features you don't need. Modularity is important to ensure you only pay for what you will use.

2. Does your recorder capture calls in single channel (mono) or dual channel (stereo)? Mono records both the agent and the customer on the same channel and overtalk can disrupt you from clearly hearing what each individual says. You want stereo recording which captures both parties on separate distinct channels which can be isolated upon playback.

3. Can the recording system scale up to hundreds or thousands of users? You don't know how many channels of recording you will need in six months, one year or even five years. You need a recorder that can scale with you as you grow. Look for a solution that can support as few or as many as you may need someday, ideally from five to hundreds of thousands.

4. Does the system offer speech analytics and transcription, and/or does it provide an easy pathway to procuring those added solutions? Not all recording solutions offer the capabilities and interfaces to enable a smooth migration to analytics and customer experience management solutions. You may need these added features as your organization evolves. The ideal recording system will provide a smooth pathway for you to enable this functionality when and if you need it.

5.  What is the process and cost for accessing/exporting my recording metadata to third party systems like CRM, conversational analytics, and customer experience management? Some vendors actually charge you to export your own data to these systems. The right vendor for you provides a REST API for easy export to these solutions without any charge. After all, it's your data. You shouldn't have to pay extra to access it.

6. Do you support cloud, premise and hybrid storage of my recordings? Your organization may have different locations (featuring different telephony environments) today or in the future. You need a recording solution that provides the flexibility to capture calls in each location and store those recordings how and where you need them stored.

7. Are there any restrictions on interoperability with my existing communications environment (PBX, ACD, etc.)? Many recording solutions are proprietary black box systems that do not play well with others. Your recording system must easily integrate with your current and future environment without significant customization or help from a costly professional services organization. 

8. Can you please provide a demonstration of the solutions administrative capabilities? Some systems require weeks or months of training to learn how to manage. Your recorder should be easy to learn and operate so you may begin capturing and replaying calls, evaluating your agents, and creating performance reports right away.

If you are unsure if your current or potential recording solution meets the bill, contact OrecX. Our open-source based recording solution is the most open, easy to administer and scalable recorder on the market, and we are now owned by CallMiner, the number one conversational analytics provider on the planet. Our solutions provide a direct pathway to transcription and analytics-as-a-service.

Health insurance providers can no longer avoid recording their customer/prospect calls. It is now the law.

CMS Final Rule 2023 (from Centers for Medicare & Medicaid Services - CMS) was announced in May of 2022 and stipulates new call recording requirements for health provider agents. This new rule is set to go into effect on October 1, 2022, and it will require all Medicare and Medicaid marketing calls (including enrollment related interactions) to be recorded and stored for 10 years. Its launch date is designed to go into effect just before the start of the 2023 Annual Election Period for Medicare Advantage and Medicare Part D plans. 

What does this mean for providers?

Health insurance companies must record all calls their agents make, including all sales calls and even enrollment-related conversations. This includes an agent talking a customer through the enrollment process live on the phone.

Recording stipulations include:

• The following interaction types (landline, mobile or Zoom) must be recorded:
  • When agents are calling leads, scheduling appointments, collecting drug and provider lists, and conducting education meetings, and phone enrollments.
  • When an agent is selling a Medicare Supplement in tandem with a Prescription Drug Plan, the call would need to be recorded due to the Part D discussion.
  • Any follow up calls related to sales and completing the enrollment process.

• All these recordings must be retained in a HIPAA compliant manner for 10 years.

The rule applies to all health care field agents and health care professionals (who enroll beneficiaries in new Medicare Advantage and/or Part D plans. 

What should you do?

All companies impacted by this new CMS ruling need to acquire robust call recording software that can capture every single call that is made by its agents and plan enrollers and can safely and securely store those recordings for 10 years. Many all-in-one telephony platforms you may be using offer very limited call recording functionality that will not suffice under the new rule.

It is important to assess your call recording needs by October 1, 2022, in time for the new ruling. 

For years, companies have been recording customer calls for quality assurance, compliance and risk mitigation. For those purposes, standard call recording solutions work fine. But, when you want to start analyzing your calls for customer insight like buying patterns, traditional call recording solutions don't always serve you the best.

For call analysis, you require certain elements in a call recording system that not all recorders have. These include:

• Dual channel audio capture - Recording the agent and the customer on separate, distinct channels so the parties can be isolated upon playback to clearly hear what each person said. Mono, or single-channel recorders, make it impossible to distinguish what was said during overtalk.

• High fidelity audio recording - Many call recorders capture audio using lower quality, low bit rate codecs. You want your calls recorded using high fidelity, wideband, high bit rate codecs like Opus. Opus is distinguished from most high-quality formats (e.g.: Vorbis, AAC, MP3) by having low delay (5 ~ 66.5 ms). It is unique from most low delay formats (e.g. Speex, G.711, GSM) by supporting high audio quality (narrow-band all the way to full-band audio). It meets or exceeds existing codecs' quality across a wide range of bitrates, and it operates at a lower delay than virtually any existing compressed format.
• Cost-free, immediate access to your recordings - Many call recording vendors charge money for you to export your call recordings to a third party analytics solution. These same companies also take time to grant you that access. In a world of immediacy, you can't afford to wait or have to pay to access your own audio data. You want to be able to export your calls freely and immediately, without any extra charge.
• Third party data interoperability - To gain the most telling customer insight from your recordings, you want to be able to easily combine your captured audio metadata with other third party data sources such as CRM, ACD and customer service solutions. This greatly enhances the value of your recorded calls and provides richer, more telling buyer intelligence. Most call recording systems don't enable you to easily combine such data. You want a recorder that has a REST API to make this process seamless and fast.

You see, when you evolve past the point of mere call recording and want to start call analyzing, you need to assess your current call recording solution to ensure it has the capabilities you need to do it successfully and immediately.

In many organizations, call recording is a means to an end. That is, upon audio capture, recorded calls are then transcribed and mined for meaningful keywords and phrases like "mad", "unhappy", or "cancel". Conversational analytics engines automatically identify these words and can alert managers, team leaders and/or quality evaluators who can use those relevant sections of an interaction to better coach underperforming agents. 

The accuracy of that analytics process not only comes down to the quality of the analytics software but also the audio that is transcribed. In fact, the transcription stage of a recorded call's life can go one of two ways, depending on the quality of the audio itself:

1. Spoken words are clearly identified and discerned from each party (e.g., customer and agent)

This happens when the recording system distinctly captures and replays both parties on the call on separate recording channels. Rather than mono audio capture, these solutions isolate each voice, enabling transcription engines to very clearly detect each voice in an isolated manner. This separation leads to significantly higher transcription accuracy and minimizes erroneous results with can mislead and waste time.

2. Spoken words are jumbled and misidentified 

This occurs often when single channel, mono recording solutions are utilized, which capture both parties on the call on the same recording channel. We all know customer service calls can become contentious, and overtalk does occur. When this happens, the transcription engine has trouble discerning what each individual said. 

Speech recognition/transcription software works by breaking down recorded audio into individual sounds. It then analyzes each sound, using custom algorithms to identify the most likely word fit. Once determined, those sounds are transcribed into text.

"Converting speech to text works through a complex machine learning model that involves several steps. Let's take a closer look at how this works:

1. When sounds come out of someone's mouth to create words, it also makes a series of vibrations. Speech to text technology works by picking up on these vibrations and translating them into a digital language through an analog to digital converter.
2. The analog-to-digital-converter takes sounds from an audio file, measures the waves in great detail, and filters them to distinguish the relevant sounds.
3. The sounds are then segmented into hundredths or thousandths of seconds and are then matched to phonemes. A phoneme is a unit of sound that distinguishes one word from another in any given language. For example, there are approximately 40 phonemes in the English language.
4. The phonemes are then run through a network via a mathematical model that compares them to well-known sentences, words, and phrases.
5. The text is then presented as text, or a computer-based demand based on the audio’s most likely version."

Let's look at an excerpt of a customer service interaction from the two perspectives:

A. Dual channel recording - 

Agent - Sorry sir, we cannot issue a refund for this. I wish we could.

Customer - This is crap! I'm going to cancel my account and go somewhere else unless you put me on with a supervisor right now who can make this happen for me.

B. Mono recording - 

Agent/Customer - Sorry sir, we cannot issue a refundcancelmy andacccount i wishunlessthisyoucrap for thisunlessyouwishisupervisor right now.

This simple, imaginary example shows how important dual channel recording can be to a business. In the dual channel instance, the transcription software will clearly translate what was spoken into words and the appropriate action can be taken to rescue the disgruntled customer.

With the mono recording/overtalk scenario, the resulting transcription will be flawed. There is no way for the speech-to-text engine to discern what was said during the portions of the call when both the agent and the customer spoke simultaneously. This can result in a lost customer.  This highly problematic as it can cost upwards of 15X more to acquire a new customer than keep an existing one.

Many call recording systems only offer mono recording. Is your call recorder dual channel or mono? If it's mono, give our dual channel/stereo recording solution a try for 30 days at no cost.

Did you know that many call recording solutions charge a substantial fee to access your own recorded audio data? It's true, even after you've already paid for their software. If you think this doesn't make sense, it doesn't. It's similar to going to McDonald's and paying for a Happy Meal, but before they give it to you, they charge you again.

So why does this happen? Well, it's due to a few reasons:

1. Many call recording software vendors are greedy
2. Some vendors claim it is complicated and time consuming to export your recordings in a playable, industry standard format - which it shouldn't be!
3. Other vendors feel the need to involve their professional services team to grant you access to your own recorded calls - this shouldn't be necessary

If you think these reasons are ridiculous, you're right! They are. Once you pay for a solution to record your customer calls, you shouldn't be charged extra to access your calls. First off, many of these same recording solutions are expensive to begin with, only to then charge you additionally for something that is already yours.

When this happens, a couple things occur:

1. You must lay out more money for something you already paid for

2. There is an unnecessary delay (days or even weeks) in accessing your recordings. This lag can cause you to lose dissatisfied customers to your competitors as well as incur costly compliance issues, because this process also delays your ability to send the recorded audio to your conversational analytics solution. Once this finally happens, you can unearth some rich customer insights that could have led to rescuing dissatisfied customers, thwarting expensive compliance infractions, mitigating costly disputes, etc. 

With all this being said, you want a recording/audio capture solution that offers unbridled access to your own recorded data (in a standard exportable format) at no extra charge. You also want the ability to collect non-audio data from CRM, ACD or agent desktop applications via a REST API, which can then be appended to audio recordings. This improves your ability to correlate, discover patterns and pinpoint specific types of interactions. Most audio capture solutions do not offer this level of openness. 

If your recording vendor charges you for access to your recordings, it may be time to look elsewhere. You don't have to be in that position.

Call recorders serve many functions, and one of their most valuable is the ability to feed recorded audio to transcription and conversational analytics engines to distill customer insights that move the needle on performance and revenue. These engines holey rely on the recorded audio file from the customer conversation to operate. With jumbled audio or two individuals speaking over one another, the resulting analytics will be flawed, at best. 

The chart above shows that for every 1,000 words of transcribed text, you could have at least 180 incorrectly transcribed words without the phonetic clarity of high fidelity, dual channel recording.

The deficiency of most recorders/analytics solutions today is the quality of that recorded audio and its portability from the recorder to the transcription/analytics engine. Many combined solutions rely on less-than-crystal-clear audio to transcribe calls and identify phonetic patterns. In fact, the use of a dual-channel/stereo recording solution (versus single channel mono) can improve transcription accuracy 15%-40%. Unfortunately, about 90% of call recorders use mono audio capture. Do you know if your recorder has stereo or mono? You should check.

It's like the chicken and the egg - which came first: the call recorder or the recording engine? Honestly, it doesn't matter because they are not necessarily inextricably linked, as they serve to very different use cases.

A call recorder is a piece of software (or hardware and software) that captures customer calls and stores them for later replay. Most recorders have a dynamic user interface to enable multi-criteria searching to quickly locate the calls you need most for compliance, dispute resolution, quality assurance, etc.

A recording engine is quite different, although you may not initially think so. Yes, a call recorder has a recording engine within it, but we aren't looking at it that way right now. For this post, we are viewing a recording engine as a distinct piece of software that captures recorded calls and sends them directly to a transcription engine, which then sends the transcribed text on to a speech analytics engine - all of which can take place in fractions of milliseconds. These solutions power real-time analytics which can identify at-risk customers before they leave, uncover compliance infractions while the agent is still on the phone, enable automated QA and so on. 

You see, one (a recorder) serves as a capture/playback device to store and replay the interactions themselves for various business purposes. The other (a recording engine) serves as a capture/streaming device, which feeds speech analytics for keyword/phrase spotting. This enables automated QA and other real-time functions (such as identifying at-risk customers or agent compliance infractions) that can drive real intraday advantages and risk mitigation.   

An important component to consider here is real-time vs. post-call in terms of how calls are handled and utilized. Real-time audio capture arms your business with some advanced capabilities that provide significant and immediate business value (e.g. rescuing customers considering defection, interceding a failing sales call to save the sale, and many more). Post-call audio capture also provides substantial value but in a less immediate manner (e.g. agent-supervisor call review, evidence for dispute resolution, and countless others).  

Questions to Ask

The question now is, which one do you need/want for your organization? The long answer is...Most large organizations have both to serve the two unique use cases. Some smaller organizations, on the other hand, choose to forego the primary call recorder altogether as they rely on the scaled-down recording functionality embedded within their telephony system. If they are satisfied with its capabilities, they may opt for a recording engine alone to power their speech analytics. 

The short answer is...You should have both. That is the best way to satisfy both use cases. 

Here are some questions to ask yourself to help you determine which is the best route for your organization:

1. How do you plan/want to leverage your customer calls?

• Real time, automated QA - recording engine
• Real time customer rescue - recording engine
• Real time compliance assurance - recording engine
• Real time order verification - recording engine
• Post-call QA - recorder
• Post-call customer win-back - recorder
• Post-call compliance verification/proof - recorder
• Post-call dispute resolution - recorder
• Post-call order verification - recorder

2. How concerned are you about intraday performance improvements?

• Very concerned - recording engine
• Not overly concerned - recorder

3. Do you plan to store your recorded customer interactions?

• Yes - recorder
• No - recording engine

Take a few minutes to think through these questions and determine where your organization stands (or wants to stand) in terms of recorder vs. recording engine use cases. Your compliance, performance and risk management efforts may depend upon it. 

Call recording provides a wealth of opportunity for businesses, most notably for agent performance management, quality assurance, customer intelligence, compliance, and risk mitigation. Recorded audio can be mined to distill critical insight which empowers each of these areas. A recorded interaction of both voice and screen, for example, can uncover workflow breakdowns that cause longer average handle times (AHT). Another recording might show an agent who is not properly complying with government regulations like GDPR, TSR, PCI, etc. Yet another could uncover important competitor information that could impact sales/marketing campaigns. 

All these benefits come from recorded conversations. However, like just about everything in business, timing is everything. Some of these insights are most valuable when discovered as the call unfolds. If a customer is indicating defection behavior, it can be too late to save the customer after the call ends. Likewise, a potentially costly dispute needs to be dealt with right away. 

As such, there are two types of call recording/audio capture methodologies to consider:

Real-time audio capture - Live audio streaming is captured as the conversation takes place in real time. When a speech analytics solution is also deployed, the recorded audio is streamed right to the transcription and analytics engines to distill critical insight as the conversation unfolds. Triggers can be set to alert supervisors immediately when a compliance infraction occurs or when a customer utters the words "cancel" or "unhappy", for instance. The proper mitigation/action can then take place while the call is still live.

Post-call audio capture - Some insights aren't as time sensitive as others, yet they still provide ample value after the call has concluded. These areas can include the discovery of product update recommendations, for example, or information about a competitor, or even evidence to settle a discrepancy with a customer. Some evidence can even be used in court. Still invaluable to a business, these insights can be gleaned after the conversation ends and still provide supervisors and management with actionable intelligence they can use to make improvements and mitigate risk. 

Do you Need Real-Time or Post-Call?

Real-time audio capture is only valuable if someone or something (transcription and speech analytics, e.g.) is listening as the call takes place. Otherwise, post-call recording will suffice, as you wouldn't be taking full advantage of its real-time, immediate value. 

Here are a couple key questions to ask yourself to determine which recording solution makes the most sense for your business:

1. Which of the following areas are most important to your contact center right now?

• Customer defection (real-time recording is best)
• Agent performance management (post-call recording will suffice, but real-time recording enables supervisors to make corrections immediately)
• Product insight (post-call recording will suffice)
• Regulatory compliance (post-call recording will suffice, but real-time recording enables supervisors to make corrections immediately)
• Dispute resolution (post-call recording will suffice, but real-time recording enables supervisors to make corrections immediately)

2. Do you have speech analytics installed or are you planning to deploy it?

Real-time streaming audio recording can feed a speech analytics engine as the call unfolds and provide invaluable insight that can be used immediately to drive service improvements, save a defecting customer, or mitigate dispute or compliance risk, for instance. AI-powered analytics can automatically score your calls and identify the most impactful insight for immediate business improvement. Furthermore, with real-time recording and speech analytics, you can trigger immediate attention and next-best action based on the most impactful customer indicators with automatic data-driven intelligence. 

AI-driven real-time analytics is revolutionizing how customer service organizations operate today, and it is all driven by real-time audio capture. Is it time for you to invest in real-time recording?